xhlm123 commited on
Commit
804e0a1
·
1 Parent(s): b7a8592

Add application file

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +0 -35
  2. AR/__pycache__/__init__.cpython-39.pyc +0 -0
  3. AR/data/bucket_sampler.py +0 -163
  4. AR/data/data_module.py +0 -76
  5. AR/data/dataset.py +0 -321
  6. AR/models/__init__.py +0 -0
  7. AR/models/__pycache__/__init__.cpython-39.pyc +0 -0
  8. AR/models/__pycache__/t2s_lightning_module.cpython-39.pyc +0 -0
  9. AR/models/__pycache__/t2s_model.cpython-39.pyc +0 -0
  10. AR/models/__pycache__/utils.cpython-39.pyc +0 -0
  11. AR/models/t2s_lightning_module.py +0 -141
  12. AR/models/t2s_lightning_module_onnx.py +0 -107
  13. AR/models/t2s_model.py +0 -588
  14. AR/models/t2s_model_onnx.py +0 -338
  15. AR/models/utils.py +0 -229
  16. AR/modules/__init__.py +0 -0
  17. AR/modules/__pycache__/__init__.cpython-39.pyc +0 -0
  18. AR/modules/__pycache__/activation.cpython-39.pyc +0 -0
  19. AR/modules/__pycache__/embedding.cpython-39.pyc +0 -0
  20. AR/modules/__pycache__/lr_schedulers.cpython-39.pyc +0 -0
  21. AR/modules/__pycache__/optim.cpython-39.pyc +0 -0
  22. AR/modules/__pycache__/patched_mha_with_cache.cpython-39.pyc +0 -0
  23. AR/modules/__pycache__/scaling.cpython-39.pyc +0 -0
  24. AR/modules/__pycache__/transformer.cpython-39.pyc +0 -0
  25. AR/modules/activation.py +0 -428
  26. AR/modules/activation_onnx.py +0 -178
  27. AR/modules/embedding.py +0 -81
  28. AR/modules/embedding_onnx.py +0 -63
  29. AR/modules/lr_schedulers.py +0 -83
  30. AR/modules/optim.py +0 -622
  31. AR/modules/patched_mha_with_cache.py +0 -465
  32. AR/modules/patched_mha_with_cache_onnx.py +0 -92
  33. AR/modules/scaling.py +0 -335
  34. AR/modules/transformer.py +0 -378
  35. AR/modules/transformer_onnx.py +0 -292
  36. AR/text_processing/__init__.py +0 -0
  37. AR/text_processing/phonemizer.py +0 -79
  38. AR/text_processing/symbols.py +0 -10
  39. AR/utils/__init__.py +0 -37
  40. AR/utils/initialize.py +0 -38
  41. AR/utils/io.py +0 -34
  42. README.md +0 -12
  43. __pycache__/utils.cpython-39.pyc +0 -0
  44. app.py +0 -678
  45. configs/s1.yaml +0 -31
  46. configs/s1big.yaml +0 -31
  47. configs/s1big2.yaml +0 -31
  48. configs/s1longer.yaml +0 -31
  49. configs/s1mq.yaml +0 -77
  50. configs/s2.json +0 -90
.gitattributes DELETED
@@ -1,35 +0,0 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
- *.tflite filter=lfs diff=lfs merge=lfs -text
30
- *.tgz filter=lfs diff=lfs merge=lfs -text
31
- *.wasm filter=lfs diff=lfs merge=lfs -text
32
- *.xz filter=lfs diff=lfs merge=lfs -text
33
- *.zip filter=lfs diff=lfs merge=lfs -text
34
- *.zst filter=lfs diff=lfs merge=lfs -text
35
- *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/__pycache__/__init__.cpython-39.pyc DELETED
Binary file (143 Bytes)
 
AR/data/bucket_sampler.py DELETED
@@ -1,163 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/data/bucket_sampler.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import itertools
4
- import math
5
- import random
6
- from random import shuffle
7
- from typing import Iterator
8
- from typing import Optional
9
- from typing import TypeVar
10
-
11
- import torch
12
- import torch.distributed as dist
13
- from torch.utils.data import Dataset
14
- from torch.utils.data import Sampler
15
-
16
- __all__ = [
17
- "DistributedBucketSampler",
18
- ]
19
-
20
- T_co = TypeVar("T_co", covariant=True)
21
-
22
-
23
- class DistributedBucketSampler(Sampler[T_co]):
24
- r"""
25
- sort the dataset wrt. input length
26
- divide samples into buckets
27
- sort within buckets
28
- divide buckets into batches
29
- sort batches
30
- """
31
-
32
- def __init__(
33
- self,
34
- dataset: Dataset,
35
- num_replicas: Optional[int] = None,
36
- rank: Optional[int] = None,
37
- shuffle: bool = True,
38
- seed: int = 0,
39
- drop_last: bool = False,
40
- batch_size: int = 32,
41
- ) -> None:
42
- if num_replicas is None:
43
- if not dist.is_available():
44
- raise RuntimeError("Requires distributed package to be available")
45
- num_replicas = dist.get_world_size() if torch.cuda.is_available() else 1
46
- if rank is None:
47
- if not dist.is_available():
48
- raise RuntimeError("Requires distributed package to be available")
49
- rank = dist.get_rank() if torch.cuda.is_available() else 0
50
- if torch.cuda.is_available():
51
- torch.cuda.set_device(rank)
52
- if rank >= num_replicas or rank < 0:
53
- raise ValueError(
54
- "Invalid rank {}, rank should be in the interval"
55
- " [0, {}]".format(rank, num_replicas - 1)
56
- )
57
- self.dataset = dataset
58
- self.num_replicas = num_replicas
59
- self.rank = rank
60
- self.epoch = 0
61
- self.drop_last = drop_last
62
- # If the dataset length is evenly divisible by # of replicas, then there
63
- # is no need to drop any data, since the dataset will be split equally.
64
- if (
65
- self.drop_last and len(self.dataset) % self.num_replicas != 0
66
- ): # type: ignore[arg-type]
67
- # Split to nearest available length that is evenly divisible.
68
- # This is to ensure each rank receives the same amount of data when
69
- # using this Sampler.
70
- self.num_samples = math.ceil(
71
- (len(self.dataset) - self.num_replicas)
72
- / self.num_replicas # type: ignore[arg-type]
73
- )
74
- else:
75
- self.num_samples = math.ceil(
76
- len(self.dataset) / self.num_replicas
77
- ) # type: ignore[arg-type]
78
- self.total_size = self.num_samples * self.num_replicas
79
- self.shuffle = shuffle
80
- self.seed = seed
81
- self.batch_size = batch_size
82
- self.id_with_length = self._get_sample_lengths()
83
- self.id_buckets = self.make_buckets(bucket_width=2.0)
84
-
85
- def _get_sample_lengths(self):
86
- id_with_lengths = []
87
- for i in range(len(self.dataset)):
88
- id_with_lengths.append((i, self.dataset.get_sample_length(i)))
89
- id_with_lengths.sort(key=lambda x: x[1])
90
- return id_with_lengths
91
-
92
- def make_buckets(self, bucket_width: float = 2.0):
93
- buckets = []
94
- cur = []
95
- max_sec = bucket_width
96
- for id, sec in self.id_with_length:
97
- if sec < max_sec:
98
- cur.append(id)
99
- else:
100
- buckets.append(cur)
101
- cur = [id]
102
- max_sec += bucket_width
103
- if len(cur) > 0:
104
- buckets.append(cur)
105
- return buckets
106
-
107
- def __iter__(self) -> Iterator[T_co]:
108
- if self.shuffle:
109
- # deterministically shuffle based on epoch and seed
110
- g = torch.Generator()
111
- g.manual_seed(self.seed + self.epoch)
112
- random.seed(self.epoch + self.seed)
113
- shuffled_bucket = []
114
- for buc in self.id_buckets:
115
- buc_copy = buc.copy()
116
- shuffle(buc_copy)
117
- shuffled_bucket.append(buc_copy)
118
- grouped_batch_size = self.batch_size * self.num_replicas
119
- shuffled_bucket = list(itertools.chain(*shuffled_bucket))
120
- n_batch = int(math.ceil(len(shuffled_bucket) / grouped_batch_size))
121
- batches = [
122
- shuffled_bucket[b * grouped_batch_size : (b + 1) * grouped_batch_size]
123
- for b in range(n_batch)
124
- ]
125
- shuffle(batches)
126
- indices = list(itertools.chain(*batches))
127
- else:
128
- # type: ignore[arg-type]
129
- indices = list(range(len(self.dataset)))
130
-
131
- if not self.drop_last:
132
- # add extra samples to make it evenly divisible
133
- padding_size = self.total_size - len(indices)
134
- if padding_size <= len(indices):
135
- indices += indices[:padding_size]
136
- else:
137
- indices += (indices * math.ceil(padding_size / len(indices)))[
138
- :padding_size
139
- ]
140
- else:
141
- # remove tail of data to make it evenly divisible.
142
- indices = indices[: self.total_size]
143
- assert len(indices) == self.total_size
144
-
145
- # subsample
146
- indices = indices[self.rank : self.total_size : self.num_replicas]
147
- assert len(indices) == self.num_samples
148
-
149
- return iter(indices)
150
-
151
- def __len__(self) -> int:
152
- return self.num_samples
153
-
154
- def set_epoch(self, epoch: int) -> None:
155
- r"""
156
- Sets the epoch for this sampler. When :attr:`shuffle=True`, this ensures all replicas
157
- use a different random ordering for each epoch. Otherwise, the next iteration of this
158
- sampler will yield the same ordering.
159
-
160
- Args:
161
- epoch (int): Epoch number.
162
- """
163
- self.epoch = epoch
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/data/data_module.py DELETED
@@ -1,76 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/data/data_module.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- from pytorch_lightning import LightningDataModule
4
- from AR.data.bucket_sampler import DistributedBucketSampler
5
- from AR.data.dataset import Text2SemanticDataset
6
- from torch.utils.data import DataLoader
7
-
8
-
9
- class Text2SemanticDataModule(LightningDataModule):
10
- def __init__(
11
- self,
12
- config,
13
- train_semantic_path,
14
- train_phoneme_path,
15
- dev_semantic_path=None,
16
- dev_phoneme_path=None,
17
- ):
18
- super().__init__()
19
- self.config = config
20
- self.train_semantic_path = train_semantic_path
21
- self.train_phoneme_path = train_phoneme_path
22
- self.dev_semantic_path = dev_semantic_path
23
- self.dev_phoneme_path = dev_phoneme_path
24
- self.num_workers = self.config["data"]["num_workers"]
25
-
26
- def prepare_data(self):
27
- pass
28
-
29
- def setup(self, stage=None, output_logs=False):
30
- self._train_dataset = Text2SemanticDataset(
31
- phoneme_path=self.train_phoneme_path,
32
- semantic_path=self.train_semantic_path,
33
- max_sec=self.config["data"]["max_sec"],
34
- pad_val=self.config["data"]["pad_val"],
35
- )
36
- self._dev_dataset = self._train_dataset
37
- # self._dev_dataset = Text2SemanticDataset(
38
- # phoneme_path=self.dev_phoneme_path,
39
- # semantic_path=self.dev_semantic_path,
40
- # max_sample=self.config['data']['max_eval_sample'],
41
- # max_sec=self.config['data']['max_sec'],
42
- # pad_val=self.config['data']['pad_val'])
43
-
44
- def train_dataloader(self):
45
- batch_size=self.config["train"]["batch_size"]//2 if self.config["train"].get("if_dpo",False)==True else self.config["train"]["batch_size"]
46
- batch_size = max(min(batch_size,len(self._train_dataset)//4),1)#防止不保存
47
- sampler = DistributedBucketSampler(self._train_dataset, batch_size=batch_size)
48
- return DataLoader(
49
- self._train_dataset,
50
- batch_size=batch_size,
51
- sampler=sampler,
52
- collate_fn=self._train_dataset.collate,
53
- num_workers=self.num_workers,
54
- persistent_workers=True,
55
- prefetch_factor=16,
56
- )
57
-
58
- def val_dataloader(self):
59
- return DataLoader(
60
- self._dev_dataset,
61
- batch_size=1,
62
- shuffle=False,
63
- collate_fn=self._train_dataset.collate,
64
- num_workers=max(self.num_workers, 12),
65
- persistent_workers=True,
66
- prefetch_factor=16,
67
- )
68
-
69
- # 这个会使用到嘛?
70
- def test_dataloader(self):
71
- return DataLoader(
72
- self._dev_dataset,
73
- batch_size=1,
74
- shuffle=False,
75
- collate_fn=self._train_dataset.collate,
76
- )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/data/dataset.py DELETED
@@ -1,321 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/data/dataset.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import pdb
4
- import sys
5
-
6
- # sys.path.append("/data/docker/liujing04/gpt-vits/mq-vits-s1bert_no_bert")
7
- import traceback, os
8
- from typing import Dict
9
- from typing import List
10
-
11
- import numpy as np
12
- import pandas as pd
13
- import torch, json
14
- from torch.utils.data import DataLoader
15
- from torch.utils.data import Dataset
16
- from transformers import AutoTokenizer
17
-
18
- from text import cleaned_text_to_sequence
19
-
20
- # from config import exp_dir
21
-
22
-
23
- def batch_sequences(sequences: List[np.array], axis: int = 0, pad_value: int = 0):
24
- seq = sequences[0]
25
- ndim = seq.ndim
26
- if axis < 0:
27
- axis += ndim
28
- dtype = seq.dtype
29
- pad_value = dtype.type(pad_value)
30
- seq_lengths = [seq.shape[axis] for seq in sequences]
31
- max_length = np.max(seq_lengths)
32
-
33
- padded_sequences = []
34
- for seq, length in zip(sequences, seq_lengths):
35
- padding = (
36
- [(0, 0)] * axis + [(0, max_length - length)] + [(0, 0)] * (ndim - axis - 1)
37
- )
38
- padded_seq = np.pad(seq, padding, mode="constant", constant_values=pad_value)
39
- padded_sequences.append(padded_seq)
40
- batch = np.stack(padded_sequences)
41
- return batch
42
-
43
-
44
- class Text2SemanticDataset(Dataset):
45
- """dataset class for text tokens to semantic model training."""
46
-
47
- def __init__(
48
- self,
49
- phoneme_path: str,
50
- semantic_path: str,
51
- max_sample: int = None,
52
- max_sec: int = 100,
53
- pad_val: int = 1024,
54
- # min value of phoneme/sec
55
- min_ps_ratio: int = 3,
56
- # max value of phoneme/sec
57
- max_ps_ratio: int = 25,
58
- ) -> None:
59
- super().__init__()
60
-
61
- self.semantic_data = pd.read_csv(
62
- semantic_path, delimiter="\t", encoding="utf-8"
63
- )
64
- # get dict
65
- self.path2 = phoneme_path # "%s/2-name2text.txt"%exp_dir#phoneme_path
66
- self.path3 = "%s/3-bert" % (
67
- os.path.basename(phoneme_path)
68
- ) # "%s/3-bert"%exp_dir#bert_dir
69
- self.path6 = semantic_path # "%s/6-name2semantic.tsv"%exp_dir#semantic_path
70
- assert os.path.exists(self.path2)
71
- assert os.path.exists(self.path6)
72
- self.phoneme_data = {}
73
- with open(self.path2, "r", encoding="utf8") as f:
74
- lines = f.read().strip("\n").split("\n")
75
-
76
- for line in lines:
77
- tmp = line.split("\t")
78
- if len(tmp) != 4:
79
- continue
80
- self.phoneme_data[tmp[0]] = [tmp[1], tmp[2], tmp[3]]
81
-
82
- # self.phoneme_data = np.load(phoneme_path, allow_pickle=True).item()
83
- # pad for semantic tokens
84
- self.PAD: int = pad_val
85
- # self.hz = 25
86
- # with open("/data/docker/liujing04/gpt-vits/mq-vits-s1bert_no_bert/configs/s2.json", "r") as f:data = f.read()
87
- # data=json.loads(data)["model"]["semantic_frame_rate"]#50hz
88
- # self.hz=int(data[:-2])#
89
- self.hz = int(os.environ.get("hz", "25hz")[:-2])
90
-
91
- # max seconds of semantic token
92
- self.max_sec = max_sec
93
- self.min_ps_ratio = min_ps_ratio
94
- self.max_ps_ratio = max_ps_ratio
95
-
96
- if max_sample is not None:
97
- self.semantic_data = self.semantic_data[:max_sample]
98
-
99
- # {idx: (semantic, phoneme)}
100
- # semantic list, phoneme list
101
- self.semantic_phoneme = []
102
- self.item_names = []
103
-
104
- self.inited = False
105
-
106
- if not self.inited:
107
- # 调用初始化函数
108
- self.init_batch()
109
- self.inited = True
110
- del self.semantic_data
111
- del self.phoneme_data
112
- # self.tokenizer = AutoTokenizer.from_pretrained("hfl/chinese-roberta-wwm-ext-large")
113
- # self.tokenizer = AutoTokenizer.from_pretrained("/data/docker/liujing04/bert-vits2/Bert-VITS2-master20231106/bert/chinese-roberta-wwm-ext-large")
114
-
115
- def init_batch(self):
116
- semantic_data_len = len(self.semantic_data)
117
- phoneme_data_len = len(self.phoneme_data.keys())
118
- print("semantic_data_len:", semantic_data_len)
119
- print("phoneme_data_len:", phoneme_data_len)
120
- print(self.semantic_data)
121
- idx = 0
122
- num_not_in = 0
123
- num_deleted_bigger = 0
124
- num_deleted_ps = 0
125
- for i in range(semantic_data_len):
126
- # 先依次遍历
127
- # get str
128
- item_name = self.semantic_data.iloc[i,0]
129
- # print(self.phoneme_data)
130
- try:
131
- phoneme, word2ph, text = self.phoneme_data[item_name]
132
- except Exception:
133
- traceback.print_exc()
134
- # print(f"{item_name} not in self.phoneme_data !")
135
- num_not_in += 1
136
- continue
137
-
138
- semantic_str = self.semantic_data.iloc[i,1]
139
- # get token list
140
- semantic_ids = [int(idx) for idx in semantic_str.split(" ")]
141
- # (T), 是否需要变成 (1, T) -> 不需要,因为需要求 len
142
- # 过滤掉太长的样本
143
- if (
144
- len(semantic_ids) > self.max_sec * self.hz
145
- ): #########1###根据token个数推测总时长过滤时长60s(config里)#40*25=1k
146
- num_deleted_bigger += 1
147
- continue
148
- # (T, ), 这个速度不会很慢,所以可以在一开始就处理,无需在 __getitem__ 里面单个处理####
149
- phoneme = phoneme.split(" ")
150
-
151
- try:
152
- phoneme_ids = cleaned_text_to_sequence(phoneme)
153
- except:
154
- traceback.print_exc()
155
- # print(f"{item_name} not in self.phoneme_data !")
156
- num_not_in += 1
157
- continue
158
- # if len(phoneme_ids) >400:###########2:改为恒定限制为semantic/2.5就行
159
- if (
160
- len(phoneme_ids) > self.max_sec * self.hz / 2.5
161
- ): ###########2:改为恒定限制为semantic/2.5就行
162
- num_deleted_ps += 1
163
- continue
164
- # if len(semantic_ids) > 1000:###########3
165
- # num_deleted_bigger += 1
166
- # continue
167
-
168
- ps_ratio = len(phoneme_ids) / (len(semantic_ids) / self.hz)
169
-
170
- if (
171
- ps_ratio > self.max_ps_ratio or ps_ratio < self.min_ps_ratio
172
- ): ##########4#3~25#每秒多少个phone
173
- num_deleted_ps += 1
174
- # print(item_name)
175
- continue
176
-
177
- self.semantic_phoneme.append((semantic_ids, phoneme_ids))
178
- idx += 1
179
- self.item_names.append(item_name)
180
-
181
- min_num = 100 # 20直接不补#30补了也不存ckpt
182
- leng = len(self.semantic_phoneme)
183
- if leng < min_num:
184
- tmp1 = self.semantic_phoneme
185
- tmp2 = self.item_names
186
- self.semantic_phoneme = []
187
- self.item_names = []
188
- for _ in range(max(2, int(min_num / leng))):
189
- self.semantic_phoneme += tmp1
190
- self.item_names += tmp2
191
- if num_not_in > 0:
192
- print(f"there are {num_not_in} semantic datas not in phoneme datas")
193
- if num_deleted_bigger > 0:
194
- print(
195
- f"deleted {num_deleted_bigger} audios who's duration are bigger than {self.max_sec} seconds"
196
- )
197
- if num_deleted_ps > 0:
198
- # 4702 for LibriTTS, LirbriTTS 是标注数据, 是否需要筛?=> 需要,有值为 100 的极端值
199
- print(
200
- f"deleted {num_deleted_ps} audios who's phoneme/sec are bigger than {self.max_ps_ratio} or smaller than {self.min_ps_ratio}"
201
- )
202
- """
203
- there are 31 semantic datas not in phoneme datas
204
- deleted 34 audios who's duration are bigger than 54 seconds
205
- deleted 3190 audios who's phoneme/sec are bigger than 25 or smaller than 3
206
- dataset.__len__(): 366463
207
-
208
- """
209
- # 345410 for LibriTTS
210
- print("dataset.__len__():", self.__len__())
211
-
212
- def __get_item_names__(self) -> List[str]:
213
- return self.item_names
214
-
215
- def __len__(self) -> int:
216
- return len(self.semantic_phoneme)
217
-
218
- def __getitem__(self, idx: int) -> Dict:
219
- semantic_ids, phoneme_ids = self.semantic_phoneme[idx]
220
- item_name = self.item_names[idx]
221
- phoneme_ids_len = len(phoneme_ids)
222
- # semantic tokens target
223
- semantic_ids_len = len(semantic_ids)
224
-
225
- flag = 0
226
- path_bert = "%s/%s.pt" % (self.path3, item_name)
227
- if os.path.exists(path_bert) == True:
228
- bert_feature = torch.load(path_bert, map_location="cpu")
229
- else:
230
- flag = 1
231
- if flag == 1:
232
- # bert_feature=torch.zeros_like(phoneme_ids,dtype=torch.float32)
233
- bert_feature = None
234
- else:
235
- assert bert_feature.shape[-1] == len(phoneme_ids)
236
- return {
237
- "idx": idx,
238
- "phoneme_ids": phoneme_ids,
239
- "phoneme_ids_len": phoneme_ids_len,
240
- "semantic_ids": semantic_ids,
241
- "semantic_ids_len": semantic_ids_len,
242
- "bert_feature": bert_feature,
243
- }
244
-
245
- def get_sample_length(self, idx: int):
246
- semantic_ids = self.semantic_phoneme[idx][0]
247
- sec = 1.0 * len(semantic_ids) / self.hz
248
- return sec
249
-
250
- def collate(self, examples: List[Dict]) -> Dict:
251
- sample_index: List[int] = []
252
- phoneme_ids: List[torch.Tensor] = []
253
- phoneme_ids_lens: List[int] = []
254
- semantic_ids: List[torch.Tensor] = []
255
- semantic_ids_lens: List[int] = []
256
- # return
257
-
258
- for item in examples:
259
- sample_index.append(item["idx"])
260
- phoneme_ids.append(np.array(item["phoneme_ids"], dtype=np.int64))
261
- semantic_ids.append(np.array(item["semantic_ids"], dtype=np.int64))
262
- phoneme_ids_lens.append(item["phoneme_ids_len"])
263
- semantic_ids_lens.append(item["semantic_ids_len"])
264
-
265
- # pad 0
266
- phoneme_ids = batch_sequences(phoneme_ids)
267
- semantic_ids = batch_sequences(semantic_ids, pad_value=self.PAD)
268
-
269
- # # convert each batch to torch.tensor
270
- phoneme_ids = torch.tensor(phoneme_ids)
271
- semantic_ids = torch.tensor(semantic_ids)
272
- phoneme_ids_lens = torch.tensor(phoneme_ids_lens)
273
- semantic_ids_lens = torch.tensor(semantic_ids_lens)
274
- bert_padded = torch.FloatTensor(len(examples), 1024, max(phoneme_ids_lens))
275
- bert_padded.zero_()
276
-
277
- for idx, item in enumerate(examples):
278
- bert = item["bert_feature"]
279
- if bert != None:
280
- bert_padded[idx, :, : bert.shape[-1]] = bert
281
-
282
- return {
283
- # List[int]
284
- "ids": sample_index,
285
- # torch.Tensor (B, max_phoneme_length)
286
- "phoneme_ids": phoneme_ids,
287
- # torch.Tensor (B)
288
- "phoneme_ids_len": phoneme_ids_lens,
289
- # torch.Tensor (B, max_semantic_ids_length)
290
- "semantic_ids": semantic_ids,
291
- # torch.Tensor (B)
292
- "semantic_ids_len": semantic_ids_lens,
293
- # torch.Tensor (B, 1024, max_phoneme_length)
294
- "bert_feature": bert_padded,
295
- }
296
-
297
-
298
- if __name__ == "__main__":
299
- root_dir = "/data/docker/liujing04/gpt-vits/prepare/dump_mix/"
300
- dataset = Text2SemanticDataset(
301
- phoneme_path=root_dir + "phoneme_train.npy",
302
- semantic_path=root_dir + "semantic_train.tsv",
303
- )
304
-
305
- batch_size = 12
306
- dataloader = DataLoader(
307
- dataset, batch_size=batch_size, collate_fn=dataset.collate, shuffle=False
308
- )
309
- for i, batch in enumerate(dataloader):
310
- if i % 1000 == 0:
311
- print(i)
312
- # if i == 0:
313
- # print('batch["ids"]:', batch["ids"])
314
- # print('batch["phoneme_ids"]:', batch["phoneme_ids"],
315
- # batch["phoneme_ids"].shape)
316
- # print('batch["phoneme_ids_len"]:', batch["phoneme_ids_len"],
317
- # batch["phoneme_ids_len"].shape)
318
- # print('batch["semantic_ids"]:', batch["semantic_ids"],
319
- # batch["semantic_ids"].shape)
320
- # print('batch["semantic_ids_len"]:', batch["semantic_ids_len"],
321
- # batch["semantic_ids_len"].shape)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/models/__init__.py DELETED
File without changes
AR/models/__pycache__/__init__.cpython-39.pyc DELETED
Binary file (150 Bytes)
 
AR/models/__pycache__/t2s_lightning_module.cpython-39.pyc DELETED
Binary file (3.22 kB)
 
AR/models/__pycache__/t2s_model.cpython-39.pyc DELETED
Binary file (12.6 kB)
 
AR/models/__pycache__/utils.cpython-39.pyc DELETED
Binary file (6.63 kB)
 
AR/models/t2s_lightning_module.py DELETED
@@ -1,141 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/t2s_lightning_module.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import os, sys
4
-
5
- now_dir = os.getcwd()
6
- sys.path.append(now_dir)
7
- from typing import Dict
8
-
9
- import torch
10
- from pytorch_lightning import LightningModule
11
- from AR.models.t2s_model import Text2SemanticDecoder
12
- from AR.modules.lr_schedulers import WarmupCosineLRSchedule
13
- from AR.modules.optim import ScaledAdam
14
-
15
- class Text2SemanticLightningModule(LightningModule):
16
- def __init__(self, config, output_dir, is_train=True):
17
- super().__init__()
18
- self.config = config
19
- self.top_k = 3
20
- self.model = Text2SemanticDecoder(config=config, top_k=self.top_k)
21
- pretrained_s1 = config.get("pretrained_s1")
22
- if pretrained_s1 and is_train:
23
- # print(self.load_state_dict(torch.load(pretrained_s1,map_location="cpu")["state_dict"]))
24
- print(
25
- self.load_state_dict(
26
- torch.load(pretrained_s1, map_location="cpu")["weight"]
27
- )
28
- )
29
- if is_train:
30
- self.automatic_optimization = False
31
- self.save_hyperparameters()
32
- self.eval_dir = output_dir / "eval"
33
- self.eval_dir.mkdir(parents=True, exist_ok=True)
34
-
35
- def training_step(self, batch: Dict, batch_idx: int):
36
- opt = self.optimizers()
37
- scheduler = self.lr_schedulers()
38
- forward=self.model.forward if self.config["train"].get("if_dpo",False)==True else self.model.forward_old
39
- loss, acc = forward(
40
- batch["phoneme_ids"],
41
- batch["phoneme_ids_len"],
42
- batch["semantic_ids"],
43
- batch["semantic_ids_len"],
44
- batch["bert_feature"],
45
- )
46
- self.manual_backward(loss)
47
- if batch_idx > 0 and batch_idx % 4 == 0:
48
- opt.step()
49
- opt.zero_grad()
50
- scheduler.step()
51
-
52
- self.log(
53
- "total_loss",
54
- loss,
55
- on_step=True,
56
- on_epoch=True,
57
- prog_bar=True,
58
- sync_dist=True,
59
- )
60
- self.log(
61
- "lr",
62
- scheduler.get_last_lr()[0],
63
- on_epoch=True,
64
- prog_bar=True,
65
- sync_dist=True,
66
- )
67
- self.log(
68
- f"top_{self.top_k}_acc",
69
- acc,
70
- on_step=True,
71
- on_epoch=True,
72
- prog_bar=True,
73
- sync_dist=True,
74
- )
75
-
76
- def validation_step(self, batch: Dict, batch_idx: int):
77
- return
78
-
79
- # # get loss
80
- # loss, acc = self.model.forward(
81
- # batch['phoneme_ids'], batch['phoneme_ids_len'],
82
- # batch['semantic_ids'], batch['semantic_ids_len'],
83
- # batch['bert_feature']
84
- # )
85
- #
86
- # self.log(
87
- # "val_total_loss",
88
- # loss,
89
- # on_step=True,
90
- # on_epoch=True,
91
- # prog_bar=True,
92
- # sync_dist=True)
93
- # self.log(
94
- # f"val_top_{self.top_k}_acc",
95
- # acc,
96
- # on_step=True,
97
- # on_epoch=True,
98
- # prog_bar=True,
99
- # sync_dist=True)
100
- #
101
- # # get infer output
102
- # semantic_len = batch['semantic_ids'].size(1)
103
- # prompt_len = min(int(semantic_len * 0.5), 150)
104
- # prompt = batch['semantic_ids'][:, :prompt_len]
105
- # pred_semantic = self.model.infer(batch['phoneme_ids'],
106
- # batch['phoneme_ids_len'], prompt,
107
- # batch['bert_feature']
108
- # )
109
- # save_name = f'semantic_toks_{batch_idx}.pt'
110
- # save_path = os.path.join(self.eval_dir, save_name)
111
- # torch.save(pred_semantic.detach().cpu(), save_path)
112
-
113
- def configure_optimizers(self):
114
- model_parameters = self.model.parameters()
115
- parameters_names = []
116
- parameters_names.append(
117
- [name_param_pair[0] for name_param_pair in self.model.named_parameters()]
118
- )
119
- lm_opt = ScaledAdam(
120
- model_parameters,
121
- lr=0.01,
122
- betas=(0.9, 0.95),
123
- clipping_scale=2.0,
124
- parameters_names=parameters_names,
125
- show_dominant_parameters=False,
126
- clipping_update_period=1000,
127
- )
128
-
129
- return {
130
- "optimizer": lm_opt,
131
- "lr_scheduler": {
132
- "scheduler": WarmupCosineLRSchedule(
133
- lm_opt,
134
- init_lr=self.config["optimizer"]["lr_init"],
135
- peak_lr=self.config["optimizer"]["lr"],
136
- end_lr=self.config["optimizer"]["lr_end"],
137
- warmup_steps=self.config["optimizer"]["warmup_steps"],
138
- total_steps=self.config["optimizer"]["decay_steps"],
139
- )
140
- },
141
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/models/t2s_lightning_module_onnx.py DELETED
@@ -1,107 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/t2s_lightning_module.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import os, sys
4
-
5
- now_dir = os.getcwd()
6
- sys.path.append(now_dir)
7
- from typing import Dict
8
-
9
- import torch
10
- from pytorch_lightning import LightningModule
11
- from AR.models.t2s_model_onnx import Text2SemanticDecoder
12
- from AR.modules.lr_schedulers import WarmupCosineLRSchedule
13
- from AR.modules.optim import ScaledAdam
14
-
15
-
16
- class Text2SemanticLightningModule(LightningModule):
17
- def __init__(self, config, output_dir, is_train=True):
18
- super().__init__()
19
- self.config = config
20
- self.top_k = 3
21
- self.model = Text2SemanticDecoder(config=config, top_k=self.top_k)
22
- pretrained_s1 = config.get("pretrained_s1")
23
- if pretrained_s1 and is_train:
24
- # print(self.load_state_dict(torch.load(pretrained_s1,map_location="cpu")["state_dict"]))
25
- print(
26
- self.load_state_dict(
27
- torch.load(pretrained_s1, map_location="cpu")["weight"]
28
- )
29
- )
30
- if is_train:
31
- self.automatic_optimization = False
32
- self.save_hyperparameters()
33
- self.eval_dir = output_dir / "eval"
34
- self.eval_dir.mkdir(parents=True, exist_ok=True)
35
-
36
- def training_step(self, batch: Dict, batch_idx: int):
37
- opt = self.optimizers()
38
- scheduler = self.lr_schedulers()
39
- loss, acc = self.model.forward(
40
- batch["phoneme_ids"],
41
- batch["phoneme_ids_len"],
42
- batch["semantic_ids"],
43
- batch["semantic_ids_len"],
44
- batch["bert_feature"],
45
- )
46
- self.manual_backward(loss)
47
- if batch_idx > 0 and batch_idx % 4 == 0:
48
- opt.step()
49
- opt.zero_grad()
50
- scheduler.step()
51
-
52
- self.log(
53
- "total_loss",
54
- loss,
55
- on_step=True,
56
- on_epoch=True,
57
- prog_bar=True,
58
- sync_dist=True,
59
- )
60
- self.log(
61
- "lr",
62
- scheduler.get_last_lr()[0],
63
- on_epoch=True,
64
- prog_bar=True,
65
- sync_dist=True,
66
- )
67
- self.log(
68
- f"top_{self.top_k}_acc",
69
- acc,
70
- on_step=True,
71
- on_epoch=True,
72
- prog_bar=True,
73
- sync_dist=True,
74
- )
75
-
76
- def validation_step(self, batch: Dict, batch_idx: int):
77
- return
78
-
79
- def configure_optimizers(self):
80
- model_parameters = self.model.parameters()
81
- parameters_names = []
82
- parameters_names.append(
83
- [name_param_pair[0] for name_param_pair in self.model.named_parameters()]
84
- )
85
- lm_opt = ScaledAdam(
86
- model_parameters,
87
- lr=0.01,
88
- betas=(0.9, 0.95),
89
- clipping_scale=2.0,
90
- parameters_names=parameters_names,
91
- show_dominant_parameters=False,
92
- clipping_update_period=1000,
93
- )
94
-
95
- return {
96
- "optimizer": lm_opt,
97
- "lr_scheduler": {
98
- "scheduler": WarmupCosineLRSchedule(
99
- lm_opt,
100
- init_lr=self.config["optimizer"]["lr_init"],
101
- peak_lr=self.config["optimizer"]["lr"],
102
- end_lr=self.config["optimizer"]["lr_end"],
103
- warmup_steps=self.config["optimizer"]["warmup_steps"],
104
- total_steps=self.config["optimizer"]["decay_steps"],
105
- )
106
- },
107
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/models/t2s_model.py DELETED
@@ -1,588 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/t2s_model.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- from typing import List
4
-
5
- import torch
6
- from tqdm import tqdm
7
-
8
- from AR.models.utils import make_pad_mask
9
- from AR.models.utils import (
10
- topk_sampling,
11
- sample,
12
- logits_to_probs,
13
- multinomial_sample_one_no_sync,
14
- dpo_loss,
15
- make_reject_y,
16
- get_batch_logps
17
- )
18
- from AR.modules.embedding import SinePositionalEmbedding
19
- from AR.modules.embedding import TokenEmbedding
20
- from AR.modules.transformer import LayerNorm
21
- from AR.modules.transformer import TransformerEncoder
22
- from AR.modules.transformer import TransformerEncoderLayer
23
- from torch import nn
24
- from torch.nn import functional as F
25
- from torchmetrics.classification import MulticlassAccuracy
26
-
27
- default_config = {
28
- "embedding_dim": 512,
29
- "hidden_dim": 512,
30
- "num_head": 8,
31
- "num_layers": 12,
32
- "num_codebook": 8,
33
- "p_dropout": 0.0,
34
- "vocab_size": 1024 + 1,
35
- "phoneme_vocab_size": 512,
36
- "EOS": 1024,
37
- }
38
-
39
-
40
- @torch.jit.script
41
- class T2SMLP:
42
- def __init__(self, w1, b1, w2, b2):
43
- self.w1 = w1
44
- self.b1 = b1
45
- self.w2 = w2
46
- self.b2 = b2
47
-
48
- def forward(self, x):
49
- x = F.relu(F.linear(x, self.w1, self.b1))
50
- x = F.linear(x, self.w2, self.b2)
51
- return x
52
-
53
-
54
- @torch.jit.script
55
- class T2SBlock:
56
- def __init__(
57
- self,
58
- num_heads,
59
- hidden_dim: int,
60
- mlp: T2SMLP,
61
- qkv_w,
62
- qkv_b,
63
- out_w,
64
- out_b,
65
- norm_w1,
66
- norm_b1,
67
- norm_eps1,
68
- norm_w2,
69
- norm_b2,
70
- norm_eps2,
71
- ):
72
- self.num_heads = num_heads
73
- self.mlp = mlp
74
- self.hidden_dim: int = hidden_dim
75
- self.qkv_w = qkv_w
76
- self.qkv_b = qkv_b
77
- self.out_w = out_w
78
- self.out_b = out_b
79
- self.norm_w1 = norm_w1
80
- self.norm_b1 = norm_b1
81
- self.norm_eps1 = norm_eps1
82
- self.norm_w2 = norm_w2
83
- self.norm_b2 = norm_b2
84
- self.norm_eps2 = norm_eps2
85
-
86
- def process_prompt(self, x, attn_mask : torch.Tensor):
87
- q, k, v = F.linear(x, self.qkv_w, self.qkv_b).chunk(3, dim=-1)
88
-
89
- batch_size = q.shape[0]
90
- q_len = q.shape[1]
91
- kv_len = k.shape[1]
92
-
93
- k_cache = k
94
- v_cache = v
95
-
96
- q = q.view(batch_size, q_len, self.num_heads, -1).transpose(1, 2)
97
- k = k_cache.view(batch_size, kv_len, self.num_heads, -1).transpose(1, 2)
98
- v = v_cache.view(batch_size, kv_len, self.num_heads, -1).transpose(1, 2)
99
-
100
- attn = F.scaled_dot_product_attention(q, k, v, ~attn_mask)
101
-
102
- attn = attn.permute(2, 0, 1, 3).reshape(batch_size, -1, self.hidden_dim)
103
- attn = F.linear(attn, self.out_w, self.out_b)
104
-
105
- x = F.layer_norm(
106
- x + attn, [self.hidden_dim], self.norm_w1, self.norm_b1, self.norm_eps1
107
- )
108
- x = F.layer_norm(
109
- x + self.mlp.forward(x),
110
- [self.hidden_dim],
111
- self.norm_w2,
112
- self.norm_b2,
113
- self.norm_eps2,
114
- )
115
- return x, k_cache, v_cache
116
-
117
- def decode_next_token(self, x, k_cache, v_cache):
118
- q, k, v = F.linear(x, self.qkv_w, self.qkv_b).chunk(3, dim=-1)
119
-
120
- k_cache = torch.cat([k_cache, k], dim=1)
121
- v_cache = torch.cat([v_cache, v], dim=1)
122
- kv_len = k_cache.shape[1]
123
-
124
- batch_size = q.shape[0]
125
- q_len = q.shape[1]
126
-
127
- q = q.view(batch_size, q_len, self.num_heads, -1).transpose(1, 2)
128
- k = k_cache.view(batch_size, kv_len, self.num_heads, -1).transpose(1, 2)
129
- v = v_cache.view(batch_size, kv_len, self.num_heads, -1).transpose(1, 2)
130
-
131
-
132
- attn = F.scaled_dot_product_attention(q, k, v)
133
-
134
- attn = attn.permute(2, 0, 1, 3).reshape(batch_size, -1, self.hidden_dim)
135
- attn = F.linear(attn, self.out_w, self.out_b)
136
-
137
- x = F.layer_norm(
138
- x + attn, [self.hidden_dim], self.norm_w1, self.norm_b1, self.norm_eps1
139
- )
140
- x = F.layer_norm(
141
- x + self.mlp.forward(x),
142
- [self.hidden_dim],
143
- self.norm_w2,
144
- self.norm_b2,
145
- self.norm_eps2,
146
- )
147
- return x, k_cache, v_cache
148
-
149
-
150
- @torch.jit.script
151
- class T2STransformer:
152
- def __init__(self, num_blocks : int, blocks: List[T2SBlock]):
153
- self.num_blocks : int = num_blocks
154
- self.blocks = blocks
155
-
156
- def process_prompt(
157
- self, x, attn_mask : torch.Tensor):
158
- k_cache : List[torch.Tensor] = []
159
- v_cache : List[torch.Tensor] = []
160
- for i in range(self.num_blocks):
161
- x, k_cache_, v_cache_ = self.blocks[i].process_prompt(x, attn_mask)
162
- k_cache.append(k_cache_)
163
- v_cache.append(v_cache_)
164
- return x, k_cache, v_cache
165
-
166
- def decode_next_token(
167
- self, x, k_cache: List[torch.Tensor], v_cache: List[torch.Tensor]
168
- ):
169
- for i in range(self.num_blocks):
170
- x, k_cache[i], v_cache[i] = self.blocks[i].decode_next_token(x, k_cache[i], v_cache[i])
171
- return x, k_cache, v_cache
172
-
173
-
174
- class Text2SemanticDecoder(nn.Module):
175
- def __init__(self, config, norm_first=False, top_k=3):
176
- super(Text2SemanticDecoder, self).__init__()
177
- self.model_dim = config["model"]["hidden_dim"]
178
- self.embedding_dim = config["model"]["embedding_dim"]
179
- self.num_head = config["model"]["head"]
180
- self.num_layers = config["model"]["n_layer"]
181
- self.norm_first = norm_first
182
- self.vocab_size = config["model"]["vocab_size"]
183
- self.phoneme_vocab_size = config["model"]["phoneme_vocab_size"]
184
- self.p_dropout = config["model"]["dropout"]
185
- self.EOS = config["model"]["EOS"]
186
- self.norm_first = norm_first
187
- assert self.EOS == self.vocab_size - 1
188
- # should be same as num of kmeans bin
189
- # assert self.EOS == 1024
190
- self.bert_proj = nn.Linear(1024, self.embedding_dim)
191
- self.ar_text_embedding = TokenEmbedding(
192
- self.embedding_dim, self.phoneme_vocab_size, self.p_dropout
193
- )
194
- self.ar_text_position = SinePositionalEmbedding(
195
- self.embedding_dim, dropout=0.1, scale=False, alpha=True
196
- )
197
- self.ar_audio_embedding = TokenEmbedding(
198
- self.embedding_dim, self.vocab_size, self.p_dropout
199
- )
200
- self.ar_audio_position = SinePositionalEmbedding(
201
- self.embedding_dim, dropout=0.1, scale=False, alpha=True
202
- )
203
-
204
- self.h = TransformerEncoder(
205
- TransformerEncoderLayer(
206
- d_model=self.model_dim,
207
- nhead=self.num_head,
208
- dim_feedforward=self.model_dim * 4,
209
- dropout=0.1,
210
- batch_first=True,
211
- norm_first=norm_first,
212
- ),
213
- num_layers=self.num_layers,
214
- norm=LayerNorm(self.model_dim) if norm_first else None,
215
- )
216
-
217
- self.ar_predict_layer = nn.Linear(self.model_dim, self.vocab_size, bias=False)
218
- self.loss_fct = nn.CrossEntropyLoss(reduction="sum")
219
-
220
- self.ar_accuracy_metric = MulticlassAccuracy(
221
- self.vocab_size,
222
- top_k=top_k,
223
- average="micro",
224
- multidim_average="global",
225
- ignore_index=self.EOS,
226
- )
227
-
228
- blocks = []
229
-
230
- for i in range(self.num_layers):
231
- layer = self.h.layers[i]
232
- t2smlp = T2SMLP(
233
- layer.linear1.weight,
234
- layer.linear1.bias,
235
- layer.linear2.weight,
236
- layer.linear2.bias
237
- )
238
-
239
- block = T2SBlock(
240
- self.num_head,
241
- self.model_dim,
242
- t2smlp,
243
- layer.self_attn.in_proj_weight,
244
- layer.self_attn.in_proj_bias,
245
- layer.self_attn.out_proj.weight,
246
- layer.self_attn.out_proj.bias,
247
- layer.norm1.weight,
248
- layer.norm1.bias,
249
- layer.norm1.eps,
250
- layer.norm2.weight,
251
- layer.norm2.bias,
252
- layer.norm2.eps
253
- )
254
-
255
- blocks.append(block)
256
-
257
- self.t2s_transformer = T2STransformer(self.num_layers, blocks)
258
-
259
- # self.t2s_transformer.process_prompt = torch.compile(self.t2s_transformer.process_prompt,mode="reduce-overhead", fullgraph=True)
260
- # self.t2s_transformer.decode_next_token = torch.compile(self.t2s_transformer.decode_next_token,mode="reduce-overhead", fullgraph=True)
261
-
262
- def make_input_data(self, x, x_lens, y, y_lens, bert_feature):
263
- x = self.ar_text_embedding(x)
264
- x = x + self.bert_proj(bert_feature.transpose(1, 2))
265
- x = self.ar_text_position(x)
266
- x_mask = make_pad_mask(x_lens)
267
-
268
- y_mask = make_pad_mask(y_lens)
269
- y_mask_int = y_mask.type(torch.int64)
270
- codes = y.type(torch.int64) * (1 - y_mask_int)
271
-
272
- # Training
273
- # AR Decoder
274
- y, targets = self.pad_y_eos(codes, y_mask_int, eos_id=self.EOS)
275
- x_len = x_lens.max()
276
- y_len = y_lens.max()
277
- y_emb = self.ar_audio_embedding(y)
278
- y_pos = self.ar_audio_position(y_emb)
279
-
280
- xy_padding_mask = torch.concat([x_mask, y_mask], dim=1)
281
-
282
- ar_xy_padding_mask = xy_padding_mask
283
-
284
- x_attn_mask = F.pad(
285
- torch.zeros((x_len, x_len), dtype=torch.bool, device=x.device),
286
- (0, y_len),
287
- value=True,
288
- )
289
-
290
- y_attn_mask = F.pad(
291
- torch.triu(
292
- torch.ones(y_len, y_len, dtype=torch.bool, device=x.device),
293
- diagonal=1,
294
- ),
295
- (x_len, 0),
296
- value=False,
297
- )
298
-
299
- xy_attn_mask = torch.concat([x_attn_mask, y_attn_mask], dim=0)
300
- bsz, src_len = x.shape[0], x_len + y_len
301
- _xy_padding_mask = (
302
- ar_xy_padding_mask.view(bsz, 1, 1, src_len)
303
- .expand(-1, self.num_head, -1, -1)
304
- .reshape(bsz * self.num_head, 1, src_len)
305
- )
306
- xy_attn_mask = xy_attn_mask.logical_or(_xy_padding_mask)
307
- new_attn_mask = torch.zeros_like(xy_attn_mask, dtype=x.dtype)
308
- new_attn_mask.masked_fill_(xy_attn_mask, float("-inf"))
309
- xy_attn_mask = new_attn_mask
310
- # x 和完整的 y 一次性输入模型
311
- xy_pos = torch.concat([x, y_pos], dim=1)
312
-
313
- return xy_pos, xy_attn_mask, targets
314
-
315
- def forward(self, x, x_lens, y, y_lens, bert_feature):
316
- """
317
- x: phoneme_ids
318
- y: semantic_ids
319
- """
320
-
321
- reject_y, reject_y_lens = make_reject_y(y, y_lens)
322
-
323
- xy_pos, xy_attn_mask, targets = self.make_input_data(x, x_lens, y, y_lens, bert_feature)
324
-
325
- xy_dec, _ = self.h(
326
- (xy_pos, None),
327
- mask=xy_attn_mask,
328
- )
329
- x_len = x_lens.max()
330
- logits = self.ar_predict_layer(xy_dec[:, x_len:])
331
-
332
- ###### DPO #############
333
- reject_xy_pos, reject_xy_attn_mask, reject_targets = self.make_input_data(x, x_lens, reject_y, reject_y_lens, bert_feature)
334
-
335
- reject_xy_dec, _ = self.h(
336
- (reject_xy_pos, None),
337
- mask=reject_xy_attn_mask,
338
- )
339
- x_len = x_lens.max()
340
- reject_logits = self.ar_predict_layer(reject_xy_dec[:, x_len:])
341
-
342
- # loss
343
- # from feiteng: 每次 duration 越多, 梯度更新也应该更多, 所以用 sum
344
-
345
- loss_1 = F.cross_entropy(logits.permute(0, 2, 1), targets, reduction="sum")
346
- acc = self.ar_accuracy_metric(logits.permute(0, 2, 1).detach(), targets).item()
347
-
348
- A_logits, R_logits = get_batch_logps(logits, reject_logits, targets, reject_targets)
349
- loss_2, _, _ = dpo_loss(A_logits, R_logits, 0, 0, 0.2, reference_free=True)
350
-
351
- loss = loss_1 + loss_2
352
-
353
- return loss, acc
354
-
355
- def forward_old(self, x, x_lens, y, y_lens, bert_feature):
356
- """
357
- x: phoneme_ids
358
- y: semantic_ids
359
- """
360
- x = self.ar_text_embedding(x)
361
- x = x + self.bert_proj(bert_feature.transpose(1, 2))
362
- x = self.ar_text_position(x)
363
- x_mask = make_pad_mask(x_lens)
364
-
365
- y_mask = make_pad_mask(y_lens)
366
- y_mask_int = y_mask.type(torch.int64)
367
- codes = y.type(torch.int64) * (1 - y_mask_int)
368
-
369
- # Training
370
- # AR Decoder
371
- y, targets = self.pad_y_eos(codes, y_mask_int, eos_id=self.EOS)
372
- x_len = x_lens.max()
373
- y_len = y_lens.max()
374
- y_emb = self.ar_audio_embedding(y)
375
- y_pos = self.ar_audio_position(y_emb)
376
-
377
- xy_padding_mask = torch.concat([x_mask, y_mask], dim=1)
378
- ar_xy_padding_mask = xy_padding_mask
379
-
380
- x_attn_mask = F.pad(
381
- torch.zeros((x_len, x_len), dtype=torch.bool, device=x.device),
382
- (0, y_len),
383
- value=True,
384
- )
385
- y_attn_mask = F.pad(
386
- torch.triu(
387
- torch.ones(y_len, y_len, dtype=torch.bool, device=x.device),
388
- diagonal=1,
389
- ),
390
- (x_len, 0),
391
- value=False,
392
- )
393
- xy_attn_mask = torch.concat([x_attn_mask, y_attn_mask], dim=0)
394
- bsz, src_len = x.shape[0], x_len + y_len
395
- _xy_padding_mask = (
396
- ar_xy_padding_mask.view(bsz, 1, 1, src_len)
397
- .expand(-1, self.num_head, -1, -1)
398
- .reshape(bsz * self.num_head, 1, src_len)
399
- )
400
- xy_attn_mask = xy_attn_mask.logical_or(_xy_padding_mask)
401
- new_attn_mask = torch.zeros_like(xy_attn_mask, dtype=x.dtype)
402
- new_attn_mask.masked_fill_(xy_attn_mask, float("-inf"))
403
- xy_attn_mask = new_attn_mask
404
- # x 和完整的 y 一次性输入模型
405
- xy_pos = torch.concat([x, y_pos], dim=1)
406
- xy_dec, _ = self.h(
407
- (xy_pos, None),
408
- mask=xy_attn_mask,
409
- )
410
- logits = self.ar_predict_layer(xy_dec[:, x_len:]).permute(0, 2, 1)
411
- # loss
412
- # from feiteng: 每次 duration 越多, 梯度更新也应该更多, 所以用 sum
413
- loss = F.cross_entropy(logits, targets, reduction="sum")
414
- acc = self.ar_accuracy_metric(logits.detach(), targets).item()
415
- return loss, acc
416
-
417
- # 需要看下这个函数和 forward 的区别以及没有 semantic 的时候 prompts 输入什么
418
- def infer(
419
- self,
420
- x,
421
- x_lens,
422
- prompts,
423
- bert_feature,
424
- top_k: int = -100,
425
- early_stop_num: int = -1,
426
- temperature: float = 1.0,
427
- ):
428
- x = self.ar_text_embedding(x)
429
- x = x + self.bert_proj(bert_feature.transpose(1, 2))
430
- x = self.ar_text_position(x)
431
-
432
- # AR Decoder
433
- y = prompts
434
- prefix_len = y.shape[1]
435
- x_len = x.shape[1]
436
- x_attn_mask = torch.zeros((x_len, x_len), dtype=torch.bool)
437
- stop = False
438
- for _ in tqdm(range(1500)):
439
- y_emb = self.ar_audio_embedding(y)
440
- y_pos = self.ar_audio_position(y_emb)
441
- # x 和逐渐增长的 y 一起输入给模型
442
- xy_pos = torch.concat([x, y_pos], dim=1)
443
- y_len = y.shape[1]
444
- x_attn_mask_pad = F.pad(
445
- x_attn_mask,
446
- (0, y_len),
447
- value=True,
448
- )
449
- y_attn_mask = F.pad(
450
- torch.triu(torch.ones(y_len, y_len, dtype=torch.bool), diagonal=1),
451
- (x_len, 0),
452
- value=False,
453
- )
454
- xy_attn_mask = torch.concat([x_attn_mask_pad, y_attn_mask], dim=0).to(
455
- y.device
456
- )
457
-
458
- xy_dec, _ = self.h(
459
- (xy_pos, None),
460
- mask=xy_attn_mask,
461
- )
462
- logits = self.ar_predict_layer(xy_dec[:, -1])
463
- samples = topk_sampling(
464
- logits, top_k=top_k, top_p=1.0, temperature=temperature
465
- )
466
-
467
- if early_stop_num != -1 and (y.shape[1] - prefix_len) > early_stop_num:
468
- print("use early stop num:", early_stop_num)
469
- stop = True
470
-
471
- if torch.argmax(logits, dim=-1)[0] == self.EOS or samples[0, 0] == self.EOS:
472
- # print(torch.argmax(logits, dim=-1)[0] == self.EOS, samples[0, 0] == self.EOS)
473
- stop = True
474
- if stop:
475
- if prompts.shape[1] == y.shape[1]:
476
- y = torch.concat([y, torch.zeros_like(samples)], dim=1)
477
- print("bad zero prediction")
478
- print(f"T2S Decoding EOS [{prefix_len} -> {y.shape[1]}]")
479
- break
480
- # 本次生成的 semantic_ids 和之前的 y 构成新的 y
481
- # print(samples.shape)#[1,1]#第一个1是bs
482
- # import os
483
- # os._exit(2333)
484
- y = torch.concat([y, samples], dim=1)
485
- return y
486
-
487
- def pad_y_eos(self, y, y_mask_int, eos_id):
488
- targets = F.pad(y, (0, 1), value=0) + eos_id * F.pad(
489
- y_mask_int, (0, 1), value=1
490
- )
491
- # 错位
492
- return targets[:, :-1], targets[:, 1:]
493
-
494
- def infer_panel(
495
- self,
496
- x, #####全部文本token
497
- x_lens,
498
- prompts, ####参考音频token
499
- bert_feature,
500
- top_k: int = -100,
501
- top_p: int = 100,
502
- early_stop_num: int = -1,
503
- temperature: float = 1.0,
504
- ):
505
- x = self.ar_text_embedding(x)
506
- x = x + self.bert_proj(bert_feature.transpose(1, 2))
507
- x = self.ar_text_position(x)
508
-
509
- # AR Decoder
510
- y = prompts
511
-
512
- x_len = x.shape[1]
513
- x_attn_mask = torch.zeros((x_len, x_len), dtype=torch.bool)
514
- stop = False
515
- # print(1111111,self.num_layers)
516
-
517
- k_cache = None
518
- v_cache = None
519
- ################### first step ##########################
520
- if y is not None:
521
- y_emb = self.ar_audio_embedding(y)
522
- y_len = y_emb.shape[1]
523
- prefix_len = y.shape[1]
524
- y_pos = self.ar_audio_position(y_emb)
525
- xy_pos = torch.concat([x, y_pos], dim=1)
526
- ref_free = False
527
- else:
528
- y_emb = None
529
- y_len = 0
530
- prefix_len = 0
531
- y_pos = None
532
- xy_pos = x
533
- y = torch.zeros(x.shape[0], 0, dtype=torch.int, device=x.device)
534
- ref_free = True
535
-
536
- x_attn_mask_pad = F.pad(
537
- x_attn_mask,
538
- (0, y_len), ###xx的纯0扩展到xx纯0+xy纯1,(x,x+y)
539
- value=True,
540
- )
541
- y_attn_mask = F.pad( ###yy的右上1扩展到左边xy的0,(y,x+y)
542
- torch.triu(torch.ones(y_len, y_len, dtype=torch.bool), diagonal=1),
543
- (x_len, 0),
544
- value=False,
545
- )
546
- xy_attn_mask = torch.concat([x_attn_mask_pad, y_attn_mask], dim=0).to(
547
- x.device
548
- )
549
-
550
- for idx in tqdm(range(1500)):
551
- if xy_attn_mask is not None:
552
- xy_dec, k_cache, v_cache = self.t2s_transformer.process_prompt(xy_pos, xy_attn_mask)
553
- else:
554
- xy_dec, k_cache, v_cache = self.t2s_transformer.decode_next_token(xy_pos, k_cache, v_cache)
555
-
556
- logits = self.ar_predict_layer(
557
- xy_dec[:, -1]
558
- )
559
-
560
- if idx == 0:
561
- xy_attn_mask = None
562
- logits = logits[:, :-1]
563
- samples = sample(
564
- logits[0], y, top_k=top_k, top_p=top_p, repetition_penalty=1.35, temperature=temperature
565
- )[0].unsqueeze(0)
566
-
567
- y = torch.concat([y, samples], dim=1)
568
-
569
- if early_stop_num != -1 and (y.shape[1] - prefix_len) > early_stop_num:
570
- print("use early stop num:", early_stop_num)
571
- stop = True
572
-
573
- if torch.argmax(logits, dim=-1)[0] == self.EOS or samples[0, 0] == self.EOS:
574
- stop = True
575
- if stop:
576
- if y.shape[1]==0:
577
- y = torch.concat([y, torch.zeros_like(samples)], dim=1)
578
- print("bad zero prediction")
579
- print(f"T2S Decoding EOS [{prefix_len} -> {y.shape[1]}]")
580
- break
581
-
582
- ####################### update next step ###################################
583
- y_emb = self.ar_audio_embedding(y[:, -1:])
584
- xy_pos = y_emb * self.ar_audio_position.x_scale + self.ar_audio_position.alpha * self.ar_audio_position.pe[:, y_len + idx].to(dtype=y_emb.dtype,device=y_emb.device)
585
-
586
- if ref_free:
587
- return y[:, :-1], 0
588
- return y[:, :-1], idx - 1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/models/t2s_model_onnx.py DELETED
@@ -1,338 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/t2s_model.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import torch
4
- from tqdm import tqdm
5
-
6
- from AR.modules.embedding_onnx import SinePositionalEmbedding
7
- from AR.modules.embedding_onnx import TokenEmbedding
8
- from AR.modules.transformer_onnx import LayerNorm
9
- from AR.modules.transformer_onnx import TransformerEncoder
10
- from AR.modules.transformer_onnx import TransformerEncoderLayer
11
- from torch import nn
12
- from torch.nn import functional as F
13
- from torchmetrics.classification import MulticlassAccuracy
14
-
15
- default_config = {
16
- "embedding_dim": 512,
17
- "hidden_dim": 512,
18
- "num_head": 8,
19
- "num_layers": 12,
20
- "num_codebook": 8,
21
- "p_dropout": 0.0,
22
- "vocab_size": 1024 + 1,
23
- "phoneme_vocab_size": 512,
24
- "EOS": 1024,
25
- }
26
-
27
- inf_tensor_value = torch.FloatTensor([-float("Inf")]).float()
28
-
29
- def logits_to_probs(
30
- logits,
31
- previous_tokens = None,
32
- temperature: float = 1.0,
33
- top_k = None,
34
- top_p = None,
35
- repetition_penalty: float = 1.0,
36
- ):
37
- previous_tokens = previous_tokens.squeeze()
38
- if previous_tokens is not None and repetition_penalty != 1.0:
39
- previous_tokens = previous_tokens.long()
40
- score = torch.gather(logits, dim=0, index=previous_tokens)
41
- score = torch.where(
42
- score < 0, score * repetition_penalty, score / repetition_penalty
43
- )
44
- logits.scatter_(dim=0, index=previous_tokens, src=score)
45
-
46
- if top_p is not None and top_p < 1.0:
47
- sorted_logits, sorted_indices = torch.sort(logits, descending=True)
48
- cum_probs = torch.cumsum(
49
- torch.nn.functional.softmax(sorted_logits, dim=-1), dim=-1
50
- )
51
- sorted_indices_to_remove = cum_probs > top_p
52
- sorted_indices_to_remove[0] = False # keep at least one option
53
- indices_to_remove = sorted_indices_to_remove.scatter(
54
- dim=0, index=sorted_indices, src=sorted_indices_to_remove
55
- )
56
- logits = logits.masked_fill(indices_to_remove, -float("Inf"))
57
-
58
- logits = logits / max(temperature, 1e-5)
59
-
60
- if top_k is not None:
61
- v, _ = torch.topk(logits, top_k)
62
- pivot = v.select(-1, -1).unsqueeze(-1)
63
- logits = torch.where(logits < pivot, inf_tensor_value, logits)
64
-
65
- probs = torch.nn.functional.softmax(logits, dim=-1)
66
- return probs
67
-
68
-
69
- def multinomial_sample_one_no_sync(
70
- probs_sort
71
- ): # Does multinomial sampling without a cuda synchronization
72
- q = torch.randn_like(probs_sort)
73
- return torch.argmax(probs_sort / q, dim=-1, keepdim=True).to(dtype=torch.int)
74
-
75
-
76
- def sample(
77
- logits,
78
- previous_tokens,
79
- **sampling_kwargs,
80
- ):
81
- probs = logits_to_probs(
82
- logits=logits, previous_tokens=previous_tokens, **sampling_kwargs
83
- )
84
- idx_next = multinomial_sample_one_no_sync(probs)
85
- return idx_next, probs
86
-
87
-
88
- class OnnxEncoder(nn.Module):
89
- def __init__(self, ar_text_embedding, bert_proj, ar_text_position):
90
- super().__init__()
91
- self.ar_text_embedding = ar_text_embedding
92
- self.bert_proj = bert_proj
93
- self.ar_text_position = ar_text_position
94
-
95
- def forward(self, x, bert_feature):
96
- x = self.ar_text_embedding(x)
97
- x = x + self.bert_proj(bert_feature.transpose(1, 2))
98
- return self.ar_text_position(x)
99
-
100
-
101
- class T2SFirstStageDecoder(nn.Module):
102
- def __init__(self, ar_audio_embedding, ar_audio_position, h, ar_predict_layer, loss_fct, ar_accuracy_metric,
103
- top_k, early_stop_num, num_layers):
104
- super().__init__()
105
- self.ar_audio_embedding = ar_audio_embedding
106
- self.ar_audio_position = ar_audio_position
107
- self.h = h
108
- self.ar_predict_layer = ar_predict_layer
109
- self.loss_fct = loss_fct
110
- self.ar_accuracy_metric = ar_accuracy_metric
111
- self.top_k = top_k
112
- self.early_stop_num = early_stop_num
113
- self.num_layers = num_layers
114
-
115
- def forward(self, x, prompt):
116
- y = prompt
117
- x_example = x[:,:,0] * 0.0
118
- #N, 1, 512
119
- cache = {
120
- "all_stage": self.num_layers,
121
- "k": None,
122
- "v": None,
123
- "y_emb": None,
124
- "first_infer": 1,
125
- "stage": 0,
126
- }
127
-
128
- y_emb = self.ar_audio_embedding(y)
129
-
130
- cache["y_emb"] = y_emb
131
- y_pos = self.ar_audio_position(y_emb)
132
-
133
- xy_pos = torch.concat([x, y_pos], dim=1)
134
-
135
- y_example = y_pos[:,:,0] * 0.0
136
- x_attn_mask = torch.matmul(x_example.transpose(0, 1) , x_example).bool()
137
- y_attn_mask = torch.ones_like(torch.matmul(y_example.transpose(0, 1), y_example), dtype=torch.int64)
138
- y_attn_mask = torch.cumsum(y_attn_mask, dim=1) - torch.cumsum(
139
- torch.ones_like(y_example.transpose(0, 1), dtype=torch.int64), dim=0
140
- )
141
- y_attn_mask = y_attn_mask > 0
142
-
143
- x_y_pad = torch.matmul(x_example.transpose(0, 1), y_example).bool()
144
- y_x_pad = torch.matmul(y_example.transpose(0, 1), x_example).bool()
145
- x_attn_mask_pad = torch.cat([x_attn_mask, torch.ones_like(x_y_pad)], dim=1)
146
- y_attn_mask = torch.cat([y_x_pad, y_attn_mask], dim=1)
147
- xy_attn_mask = torch.concat([x_attn_mask_pad, y_attn_mask], dim=0)
148
- cache["k"] = torch.matmul(x_attn_mask_pad[0].float().unsqueeze(-1), torch.zeros((1, 512)))\
149
- .unsqueeze(1).repeat(self.num_layers, 1, 1, 1)
150
- cache["v"] = torch.matmul(x_attn_mask_pad[0].float().unsqueeze(-1), torch.zeros((1, 512)))\
151
- .unsqueeze(1).repeat(self.num_layers, 1, 1, 1)
152
-
153
- xy_dec = self.h(xy_pos, mask=xy_attn_mask, cache=cache)
154
- logits = self.ar_predict_layer(xy_dec[:, -1])
155
- samples = sample(logits[0], y, top_k=self.top_k, top_p=1.0, repetition_penalty=1.35)[0].unsqueeze(0)
156
-
157
- y = torch.concat([y, samples], dim=1)
158
-
159
- return y, cache["k"], cache["v"], cache["y_emb"], x_example
160
-
161
-
162
- class T2SStageDecoder(nn.Module):
163
- def __init__(self, ar_audio_embedding, ar_audio_position, h, ar_predict_layer, loss_fct, ar_accuracy_metric,
164
- top_k, early_stop_num, num_layers):
165
- super().__init__()
166
- self.ar_audio_embedding = ar_audio_embedding
167
- self.ar_audio_position = ar_audio_position
168
- self.h = h
169
- self.ar_predict_layer = ar_predict_layer
170
- self.loss_fct = loss_fct
171
- self.ar_accuracy_metric = ar_accuracy_metric
172
- self.top_k = top_k
173
- self.early_stop_num = early_stop_num
174
- self.num_layers = num_layers
175
-
176
- def forward(self, y, k, v, y_emb, x_example):
177
- cache = {
178
- "all_stage": self.num_layers,
179
- "k": torch.nn.functional.pad(k, (0, 0, 0, 0, 0, 1)),
180
- "v": torch.nn.functional.pad(v, (0, 0, 0, 0, 0, 1)),
181
- "y_emb": y_emb,
182
- "first_infer": 0,
183
- "stage": 0,
184
- }
185
-
186
- y_emb = torch.cat(
187
- [cache["y_emb"], self.ar_audio_embedding(y[:, -1:])], 1
188
- )
189
- cache["y_emb"] = y_emb
190
- y_pos = self.ar_audio_position(y_emb)
191
-
192
- xy_pos = y_pos[:, -1:]
193
-
194
- y_example = y_pos[:,:,0] * 0.0
195
-
196
- xy_attn_mask = torch.cat([x_example, y_example], dim=1)
197
- xy_attn_mask = torch.zeros_like(xy_attn_mask, dtype=torch.bool)
198
-
199
- xy_dec = self.h(xy_pos, mask=xy_attn_mask, cache=cache)
200
- logits = self.ar_predict_layer(xy_dec[:, -1])
201
- samples = sample(logits[0], y, top_k=self.top_k, top_p=1.0, repetition_penalty=1.35)[0].unsqueeze(0)
202
-
203
- y = torch.concat([y, samples], dim=1)
204
-
205
- return y, cache["k"], cache["v"], cache["y_emb"], logits, samples
206
-
207
-
208
- class Text2SemanticDecoder(nn.Module):
209
- def __init__(self, config, norm_first=False, top_k=3):
210
- super(Text2SemanticDecoder, self).__init__()
211
- self.model_dim = config["model"]["hidden_dim"]
212
- self.embedding_dim = config["model"]["embedding_dim"]
213
- self.num_head = config["model"]["head"]
214
- self.num_layers = config["model"]["n_layer"]
215
- self.norm_first = norm_first
216
- self.vocab_size = config["model"]["vocab_size"]
217
- self.phoneme_vocab_size = config["model"]["phoneme_vocab_size"]
218
- self.p_dropout = float(config["model"]["dropout"])
219
- self.EOS = config["model"]["EOS"]
220
- self.norm_first = norm_first
221
- assert self.EOS == self.vocab_size - 1
222
- self.bert_proj = nn.Linear(1024, self.embedding_dim)
223
- self.ar_text_embedding = TokenEmbedding(self.embedding_dim, self.phoneme_vocab_size, self.p_dropout)
224
- self.ar_text_position = SinePositionalEmbedding(self.embedding_dim, dropout=0.1, scale=False, alpha=True)
225
- self.ar_audio_embedding = TokenEmbedding(self.embedding_dim, self.vocab_size, self.p_dropout)
226
- self.ar_audio_position = SinePositionalEmbedding(self.embedding_dim, dropout=0.1, scale=False, alpha=True)
227
- self.h = TransformerEncoder(
228
- TransformerEncoderLayer(
229
- d_model=self.model_dim,
230
- nhead=self.num_head,
231
- dim_feedforward=self.model_dim * 4,
232
- dropout=0.1,
233
- batch_first=True,
234
- norm_first=norm_first,
235
- ),
236
- num_layers=self.num_layers,
237
- norm=LayerNorm(self.model_dim) if norm_first else None,
238
- )
239
- self.ar_predict_layer = nn.Linear(self.model_dim, self.vocab_size, bias=False)
240
- self.loss_fct = nn.CrossEntropyLoss(reduction="sum")
241
- self.ar_accuracy_metric = MulticlassAccuracy(
242
- self.vocab_size,
243
- top_k=top_k,
244
- average="micro",
245
- multidim_average="global",
246
- ignore_index=self.EOS,
247
- )
248
- self.top_k = torch.LongTensor([1])
249
- self.early_stop_num = torch.LongTensor([-1])
250
-
251
- def init_onnx(self):
252
- self.onnx_encoder = OnnxEncoder(self.ar_text_embedding, self.bert_proj, self.ar_text_position)
253
- self.first_stage_decoder = T2SFirstStageDecoder(self.ar_audio_embedding, self.ar_audio_position, self.h,
254
- self.ar_predict_layer, self.loss_fct, self.ar_accuracy_metric, self.top_k, self.early_stop_num,
255
- self.num_layers)
256
- self.stage_decoder = T2SStageDecoder(self.ar_audio_embedding, self.ar_audio_position, self.h,
257
- self.ar_predict_layer, self.loss_fct, self.ar_accuracy_metric, self.top_k, self.early_stop_num,
258
- self.num_layers)
259
-
260
- def forward(self, x, prompts, bert_feature):
261
- early_stop_num = self.early_stop_num
262
- prefix_len = prompts.shape[1]
263
-
264
- x = self.onnx_encoder(x, bert_feature)
265
- y, k, v, y_emb, stage, x_example = self.first_stage_decoder(x, prompts)
266
-
267
- stop = False
268
- for idx in range(1, 1500):
269
- enco = self.stage_decoder(y, k, v, y_emb, stage, x_example)
270
- y, k, v, y_emb, stage, logits, samples = enco
271
- if early_stop_num != -1 and (y.shape[1] - prefix_len) > early_stop_num:
272
- stop = True
273
- if torch.argmax(logits, dim=-1)[0] == self.EOS or samples[0, 0] == self.EOS:
274
- stop = True
275
- if stop:
276
- break
277
- y[0, -1] = 0
278
- return y, idx
279
-
280
- def infer(self, x, prompts, bert_feature):
281
- top_k = self.top_k
282
- early_stop_num = self.early_stop_num
283
-
284
- x = self.onnx_encoder(x, bert_feature)
285
-
286
- y = prompts
287
- prefix_len = y.shape[1]
288
- x_len = x.shape[1]
289
- x_example = x[:,:,0] * 0.0
290
- x_attn_mask = torch.matmul(x_example.transpose(0, 1), x_example)
291
- x_attn_mask = torch.zeros_like(x_attn_mask, dtype=torch.bool)
292
-
293
- stop = False
294
- cache = {
295
- "all_stage": self.num_layers,
296
- "k": [None] * self.num_layers,
297
- "v": [None] * self.num_layers,
298
- "y_emb": None,
299
- "first_infer": 1,
300
- "stage": 0,
301
- }
302
- for idx in range(1500):
303
- if cache["first_infer"] == 1:
304
- y_emb = self.ar_audio_embedding(y)
305
- else:
306
- y_emb = torch.cat(
307
- [cache["y_emb"], self.ar_audio_embedding(y[:, -1:])], 1
308
- )
309
- cache["y_emb"] = y_emb
310
- y_pos = self.ar_audio_position(y_emb)
311
- if cache["first_infer"] == 1:
312
- xy_pos = torch.concat([x, y_pos], dim=1)
313
- else:
314
- xy_pos = y_pos[:, -1:]
315
- y_len = y_pos.shape[1]
316
- if cache["first_infer"] == 1:
317
- x_attn_mask_pad = F.pad(x_attn_mask, (0, y_len), value=True)
318
- y_attn_mask = F.pad(
319
- torch.triu(torch.ones(y_len, y_len, dtype=torch.bool), diagonal=1),
320
- (x_len, 0), value=False
321
- )
322
- xy_attn_mask = torch.concat([x_attn_mask_pad, y_attn_mask], dim=0)
323
- else:
324
- xy_attn_mask = torch.zeros((1, x_len + y_len), dtype=torch.bool)
325
- xy_dec = self.h(xy_pos, mask=xy_attn_mask, cache=cache)
326
- logits = self.ar_predict_layer(xy_dec[:, -1])
327
- samples = sample(logits[0], y, top_k=top_k, top_p=1.0, repetition_penalty=1.35)[0].unsqueeze(0)
328
- if early_stop_num != -1 and (y.shape[1] - prefix_len) > early_stop_num:
329
- stop = True
330
- if torch.argmax(logits, dim=-1)[0] == self.EOS or samples[0, 0] == self.EOS:
331
- stop = True
332
- if stop:
333
- if prompts.shape[1] == y.shape[1]:
334
- y = torch.concat([y, torch.zeros_like(samples)], dim=1)
335
- break
336
- y = torch.concat([y, samples], dim=1)
337
- cache["first_infer"] = 0
338
- return y, idx
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/models/utils.py DELETED
@@ -1,229 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/models/utils.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import torch
4
- import torch.nn.functional as F
5
- from typing import Tuple
6
-
7
- def sequence_mask(length, max_length=None):
8
- if max_length is None:
9
- max_length = length.max()
10
- x = torch.arange(max_length, dtype=length.dtype, device=length.device)
11
- return x.unsqueeze(0) < length.unsqueeze(1)
12
-
13
-
14
- def make_pad_mask(lengths: torch.Tensor, max_len: int = 0) -> torch.Tensor:
15
- """
16
- Args:
17
- lengths:
18
- A 1-D tensor containing sentence lengths.
19
- max_len:
20
- The length of masks.
21
- Returns:
22
- Return a 2-D bool tensor, where masked positions
23
- are filled with `True` and non-masked positions are
24
- filled with `False`.
25
-
26
- #>>> lengths = torch.tensor([1, 3, 2, 5])
27
- #>>> make_pad_mask(lengths)
28
- tensor([[False, True, True, True, True],
29
- [False, False, False, True, True],
30
- [False, False, True, True, True],
31
- [False, False, False, False, False]])
32
- """
33
- assert lengths.ndim == 1, lengths.ndim
34
- max_len = max(max_len, lengths.max())
35
- n = lengths.size(0)
36
- seq_range = torch.arange(0, max_len, device=lengths.device)
37
- expaned_lengths = seq_range.unsqueeze(0).expand(n, max_len)
38
-
39
- return expaned_lengths >= lengths.unsqueeze(-1)
40
-
41
-
42
- # https://github.com/microsoft/unilm/blob/master/xtune/src/transformers/modeling_utils.py
43
- def top_k_top_p_filtering(
44
- logits, top_k=0, top_p=1.0, filter_value=-float("Inf"), min_tokens_to_keep=1
45
- ):
46
- """Filter a distribution of logits using top-k and/or nucleus (top-p) filtering
47
- Args:
48
- logits: logits distribution shape (batch size, vocabulary size)
49
- if top_k > 0: keep only top k tokens with highest probability (top-k filtering).
50
- if top_p < 1.0: keep the top tokens with cumulative probability >= top_p (nucleus filtering).
51
- Nucleus filtering is described in Holtzman et al. (http://arxiv.org/abs/1904.09751)
52
- Make sure we keep at least min_tokens_to_keep per batch example in the output
53
- From: https://gist.github.com/thomwolf/1a5a29f6962089e871b94cbd09daf317
54
- """
55
- if top_k > 0:
56
- top_k = min(max(top_k, min_tokens_to_keep), logits.size(-1)) # Safety check
57
- # Remove all tokens with a probability less than the last token of the top-k
58
- indices_to_remove = logits < torch.topk(logits, top_k)[0][..., -1, None]
59
- logits[indices_to_remove] = filter_value
60
-
61
- if top_p < 1.0:
62
- sorted_logits, sorted_indices = torch.sort(logits, descending=True)
63
- cumulative_probs = torch.cumsum(F.softmax(sorted_logits, dim=-1), dim=-1)
64
-
65
- # Remove tokens with cumulative probability above the threshold (token with 0 are kept)
66
- sorted_indices_to_remove = cumulative_probs > top_p
67
- if min_tokens_to_keep > 1:
68
- # Keep at least min_tokens_to_keep (set to min_tokens_to_keep-1 because we add the first one below)
69
- sorted_indices_to_remove[..., :min_tokens_to_keep] = 0
70
- # Shift the indices to the right to keep also the first token above the threshold
71
- sorted_indices_to_remove[..., 1:] = sorted_indices_to_remove[..., :-1].clone()
72
- sorted_indices_to_remove[..., 0] = 0
73
-
74
- # scatter sorted tensors to original indexing
75
- indices_to_remove = sorted_indices_to_remove.scatter(
76
- 1, sorted_indices, sorted_indices_to_remove
77
- )
78
- logits[indices_to_remove] = filter_value
79
- return logits
80
-
81
-
82
- def topk_sampling(logits, top_k=10, top_p=1.0, temperature=1.0):
83
- # temperature: (`optional`) float
84
- # The value used to module the next token probabilities. Must be strictly positive. Default to 1.0.
85
- # top_k: (`optional`) int
86
- # The number of highest probability vocabulary tokens to keep for top-k-filtering. Between 1 and infinity. Default to 50.
87
- # top_p: (`optional`) float
88
- # The cumulative probability of parameter highest probability vocabulary tokens to keep for nucleus sampling. Must be between 0 and 1. Default to 1.
89
-
90
- # Temperature (higher temperature => more likely to sample low probability tokens)
91
- if temperature != 1.0:
92
- logits = logits / temperature
93
- # Top-p/top-k filtering
94
- logits = top_k_top_p_filtering(logits, top_k=top_k, top_p=top_p)
95
- # Sample
96
- token = torch.multinomial(F.softmax(logits, dim=-1), num_samples=1)
97
- return token
98
-
99
-
100
- from typing import Optional, Tuple
101
-
102
-
103
- def multinomial_sample_one_no_sync(
104
- probs_sort,
105
- ): # Does multinomial sampling without a cuda synchronization
106
- q = torch.empty_like(probs_sort).exponential_(1)
107
- return torch.argmax(probs_sort / q, dim=-1, keepdim=True).to(dtype=torch.int)
108
-
109
-
110
- def logits_to_probs(
111
- logits,
112
- previous_tokens: Optional[torch.Tensor] = None,
113
- temperature: float = 1.0,
114
- top_k: Optional[int] = None,
115
- top_p: Optional[int] = None,
116
- repetition_penalty: float = 1.0,
117
- ):
118
- if previous_tokens is not None:
119
- previous_tokens = previous_tokens.squeeze()
120
- # print(logits.shape,previous_tokens.shape)
121
- # pdb.set_trace()
122
- if previous_tokens is not None and repetition_penalty != 1.0:
123
- previous_tokens = previous_tokens.long()
124
- score = torch.gather(logits, dim=0, index=previous_tokens)
125
- score = torch.where(
126
- score < 0, score * repetition_penalty, score / repetition_penalty
127
- )
128
- logits.scatter_(dim=0, index=previous_tokens, src=score)
129
-
130
- if top_p is not None and top_p < 1.0:
131
- sorted_logits, sorted_indices = torch.sort(logits, descending=True)
132
- cum_probs = torch.cumsum(
133
- torch.nn.functional.softmax(sorted_logits, dim=-1), dim=-1
134
- )
135
- sorted_indices_to_remove = cum_probs > top_p
136
- sorted_indices_to_remove[0] = False # keep at least one option
137
- indices_to_remove = sorted_indices_to_remove.scatter(
138
- dim=0, index=sorted_indices, src=sorted_indices_to_remove
139
- )
140
- logits = logits.masked_fill(indices_to_remove, -float("Inf"))
141
-
142
- logits = logits / max(temperature, 1e-5)
143
-
144
- if top_k is not None:
145
- v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
146
- pivot = v.select(-1, -1).unsqueeze(-1)
147
- logits = torch.where(logits < pivot, -float("Inf"), logits)
148
-
149
- probs = torch.nn.functional.softmax(logits, dim=-1)
150
- return probs
151
-
152
-
153
- def sample(
154
- logits,
155
- previous_tokens: Optional[torch.Tensor] = None,
156
- **sampling_kwargs,
157
- ) -> Tuple[torch.Tensor, torch.Tensor]:
158
- probs = logits_to_probs(
159
- logits=logits, previous_tokens=previous_tokens, **sampling_kwargs
160
- )
161
- idx_next = multinomial_sample_one_no_sync(probs)
162
- return idx_next, probs
163
-
164
- def dpo_loss(policy_chosen_logps: torch.FloatTensor,
165
- policy_rejected_logps: torch.FloatTensor,
166
- reference_chosen_logps: torch.FloatTensor,
167
- reference_rejected_logps: torch.FloatTensor,
168
- beta: float,
169
- reference_free: bool = False) -> Tuple[torch.FloatTensor, torch.FloatTensor, torch.FloatTensor]:
170
- pi_logratios = policy_chosen_logps - policy_rejected_logps
171
- ref_logratios = reference_chosen_logps - reference_rejected_logps
172
-
173
- if reference_free:
174
- ref_logratios = 0
175
-
176
- logits = pi_logratios - ref_logratios
177
-
178
- losses = -F.logsigmoid(beta * logits)
179
- chosen_rewards = beta * (policy_chosen_logps - reference_chosen_logps).detach()
180
- rejected_rewards = beta * (policy_rejected_logps - reference_rejected_logps).detach()
181
-
182
- return losses.mean(), chosen_rewards, rejected_rewards
183
-
184
- def get_batch_logps(logits_target: torch.FloatTensor, logits_reject: torch.FloatTensor, labels_target: torch.LongTensor, labels_reject: torch.LongTensor, average_log_prob: bool = False) -> Tuple[torch.FloatTensor, torch.FloatTensor]:
185
-
186
- # dummy token; we'll ignore the losses on these tokens later
187
-
188
- per_token_logps_target = torch.gather(logits_target.log_softmax(-1), dim=2, index=labels_target.unsqueeze(2)).squeeze(2)
189
- per_token_logps_reject = torch.gather(logits_reject.log_softmax(-1), dim=2, index=labels_reject.unsqueeze(2)).squeeze(2)
190
-
191
- return per_token_logps_target.sum(-1), per_token_logps_reject.sum(-1)
192
-
193
- def make_reject_y(y_o, y_lens):
194
- def repeat_P(y):
195
- range_idx, _ = torch.randint(0, len(y), size=(2,)).sort()
196
- pre = y[:range_idx[0]]
197
- shf = y[range_idx[1]:]
198
- range_text = y[range_idx[0]:range_idx[1]]
199
- new_y = torch.cat([pre, range_text, range_text, shf])
200
- return new_y
201
- def lost_P(y):
202
- range_idx, _ = torch.randint(0, len(y), size=(2,)).sort()
203
- pre = y[:range_idx[0]]
204
- shf = y[range_idx[1]:]
205
- range_text = y[range_idx[0]:range_idx[1]]
206
- new_y = torch.cat([pre, shf])
207
- return new_y
208
- bs = len(y_lens)
209
- reject_y = []
210
- reject_y_lens = []
211
- for b in range(bs):
212
- process_item_idx = torch.randint(0, 1, size=(1, ))[0]
213
- if process_item_idx == 0:
214
- new_y = repeat_P(y_o[b])
215
- reject_y.append(new_y)
216
- reject_y_lens.append(len(new_y))
217
- elif process_item_idx==1:
218
- new_y = lost_P(y_o[b])
219
- reject_y.append(new_y)
220
- reject_y_lens.append(len(new_y))
221
- max_length = max(reject_y_lens)
222
- for b in range(bs):
223
- pad_length = max_length - reject_y_lens[b]
224
- reject_y[b] = torch.cat([reject_y[b], torch.zeros(pad_length, dtype=y_o.dtype, device=y_o.device)], dim=0)
225
-
226
- reject_y = torch.stack(reject_y, dim = 0)
227
- reject_y_lens = torch.tensor(reject_y_lens, device=y_lens.device)
228
-
229
- return reject_y, reject_y_lens
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/__init__.py DELETED
File without changes
AR/modules/__pycache__/__init__.cpython-39.pyc DELETED
Binary file (151 Bytes)
 
AR/modules/__pycache__/activation.cpython-39.pyc DELETED
Binary file (14 kB)
 
AR/modules/__pycache__/embedding.cpython-39.pyc DELETED
Binary file (3.05 kB)
 
AR/modules/__pycache__/lr_schedulers.cpython-39.pyc DELETED
Binary file (2.37 kB)
 
AR/modules/__pycache__/optim.cpython-39.pyc DELETED
Binary file (18.9 kB)
 
AR/modules/__pycache__/patched_mha_with_cache.cpython-39.pyc DELETED
Binary file (11.3 kB)
 
AR/modules/__pycache__/scaling.cpython-39.pyc DELETED
Binary file (9.51 kB)
 
AR/modules/__pycache__/transformer.cpython-39.pyc DELETED
Binary file (9.79 kB)
 
AR/modules/activation.py DELETED
@@ -1,428 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/activation.py
2
- from typing import Optional
3
- from typing import Tuple
4
- import torch
5
- from torch import Tensor
6
- from torch.nn import Linear
7
- from torch.nn import Module
8
- from torch.nn.init import constant_
9
- from torch.nn.init import xavier_normal_
10
- from torch.nn.init import xavier_uniform_
11
- from torch.nn.modules.linear import NonDynamicallyQuantizableLinear
12
- from torch.nn.parameter import Parameter
13
-
14
- from torch.nn import functional as F
15
- from AR.modules.patched_mha_with_cache import multi_head_attention_forward_patched
16
-
17
- F.multi_head_attention_forward = multi_head_attention_forward_patched
18
-
19
-
20
- class MultiheadAttention(Module):
21
- r"""Allows the model to jointly attend to information
22
- from different representation subspaces as described in the paper:
23
- `Attention Is All You Need <https://arxiv.org/abs/1706.03762>`_.
24
-
25
- Multi-Head Attention is defined as:
26
-
27
- .. math::
28
- \text{MultiHead}(Q, K, V) = \text{Concat}(head_1,\dots,head_h)W^O
29
-
30
- where :math:`head_i = \text{Attention}(QW_i^Q, KW_i^K, VW_i^V)`.
31
-
32
- ``forward()`` will use a special optimized implementation if all of the following
33
- conditions are met:
34
-
35
- - self attention is being computed (i.e., ``query``, ``key``, and ``value`` are the same tensor. This
36
- restriction will be loosened in the future.)
37
- - Either autograd is disabled (using ``torch.inference_mode`` or ``torch.no_grad``) or no tensor argument ``requires_grad``
38
- - training is disabled (using ``.eval()``)
39
- - dropout is 0
40
- - ``add_bias_kv`` is ``False``
41
- - ``add_zero_attn`` is ``False``
42
- - ``batch_first`` is ``True`` and the input is batched
43
- - ``kdim`` and ``vdim`` are equal to ``embed_dim``
44
- - at most one of ``key_padding_mask`` or ``attn_mask`` is passed
45
- - if a `NestedTensor <https://pytorch.org/docs/stable/nested.html>`_ is passed, neither ``key_padding_mask``
46
- nor ``attn_mask`` is passed
47
-
48
- If the optimized implementation is in use, a
49
- `NestedTensor <https://pytorch.org/docs/stable/nested.html>`_ can be passed for
50
- ``query``/``key``/``value`` to represent padding more efficiently than using a
51
- padding mask. In this case, a `NestedTensor <https://pytorch.org/docs/stable/nested.html>`_
52
- will be returned, and an additional speedup proportional to the fraction of the input
53
- that is padding can be expected.
54
-
55
- Args:
56
- embed_dim: Total dimension of the model.
57
- num_heads: Number of parallel attention heads. Note that ``embed_dim`` will be split
58
- across ``num_heads`` (i.e. each head will have dimension ``embed_dim // num_heads``).
59
- dropout: Dropout probability on ``attn_output_weights``. Default: ``0.0`` (no dropout).
60
- bias: If specified, adds bias to input / output projection layers. Default: ``True``.
61
- add_bias_kv: If specified, adds bias to the key and value sequences at dim=0. Default: ``False``.
62
- add_zero_attn: If specified, adds a new batch of zeros to the key and value sequences at dim=1.
63
- Default: ``False``.
64
- kdim: Total number of features for keys. Default: ``None`` (uses ``kdim=embed_dim``).
65
- vdim: Total number of features for values. Default: ``None`` (uses ``vdim=embed_dim``).
66
- batch_first: If ``True``, then the input and output tensors are provided
67
- as (batch, seq, feature). Default: ``False`` (seq, batch, feature).
68
-
69
- Examples::
70
-
71
- >>> # xdoctest: +SKIP
72
- >>> multihead_attn = nn.MultiheadAttention(embed_dim, num_heads)
73
- >>> attn_output, attn_output_weights = multihead_attn(query, key, value)
74
-
75
- """
76
- __constants__ = ["batch_first"]
77
- bias_k: Optional[torch.Tensor]
78
- bias_v: Optional[torch.Tensor]
79
-
80
- def __init__(
81
- self,
82
- embed_dim,
83
- num_heads,
84
- dropout=0.0,
85
- bias=True,
86
- add_bias_kv=False,
87
- add_zero_attn=False,
88
- kdim=None,
89
- vdim=None,
90
- batch_first=False,
91
- linear1_cls=Linear,
92
- linear2_cls=Linear,
93
- device=None,
94
- dtype=None,
95
- ) -> None:
96
- factory_kwargs = {"device": device, "dtype": dtype}
97
- super(MultiheadAttention, self).__init__()
98
- self.embed_dim = embed_dim
99
- self.kdim = kdim if kdim is not None else embed_dim
100
- self.vdim = vdim if vdim is not None else embed_dim
101
- self._qkv_same_embed_dim = self.kdim == embed_dim and self.vdim == embed_dim
102
-
103
- self.num_heads = num_heads
104
- self.dropout = dropout
105
- self.batch_first = batch_first
106
- self.head_dim = embed_dim // num_heads
107
- assert (
108
- self.head_dim * num_heads == self.embed_dim
109
- ), "embed_dim must be divisible by num_heads"
110
-
111
- if add_bias_kv:
112
- self.bias_k = Parameter(torch.empty((1, 1, embed_dim), **factory_kwargs))
113
- self.bias_v = Parameter(torch.empty((1, 1, embed_dim), **factory_kwargs))
114
- else:
115
- self.bias_k = self.bias_v = None
116
-
117
- if linear1_cls == Linear:
118
- if not self._qkv_same_embed_dim:
119
- self.q_proj_weight = Parameter(
120
- torch.empty((embed_dim, embed_dim), **factory_kwargs)
121
- )
122
- self.k_proj_weight = Parameter(
123
- torch.empty((embed_dim, self.kdim), **factory_kwargs)
124
- )
125
- self.v_proj_weight = Parameter(
126
- torch.empty((embed_dim, self.vdim), **factory_kwargs)
127
- )
128
- self.register_parameter("in_proj_weight", None)
129
- else:
130
- self.in_proj_weight = Parameter(
131
- torch.empty((3 * embed_dim, embed_dim), **factory_kwargs)
132
- )
133
- self.register_parameter("q_proj_weight", None)
134
- self.register_parameter("k_proj_weight", None)
135
- self.register_parameter("v_proj_weight", None)
136
-
137
- if bias:
138
- self.in_proj_bias = Parameter(
139
- torch.empty(3 * embed_dim, **factory_kwargs)
140
- )
141
- else:
142
- self.register_parameter("in_proj_bias", None)
143
- self.out_proj = NonDynamicallyQuantizableLinear(
144
- embed_dim, embed_dim, bias=bias, **factory_kwargs
145
- )
146
-
147
- self._reset_parameters()
148
- else:
149
- if not self._qkv_same_embed_dim:
150
- raise NotImplementedError
151
- else:
152
- self.in_proj_linear = linear1_cls(
153
- embed_dim, 3 * embed_dim, bias=bias, **factory_kwargs
154
- )
155
- self.in_proj_weight = self.in_proj_linear.weight
156
-
157
- self.register_parameter("q_proj_weight", None)
158
- self.register_parameter("k_proj_weight", None)
159
- self.register_parameter("v_proj_weight", None)
160
-
161
- if bias:
162
- self.in_proj_bias = self.in_proj_linear.bias
163
- else:
164
- self.register_parameter("in_proj_bias", None)
165
-
166
- self.out_proj = linear2_cls(
167
- embed_dim, embed_dim, bias=bias, **factory_kwargs
168
- )
169
-
170
- if self.bias_k is not None:
171
- xavier_normal_(self.bias_k)
172
- if self.bias_v is not None:
173
- xavier_normal_(self.bias_v)
174
-
175
- self.add_zero_attn = add_zero_attn
176
-
177
- def _reset_parameters(self):
178
- if self._qkv_same_embed_dim:
179
- xavier_uniform_(self.in_proj_weight)
180
- else:
181
- xavier_uniform_(self.q_proj_weight)
182
- xavier_uniform_(self.k_proj_weight)
183
- xavier_uniform_(self.v_proj_weight)
184
-
185
- if self.in_proj_bias is not None:
186
- constant_(self.in_proj_bias, 0.0)
187
- constant_(self.out_proj.bias, 0.0)
188
-
189
- if self.bias_k is not None:
190
- xavier_normal_(self.bias_k)
191
- if self.bias_v is not None:
192
- xavier_normal_(self.bias_v)
193
-
194
- def __setstate__(self, state):
195
- # Support loading old MultiheadAttention checkpoints generated by v1.1.0
196
- if "_qkv_same_embed_dim" not in state:
197
- state["_qkv_same_embed_dim"] = True
198
-
199
- super(MultiheadAttention, self).__setstate__(state)
200
-
201
- def forward(
202
- self,
203
- query: Tensor,
204
- key: Tensor,
205
- value: Tensor,
206
- key_padding_mask: Optional[Tensor] = None,
207
- need_weights: bool = True,
208
- attn_mask: Optional[Tensor] = None,
209
- average_attn_weights: bool = True,
210
- cache=None,
211
- ) -> Tuple[Tensor, Optional[Tensor]]:
212
- r"""
213
- Args:
214
- query: Query embeddings of shape :math:`(L, E_q)` for unbatched input, :math:`(L, N, E_q)` when ``batch_first=False``
215
- or :math:`(N, L, E_q)` when ``batch_first=True``, where :math:`L` is the target sequence length,
216
- :math:`N` is the batch size, and :math:`E_q` is the query embedding dimension ``embed_dim``.
217
- Queries are compared against key-value pairs to produce the output.
218
- See "Attention Is All You Need" for more details.
219
- key: Key embeddings of shape :math:`(S, E_k)` for unbatched input, :math:`(S, N, E_k)` when ``batch_first=False``
220
- or :math:`(N, S, E_k)` when ``batch_first=True``, where :math:`S` is the source sequence length,
221
- :math:`N` is the batch size, and :math:`E_k` is the key embedding dimension ``kdim``.
222
- See "Attention Is All You Need" for more details.
223
- value: Value embeddings of shape :math:`(S, E_v)` for unbatched input, :math:`(S, N, E_v)` when
224
- ``batch_first=False`` or :math:`(N, S, E_v)` when ``batch_first=True``, where :math:`S` is the source
225
- sequence length, :math:`N` is the batch size, and :math:`E_v` is the value embedding dimension ``vdim``.
226
- See "Attention Is All You Need" for more details.
227
- key_padding_mask: If specified, a mask of shape :math:`(N, S)` indicating which elements within ``key``
228
- to ignore for the purpose of attention (i.e. treat as "padding"). For unbatched `query`, shape should be :math:`(S)`.
229
- Binary and byte masks are supported.
230
- For a binary mask, a ``True`` value indicates that the corresponding ``key`` value will be ignored for
231
- the purpose of attention. For a float mask, it will be directly added to the corresponding ``key`` value.
232
- need_weights: If specified, returns ``attn_output_weights`` in addition to ``attn_outputs``.
233
- Default: ``True``.
234
- attn_mask: If specified, a 2D or 3D mask preventing attention to certain positions. Must be of shape
235
- :math:`(L, S)` or :math:`(N\cdot\text{num\_heads}, L, S)`, where :math:`N` is the batch size,
236
- :math:`L` is the target sequence length, and :math:`S` is the source sequence length. A 2D mask will be
237
- broadcasted across the batch while a 3D mask allows for a different mask for each entry in the batch.
238
- Binary, byte, and float masks are supported. For a binary mask, a ``True`` value indicates that the
239
- corresponding position is not allowed to attend. For a byte mask, a non-zero value indicates that the
240
- corresponding position is not allowed to attend. For a float mask, the mask values will be added to
241
- the attention weight.
242
- average_attn_weights: If true, indicates that the returned ``attn_weights`` should be averaged across
243
- heads. Otherwise, ``attn_weights`` are provided separately per head. Note that this flag only has an
244
- effect when ``need_weights=True``. Default: ``True`` (i.e. average weights across heads)
245
-
246
- Outputs:
247
- - **attn_output** - Attention outputs of shape :math:`(L, E)` when input is unbatched,
248
- :math:`(L, N, E)` when ``batch_first=False`` or :math:`(N, L, E)` when ``batch_first=True``,
249
- where :math:`L` is the target sequence length, :math:`N` is the batch size, and :math:`E` is the
250
- embedding dimension ``embed_dim``.
251
- - **attn_output_weights** - Only returned when ``need_weights=True``. If ``average_attn_weights=True``,
252
- returns attention weights averaged across heads of shape :math:`(L, S)` when input is unbatched or
253
- :math:`(N, L, S)`, where :math:`N` is the batch size, :math:`L` is the target sequence length, and
254
- :math:`S` is the source sequence length. If ``average_attn_weights=False``, returns attention weights per
255
- head of shape :math:`(\text{num\_heads}, L, S)` when input is unbatched or :math:`(N, \text{num\_heads}, L, S)`.
256
-
257
- .. note::
258
- `batch_first` argument is ignored for unbatched inputs.
259
- """
260
- is_batched = query.dim() == 3
261
- if key_padding_mask is not None:
262
- _kpm_dtype = key_padding_mask.dtype
263
- if _kpm_dtype != torch.bool and not torch.is_floating_point(
264
- key_padding_mask
265
- ):
266
- raise AssertionError(
267
- "only bool and floating types of key_padding_mask are supported"
268
- )
269
- why_not_fast_path = ""
270
- if not is_batched:
271
- why_not_fast_path = (
272
- f"input not batched; expected query.dim() of 3 but got {query.dim()}"
273
- )
274
- elif query is not key or key is not value:
275
- # When lifting this restriction, don't forget to either
276
- # enforce that the dtypes all match or test cases where
277
- # they don't!
278
- why_not_fast_path = "non-self attention was used (query, key, and value are not the same Tensor)"
279
- elif self.in_proj_bias is not None and query.dtype != self.in_proj_bias.dtype:
280
- why_not_fast_path = f"dtypes of query ({query.dtype}) and self.in_proj_bias ({self.in_proj_bias.dtype}) don't match"
281
- elif (
282
- self.in_proj_weight is not None and query.dtype != self.in_proj_weight.dtype
283
- ):
284
- # this case will fail anyway, but at least they'll get a useful error message.
285
- why_not_fast_path = f"dtypes of query ({query.dtype}) and self.in_proj_weight ({self.in_proj_weight.dtype}) don't match"
286
- elif self.training:
287
- why_not_fast_path = "training is enabled"
288
- elif not self.batch_first:
289
- why_not_fast_path = "batch_first was not True"
290
- elif self.bias_k is not None:
291
- why_not_fast_path = "self.bias_k was not None"
292
- elif self.bias_v is not None:
293
- why_not_fast_path = "self.bias_v was not None"
294
- elif self.dropout:
295
- why_not_fast_path = f"dropout was {self.dropout}, required zero"
296
- elif self.add_zero_attn:
297
- why_not_fast_path = "add_zero_attn was enabled"
298
- elif not self._qkv_same_embed_dim:
299
- why_not_fast_path = "_qkv_same_embed_dim was not True"
300
- elif attn_mask is not None:
301
- why_not_fast_path = "attn_mask was not None"
302
- elif query.is_nested and key_padding_mask is not None:
303
- why_not_fast_path = (
304
- "key_padding_mask is not supported with NestedTensor input"
305
- )
306
- elif self.num_heads % 2 == 1:
307
- why_not_fast_path = "num_heads is odd"
308
- elif torch.is_autocast_enabled():
309
- why_not_fast_path = "autocast is enabled"
310
-
311
- if not why_not_fast_path:
312
- tensor_args = (
313
- query,
314
- key,
315
- value,
316
- self.in_proj_weight,
317
- self.in_proj_bias,
318
- self.out_proj.weight,
319
- self.out_proj.bias,
320
- )
321
- # We have to use list comprehensions below because TorchScript does not support
322
- # generator expressions.
323
- if torch.overrides.has_torch_function(tensor_args):
324
- why_not_fast_path = "some Tensor argument has_torch_function"
325
- elif not all(
326
- [
327
- (x is None or x.is_cuda or "cpu" in str(x.device))
328
- for x in tensor_args
329
- ]
330
- ):
331
- why_not_fast_path = "some Tensor argument is neither CUDA nor CPU"
332
- elif torch.is_grad_enabled() and any(
333
- [x is not None and x.requires_grad for x in tensor_args]
334
- ):
335
- why_not_fast_path = (
336
- "grad is enabled and at least one of query or the "
337
- "input/output projection weights or biases requires_grad"
338
- )
339
- if not why_not_fast_path:
340
- return torch._native_multi_head_attention(
341
- query,
342
- key,
343
- value,
344
- self.embed_dim,
345
- self.num_heads,
346
- self.in_proj_weight,
347
- self.in_proj_bias,
348
- self.out_proj.weight,
349
- self.out_proj.bias,
350
- key_padding_mask if key_padding_mask is not None else attn_mask,
351
- need_weights,
352
- average_attn_weights,
353
- 1
354
- if key_padding_mask is not None
355
- else 0
356
- if attn_mask is not None
357
- else None,
358
- )
359
-
360
- any_nested = query.is_nested or key.is_nested or value.is_nested
361
- assert not any_nested, (
362
- "MultiheadAttention does not support NestedTensor outside of its fast path. "
363
- + f"The fast path was not hit because {why_not_fast_path}"
364
- )
365
-
366
- if self.batch_first and is_batched:
367
- # make sure that the transpose op does not affect the "is" property
368
- if key is value:
369
- if query is key:
370
- query = key = value = query.transpose(1, 0)
371
- else:
372
- query, key = [x.transpose(1, 0) for x in (query, key)]
373
- value = key
374
- else:
375
- query, key, value = [x.transpose(1, 0) for x in (query, key, value)]
376
-
377
- if not self._qkv_same_embed_dim:
378
- attn_output, attn_output_weights = F.multi_head_attention_forward(
379
- query,
380
- key,
381
- value,
382
- self.embed_dim,
383
- self.num_heads,
384
- self.in_proj_weight,
385
- self.in_proj_bias,
386
- self.bias_k,
387
- self.bias_v,
388
- self.add_zero_attn,
389
- self.dropout,
390
- self.out_proj.weight,
391
- self.out_proj.bias,
392
- training=self.training,
393
- key_padding_mask=key_padding_mask,
394
- need_weights=need_weights,
395
- attn_mask=attn_mask,
396
- use_separate_proj_weight=True,
397
- q_proj_weight=self.q_proj_weight,
398
- k_proj_weight=self.k_proj_weight,
399
- v_proj_weight=self.v_proj_weight,
400
- average_attn_weights=average_attn_weights,
401
- cache=cache,
402
- )
403
- else:
404
- attn_output, attn_output_weights = F.multi_head_attention_forward(
405
- query,
406
- key,
407
- value,
408
- self.embed_dim,
409
- self.num_heads,
410
- self.in_proj_weight,
411
- self.in_proj_bias,
412
- self.bias_k,
413
- self.bias_v,
414
- self.add_zero_attn,
415
- self.dropout,
416
- self.out_proj.weight,
417
- self.out_proj.bias,
418
- training=self.training,
419
- key_padding_mask=key_padding_mask,
420
- need_weights=need_weights,
421
- attn_mask=attn_mask,
422
- average_attn_weights=average_attn_weights,
423
- cache=cache,
424
- )
425
- if self.batch_first and is_batched:
426
- return attn_output.transpose(1, 0), attn_output_weights
427
- else:
428
- return attn_output, attn_output_weights
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/activation_onnx.py DELETED
@@ -1,178 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/activation.py
2
- from typing import Optional
3
- from typing import Tuple
4
- import torch
5
- from torch import Tensor
6
- from torch.nn import Linear
7
- from torch.nn import Module
8
- from torch.nn.init import constant_
9
- from torch.nn.init import xavier_normal_
10
- from torch.nn.init import xavier_uniform_
11
- from torch.nn.modules.linear import NonDynamicallyQuantizableLinear
12
- from torch.nn.parameter import Parameter
13
-
14
- from torch.nn import functional as F
15
- from AR.modules.patched_mha_with_cache_onnx import multi_head_attention_forward_patched
16
-
17
-
18
- class MultiheadAttention(Module):
19
- __constants__ = ["batch_first"]
20
- bias_k: Optional[torch.Tensor]
21
- bias_v: Optional[torch.Tensor]
22
-
23
- def __init__(
24
- self,
25
- embed_dim,
26
- num_heads,
27
- dropout=0.0,
28
- bias=True,
29
- add_bias_kv=False,
30
- add_zero_attn=False,
31
- kdim=None,
32
- vdim=None,
33
- batch_first=False,
34
- linear1_cls=Linear,
35
- linear2_cls=Linear,
36
- device=None,
37
- dtype=None,
38
- ) -> None:
39
- factory_kwargs = {"device": device, "dtype": dtype}
40
- super(MultiheadAttention, self).__init__()
41
- self.embed_dim = embed_dim
42
- self.kdim = kdim if kdim is not None else embed_dim
43
- self.vdim = vdim if vdim is not None else embed_dim
44
- self._qkv_same_embed_dim = self.kdim == embed_dim and self.vdim == embed_dim
45
-
46
- self.num_heads = num_heads
47
- self.dropout = dropout
48
- self.batch_first = batch_first
49
- self.head_dim = embed_dim // num_heads
50
- assert (
51
- self.head_dim * num_heads == self.embed_dim
52
- ), "embed_dim must be divisible by num_heads"
53
-
54
- if add_bias_kv:
55
- self.bias_k = Parameter(torch.empty((1, 1, embed_dim), **factory_kwargs))
56
- self.bias_v = Parameter(torch.empty((1, 1, embed_dim), **factory_kwargs))
57
- else:
58
- self.bias_k = self.bias_v = None
59
-
60
- if linear1_cls == Linear:
61
- if not self._qkv_same_embed_dim:
62
- self.q_proj_weight = Parameter(
63
- torch.empty((embed_dim, embed_dim), **factory_kwargs)
64
- )
65
- self.k_proj_weight = Parameter(
66
- torch.empty((embed_dim, self.kdim), **factory_kwargs)
67
- )
68
- self.v_proj_weight = Parameter(
69
- torch.empty((embed_dim, self.vdim), **factory_kwargs)
70
- )
71
- self.register_parameter("in_proj_weight", None)
72
- else:
73
- self.in_proj_weight = Parameter(
74
- torch.empty((3 * embed_dim, embed_dim), **factory_kwargs)
75
- )
76
- self.register_parameter("q_proj_weight", None)
77
- self.register_parameter("k_proj_weight", None)
78
- self.register_parameter("v_proj_weight", None)
79
-
80
- if bias:
81
- self.in_proj_bias = Parameter(
82
- torch.empty(3 * embed_dim, **factory_kwargs)
83
- )
84
- else:
85
- self.register_parameter("in_proj_bias", None)
86
- self.out_proj = NonDynamicallyQuantizableLinear(
87
- embed_dim, embed_dim, bias=bias, **factory_kwargs
88
- )
89
-
90
- self._reset_parameters()
91
- else:
92
- if not self._qkv_same_embed_dim:
93
- raise NotImplementedError
94
- else:
95
- self.in_proj_linear = linear1_cls(
96
- embed_dim, 3 * embed_dim, bias=bias, **factory_kwargs
97
- )
98
- self.in_proj_weight = self.in_proj_linear.weight
99
-
100
- self.register_parameter("q_proj_weight", None)
101
- self.register_parameter("k_proj_weight", None)
102
- self.register_parameter("v_proj_weight", None)
103
-
104
- if bias:
105
- self.in_proj_bias = self.in_proj_linear.bias
106
- else:
107
- self.register_parameter("in_proj_bias", None)
108
-
109
- self.out_proj = linear2_cls(
110
- embed_dim, embed_dim, bias=bias, **factory_kwargs
111
- )
112
-
113
- if self.bias_k is not None:
114
- xavier_normal_(self.bias_k)
115
- if self.bias_v is not None:
116
- xavier_normal_(self.bias_v)
117
-
118
- self.add_zero_attn = add_zero_attn
119
-
120
- def _reset_parameters(self):
121
- if self._qkv_same_embed_dim:
122
- xavier_uniform_(self.in_proj_weight)
123
- else:
124
- xavier_uniform_(self.q_proj_weight)
125
- xavier_uniform_(self.k_proj_weight)
126
- xavier_uniform_(self.v_proj_weight)
127
-
128
- if self.in_proj_bias is not None:
129
- constant_(self.in_proj_bias, 0.0)
130
- constant_(self.out_proj.bias, 0.0)
131
-
132
- if self.bias_k is not None:
133
- xavier_normal_(self.bias_k)
134
- if self.bias_v is not None:
135
- xavier_normal_(self.bias_v)
136
-
137
- def __setstate__(self, state):
138
- # Support loading old MultiheadAttention checkpoints generated by v1.1.0
139
- if "_qkv_same_embed_dim" not in state:
140
- state["_qkv_same_embed_dim"] = True
141
-
142
- super(MultiheadAttention, self).__setstate__(state)
143
-
144
- def forward(
145
- self,
146
- query: Tensor,
147
- key: Tensor,
148
- value: Tensor,
149
- key_padding_mask: Optional[Tensor] = None,
150
- need_weights: bool = True,
151
- attn_mask: Optional[Tensor] = None,
152
- average_attn_weights: bool = True,
153
- cache=None,
154
- ) -> Tuple[Tensor, Optional[Tensor]]:
155
- any_nested = query.is_nested or key.is_nested or value.is_nested
156
- query = key = value = query.transpose(1, 0)
157
- attn_output = multi_head_attention_forward_patched(
158
- query,
159
- key,
160
- value,
161
- self.embed_dim,
162
- self.num_heads,
163
- self.in_proj_weight,
164
- self.in_proj_bias,
165
- self.bias_k,
166
- self.bias_v,
167
- self.add_zero_attn,
168
- self.dropout,
169
- self.out_proj.weight,
170
- self.out_proj.bias,
171
- training=self.training,
172
- key_padding_mask=key_padding_mask,
173
- need_weights=need_weights,
174
- attn_mask=attn_mask,
175
- average_attn_weights=average_attn_weights,
176
- cache=cache,
177
- )
178
- return attn_output.transpose(1, 0)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/embedding.py DELETED
@@ -1,81 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/embedding.py
2
- import math
3
-
4
- import torch
5
- from torch import nn
6
-
7
-
8
- class TokenEmbedding(nn.Module):
9
- def __init__(
10
- self,
11
- embedding_dim: int,
12
- vocab_size: int,
13
- dropout: float = 0.0,
14
- ):
15
- super().__init__()
16
-
17
- self.vocab_size = vocab_size
18
- self.embedding_dim = embedding_dim
19
-
20
- self.dropout = torch.nn.Dropout(p=dropout)
21
- self.word_embeddings = nn.Embedding(self.vocab_size, self.embedding_dim)
22
-
23
- @property
24
- def weight(self) -> torch.Tensor:
25
- return self.word_embeddings.weight
26
-
27
- def embedding(self, index: int) -> torch.Tensor:
28
- return self.word_embeddings.weight[index : index + 1]
29
-
30
- def forward(self, x: torch.Tensor):
31
- x = self.word_embeddings(x)
32
- x = self.dropout(x)
33
- return x
34
-
35
-
36
- class SinePositionalEmbedding(nn.Module):
37
- def __init__(
38
- self,
39
- embedding_dim: int,
40
- dropout: float = 0.0,
41
- scale: bool = False,
42
- alpha: bool = False,
43
- ):
44
- super().__init__()
45
- self.embedding_dim = embedding_dim
46
- self.x_scale = math.sqrt(embedding_dim) if scale else 1.0
47
- self.alpha = nn.Parameter(torch.ones(1), requires_grad=alpha)
48
- self.dropout = torch.nn.Dropout(p=dropout)
49
-
50
- self.reverse = False
51
- self.pe = None
52
- self.extend_pe(torch.tensor(0.0).expand(1, 4000))
53
-
54
- def extend_pe(self, x):
55
- """Reset the positional encodings."""
56
- if self.pe is not None:
57
- if self.pe.size(1) >= x.size(1):
58
- if self.pe.dtype != x.dtype or self.pe.device != x.device:
59
- self.pe = self.pe.to(dtype=x.dtype, device=x.device)
60
- return
61
- pe = torch.zeros(x.size(1), self.embedding_dim)
62
- if self.reverse:
63
- position = torch.arange(
64
- x.size(1) - 1, -1, -1.0, dtype=torch.float32
65
- ).unsqueeze(1)
66
- else:
67
- position = torch.arange(0, x.size(1), dtype=torch.float32).unsqueeze(1)
68
- div_term = torch.exp(
69
- torch.arange(0, self.embedding_dim, 2, dtype=torch.float32)
70
- * -(math.log(10000.0) / self.embedding_dim)
71
- )
72
- pe[:, 0::2] = torch.sin(position * div_term)
73
- pe[:, 1::2] = torch.cos(position * div_term)
74
- pe = pe.unsqueeze(0)
75
- self.pe = pe.to(device=x.device, dtype=x.dtype).detach()
76
-
77
- def forward(self, x: torch.Tensor) -> torch.Tensor:
78
- self.extend_pe(x)
79
- output = x.unsqueeze(-1) if x.ndim == 2 else x
80
- output = output * self.x_scale + self.alpha * self.pe[:, : x.size(1)]
81
- return self.dropout(output)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/embedding_onnx.py DELETED
@@ -1,63 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/embedding.py
2
- import math
3
-
4
- import torch
5
- from torch import nn
6
-
7
-
8
- class TokenEmbedding(nn.Module):
9
- def __init__(
10
- self,
11
- embedding_dim: int,
12
- vocab_size: int,
13
- dropout: float = 0.0,
14
- ):
15
- super().__init__()
16
-
17
- self.vocab_size = vocab_size
18
- self.embedding_dim = embedding_dim
19
-
20
- self.dropout = torch.nn.Dropout(p=dropout)
21
- self.word_embeddings = nn.Embedding(self.vocab_size, self.embedding_dim)
22
-
23
- @property
24
- def weight(self) -> torch.Tensor:
25
- return self.word_embeddings.weight
26
-
27
- def embedding(self, index: int) -> torch.Tensor:
28
- return self.word_embeddings.weight[index : index + 1]
29
-
30
- def forward(self, x: torch.Tensor):
31
- x = self.word_embeddings(x)
32
- x = self.dropout(x)
33
- return x
34
-
35
-
36
- class SinePositionalEmbedding(nn.Module):
37
- def __init__(
38
- self,
39
- embedding_dim: int,
40
- dropout: float = 0.0,
41
- scale: bool = False,
42
- alpha: bool = False,
43
- ):
44
- super().__init__()
45
- self.embedding_dim = embedding_dim
46
- self.x_scale = math.sqrt(embedding_dim) if scale else 1.0
47
- self.alpha = nn.Parameter(torch.ones(1), requires_grad=alpha)
48
- self.dropout = torch.nn.Dropout(p=dropout)
49
- self.reverse = False
50
- self.div_term = torch.exp(torch.arange(0, self.embedding_dim, 2) * -(math.log(10000.0) / self.embedding_dim))
51
-
52
- def extend_pe(self, x):
53
- position = torch.cumsum(torch.ones_like(x[:,:,0]), dim=1).transpose(0, 1)
54
- scpe = (position * self.div_term).unsqueeze(0)
55
- pe = torch.cat([torch.sin(scpe), torch.cos(scpe)]).permute(1, 2, 0)
56
- pe = pe.contiguous().view(1, -1, self.embedding_dim)
57
- return pe
58
-
59
- def forward(self, x: torch.Tensor) -> torch.Tensor:
60
- pe = self.extend_pe(x)
61
- output = x.unsqueeze(-1) if x.ndim == 2 else x
62
- output = output * self.x_scale + self.alpha * pe
63
- return self.dropout(output)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/lr_schedulers.py DELETED
@@ -1,83 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/modules/lr_schedulers.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import math
4
-
5
- import torch
6
- from matplotlib import pyplot as plt
7
- from torch import nn
8
- from torch.optim import Adam
9
-
10
-
11
- class WarmupCosineLRSchedule(torch.optim.lr_scheduler._LRScheduler):
12
- """
13
- Implements Warmup learning rate schedule until 'warmup_steps', going from 'init_lr' to 'peak_lr' for multiple optimizers.
14
- """
15
-
16
- def __init__(
17
- self,
18
- optimizer,
19
- init_lr,
20
- peak_lr,
21
- end_lr,
22
- warmup_steps=10000,
23
- total_steps=400000,
24
- current_step=0,
25
- ):
26
- self.init_lr = init_lr
27
- self.peak_lr = peak_lr
28
- self.end_lr = end_lr
29
- self.optimizer = optimizer
30
- self._warmup_rate = (peak_lr - init_lr) / warmup_steps
31
- self._decay_rate = (end_lr - peak_lr) / (total_steps - warmup_steps)
32
- self._current_step = current_step
33
- self.lr = init_lr
34
- self.warmup_steps = warmup_steps
35
- self.total_steps = total_steps
36
- self._last_lr = [self.lr]
37
-
38
- def set_lr(self, lr):
39
- self._last_lr = [g["lr"] for g in self.optimizer.param_groups]
40
- for g in self.optimizer.param_groups:
41
- # g['lr'] = lr
42
- g["lr"] = self.end_lr ###锁定用线性
43
-
44
- def step(self):
45
- if self._current_step < self.warmup_steps:
46
- lr = self.init_lr + self._warmup_rate * self._current_step
47
-
48
- elif self._current_step > self.total_steps:
49
- lr = self.end_lr
50
-
51
- else:
52
- decay_ratio = (self._current_step - self.warmup_steps) / (
53
- self.total_steps - self.warmup_steps
54
- )
55
- if decay_ratio < 0.0 or decay_ratio > 1.0:
56
- raise RuntimeError(
57
- "Decay ratio must be in [0.0, 1.0]. Fix LR scheduler settings."
58
- )
59
- coeff = 0.5 * (1.0 + math.cos(math.pi * decay_ratio))
60
- lr = self.end_lr + coeff * (self.peak_lr - self.end_lr)
61
-
62
- self.lr = lr = self.end_lr = 0.002 ###锁定用线性###不听话,直接锁定!
63
- self.set_lr(lr)
64
- self.lr = lr
65
- self._current_step += 1
66
- return self.lr
67
-
68
-
69
- if __name__ == "__main__":
70
- m = nn.Linear(10, 10)
71
- opt = Adam(m.parameters(), lr=1e-4)
72
- s = WarmupCosineLRSchedule(
73
- opt, 1e-6, 2e-4, 1e-6, warmup_steps=2000, total_steps=20000, current_step=0
74
- )
75
- lrs = []
76
- for i in range(25000):
77
- s.step()
78
- lrs.append(s.lr)
79
- print(s.lr)
80
-
81
- plt.plot(lrs)
82
- plt.plot(range(0, 25000), lrs)
83
- plt.show()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/optim.py DELETED
@@ -1,622 +0,0 @@
1
- # Copyright 2022 Xiaomi Corp. (authors: Daniel Povey)
2
- #
3
- # See ../LICENSE for clarification regarding multiple authors
4
- #
5
- # Licensed under the Apache License, Version 2.0 (the "License");
6
- # you may not use this file except in compliance with the License.
7
- # You may obtain a copy of the License at
8
- #
9
- # http://www.apache.org/licenses/LICENSE-2.0
10
- #
11
- # Unless required by applicable law or agreed to in writing, software
12
- # distributed under the License is distributed on an "AS IS" BASIS,
13
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
- # See the License for the specific language governing permissions and
15
- # limitations under the License.
16
- import contextlib
17
- import logging
18
- from collections import defaultdict
19
- from typing import List
20
- from typing import Tuple
21
-
22
- import torch
23
- from torch import Tensor
24
- from torch.optim import Optimizer
25
-
26
-
27
- class BatchedOptimizer(Optimizer):
28
- """
29
- This class adds to class Optimizer the capability to optimize parameters in batches:
30
- it will stack the parameters and their grads for you so the optimizer can work
31
- on tensors with an extra leading dimension. This is intended for speed with GPUs,
32
- as it reduces the number of kernels launched in the optimizer.
33
-
34
- Args:
35
- params:
36
- """
37
-
38
- def __init__(self, params, defaults):
39
- super(BatchedOptimizer, self).__init__(params, defaults)
40
-
41
- @contextlib.contextmanager
42
- def batched_params(self, param_group, group_params_names):
43
- """
44
- This function returns (technically, yields) a list of
45
- of tuples (p, state), where
46
- p is a `fake` parameter that is stacked (over axis 0) from real parameters
47
- that share the same shape, and its gradient is also stacked;
48
- `state` is the state corresponding to this batch of parameters
49
- (it will be physically located in the "state" for one of the real
50
- parameters, the last one that has any particular shape and dtype).
51
-
52
- This function is decorated as a context manager so that it can
53
- write parameters back to their "real" locations.
54
-
55
- The idea is, instead of doing:
56
- <code>
57
- for p in group["params"]:
58
- state = self.state[p]
59
- ...
60
- </code>
61
- you can do:
62
- <code>
63
- with self.batched_params(group["params"]) as batches:
64
- for p, state, p_names in batches:
65
- ...
66
- </code>
67
-
68
- Args:
69
- group: a parameter group, which is a list of parameters; should be
70
- one of self.param_groups.
71
- group_params_names: name for each parameter in group,
72
- which is List[str].
73
- """
74
- batches = defaultdict(
75
- list
76
- ) # `batches` maps from tuple (dtype_as_str,*shape) to list of nn.Parameter
77
- batches_names = defaultdict(
78
- list
79
- ) # `batches` maps from tuple (dtype_as_str,*shape) to list of str
80
-
81
- assert len(param_group) == len(group_params_names)
82
- for p, named_p in zip(param_group, group_params_names):
83
- key = (str(p.dtype), *p.shape)
84
- batches[key].append(p)
85
- batches_names[key].append(named_p)
86
-
87
- batches_names_keys = list(batches_names.keys())
88
- sorted_idx = sorted(
89
- range(len(batches_names)), key=lambda i: batches_names_keys[i])
90
- batches_names = [
91
- batches_names[batches_names_keys[idx]] for idx in sorted_idx
92
- ]
93
- batches = [batches[batches_names_keys[idx]] for idx in sorted_idx]
94
-
95
- stacked_params_dict = dict()
96
-
97
- # turn batches into a list, in deterministic order.
98
- # tuples will contain tuples of (stacked_param, state, stacked_params_names),
99
- # one for each batch in `batches`.
100
- tuples = []
101
-
102
- for batch, batch_names in zip(batches, batches_names):
103
- p = batch[0]
104
- # we arbitrarily store the state in the
105
- # state corresponding to the 1st parameter in the
106
- # group. class Optimizer will take care of saving/loading state.
107
- state = self.state[p]
108
- p_stacked = torch.stack(batch)
109
- grad = torch.stack([
110
- torch.zeros_like(p) if p.grad is None else p.grad for p in batch
111
- ])
112
- p_stacked.grad = grad
113
- stacked_params_dict[key] = p_stacked
114
- tuples.append((p_stacked, state, batch_names))
115
-
116
- yield tuples # <-- calling code will do the actual optimization here!
117
-
118
- for ((stacked_params, _state, _names), batch) in zip(tuples, batches):
119
- for i, p in enumerate(batch): # batch is list of Parameter
120
- p.copy_(stacked_params[i])
121
-
122
-
123
- class ScaledAdam(BatchedOptimizer):
124
- """
125
- Implements 'Scaled Adam', a variant of Adam where we scale each parameter's update
126
- proportional to the norm of that parameter; and also learn the scale of the parameter,
127
- in log space, subject to upper and lower limits (as if we had factored each parameter as
128
- param = underlying_param * log_scale.exp())
129
-
130
-
131
- Args:
132
- params: The parameters or param_groups to optimize (like other Optimizer subclasses)
133
- lr: The learning rate. We will typically use a learning rate schedule that starts
134
- at 0.03 and decreases over time, i.e. much higher than other common
135
- optimizers.
136
- clipping_scale: (e.g. 2.0)
137
- A scale for gradient-clipping: if specified, the normalized gradients
138
- over the whole model will be clipped to have 2-norm equal to
139
- `clipping_scale` times the median 2-norm over the most recent period
140
- of `clipping_update_period` minibatches. By "normalized gradients",
141
- we mean after multiplying by the rms parameter value for this tensor
142
- [for non-scalars]; this is appropriate because our update is scaled
143
- by this quantity.
144
- betas: beta1,beta2 are momentum constants for regular momentum, and moving sum-sq grad.
145
- Must satisfy 0 < beta <= beta2 < 1.
146
- scalar_lr_scale: A scaling factor on the learning rate, that we use to update the
147
- scale of each parameter tensor and scalar parameters of the mode..
148
- If each parameter were decomposed
149
- as p * p_scale.exp(), where (p**2).mean().sqrt() == 1.0, scalar_lr_scale
150
- would be a the scaling factor on the learning rate of p_scale.
151
- eps: A general-purpose epsilon to prevent division by zero
152
- param_min_rms: Minimum root-mean-square value of parameter tensor, for purposes of
153
- learning the scale on the parameters (we'll constrain the rms of each non-scalar
154
- parameter tensor to be >= this value)
155
- param_max_rms: Maximum root-mean-square value of parameter tensor, for purposes of
156
- learning the scale on the parameters (we'll constrain the rms of each non-scalar
157
- parameter tensor to be <= this value)
158
- scalar_max: Maximum absolute value for scalar parameters (applicable if your
159
- model has any parameters with numel() == 1).
160
- size_update_period: The periodicity, in steps, with which we update the size (scale)
161
- of the parameter tensor. This is provided to save a little time
162
- in the update.
163
- clipping_update_period: if clipping_scale is specified, this is the period
164
- """
165
-
166
- def __init__(
167
- self,
168
- params,
169
- lr=3e-02,
170
- clipping_scale=None,
171
- betas=(0.9, 0.98),
172
- scalar_lr_scale=0.1,
173
- eps=1.0e-08,
174
- param_min_rms=1.0e-05,
175
- param_max_rms=3.0,
176
- scalar_max=10.0,
177
- size_update_period=4,
178
- clipping_update_period=100,
179
- parameters_names=None,
180
- show_dominant_parameters=True, ):
181
-
182
- assert parameters_names is not None, (
183
- "Please prepare parameters_names,"
184
- "which is a List[List[str]]. Each List[str] is for a group"
185
- "and each str is for a parameter")
186
- defaults = dict(
187
- lr=lr,
188
- clipping_scale=clipping_scale,
189
- betas=betas,
190
- scalar_lr_scale=scalar_lr_scale,
191
- eps=eps,
192
- param_min_rms=param_min_rms,
193
- param_max_rms=param_max_rms,
194
- scalar_max=scalar_max,
195
- size_update_period=size_update_period,
196
- clipping_update_period=clipping_update_period, )
197
-
198
- super(ScaledAdam, self).__init__(params, defaults)
199
- assert len(self.param_groups) == len(parameters_names)
200
- self.parameters_names = parameters_names
201
- self.show_dominant_parameters = show_dominant_parameters
202
-
203
- def __setstate__(self, state):
204
- super(ScaledAdam, self).__setstate__(state)
205
-
206
- @torch.no_grad()
207
- def step(self, closure=None):
208
- """Performs a single optimization step.
209
-
210
- Arguments:
211
- closure (callable, optional): A closure that reevaluates the model
212
- and returns the loss.
213
- """
214
- loss = None
215
- if closure is not None:
216
- with torch.enable_grad():
217
- loss = closure()
218
-
219
- batch = True
220
-
221
- for group, group_params_names in zip(self.param_groups,
222
- self.parameters_names):
223
-
224
- with self.batched_params(group["params"],
225
- group_params_names) as batches:
226
-
227
- # batches is list of pairs (stacked_param, state). stacked_param is like
228
- # a regular parameter, and will have a .grad, but the 1st dim corresponds to
229
- # a stacking dim, it is not a real dim.
230
-
231
- if (len(batches[0][1]) ==
232
- 0): # if len(first state) == 0: not yet initialized
233
- clipping_scale = 1
234
- else:
235
- clipping_scale = self._get_clipping_scale(group, batches)
236
-
237
- for p, state, _ in batches:
238
- # Perform optimization step.
239
- # grad is not going to be None, we handled that when creating the batches.
240
- grad = p.grad
241
- if grad.is_sparse:
242
- raise RuntimeError(
243
- "ScaledAdam optimizer does not support sparse gradients"
244
- )
245
- # State initialization
246
- if len(state) == 0:
247
- self._init_state(group, p, state)
248
-
249
- self._step_one_batch(group, p, state, clipping_scale)
250
-
251
- return loss
252
-
253
- def _init_state(self, group: dict, p: Tensor, state: dict):
254
- """
255
- Initializes state dict for parameter 'p'. Assumes that dim 0 of tensor p
256
- is actually the batch dimension, corresponding to batched-together
257
- parameters of a given shape.
258
-
259
-
260
- Args:
261
- group: Dict to look up configuration values.
262
- p: The parameter that we are initializing the state for
263
- state: Dict from string to whatever state we are initializing
264
- """
265
- size_update_period = group["size_update_period"]
266
-
267
- state["step"] = 0
268
-
269
- kwargs = {"device": p.device, "dtype": p.dtype}
270
-
271
- # 'delta' implements conventional momentum. There are
272
- # several different kinds of update going on, so rather than
273
- # compute "exp_avg" like in Adam, we store and decay a
274
- # parameter-change "delta", which combines all forms of
275
- # update. this is equivalent to how it's done in Adam,
276
- # except for the first few steps.
277
- state["delta"] = torch.zeros_like(
278
- p, memory_format=torch.preserve_format)
279
-
280
- batch_size = p.shape[0]
281
- numel = p.numel() // batch_size
282
- numel = p.numel()
283
-
284
- if numel > 1:
285
- # "param_rms" just periodically records the scalar root-mean-square value of
286
- # the parameter tensor.
287
- # it has a shape like (batch_size, 1, 1, 1, 1)
288
- param_rms = (
289
- (p**2).mean(dim=list(range(1, p.ndim)), keepdim=True).sqrt())
290
- state["param_rms"] = param_rms
291
-
292
- state["scale_exp_avg_sq"] = torch.zeros_like(param_rms)
293
- state["scale_grads"] = torch.zeros(size_update_period,
294
- *param_rms.shape, **kwargs)
295
-
296
- # exp_avg_sq is the weighted sum of scaled gradients. as in Adam.
297
- state["exp_avg_sq"] = torch.zeros_like(
298
- p, memory_format=torch.preserve_format)
299
-
300
- def _get_clipping_scale(self,
301
- group: dict,
302
- tuples: List[Tuple[Tensor, dict, List[str]]]
303
- ) -> float:
304
- """
305
- Returns a scalar factor <= 1.0 that dictates gradient clipping, i.e. we will scale the gradients
306
- by this amount before applying the rest of the update.
307
-
308
- Args:
309
- group: the parameter group, an item in self.param_groups
310
- tuples: a list of tuples of (param, state, param_names)
311
- where param is a batched set of parameters,
312
- with a .grad (1st dim is batch dim)
313
- and state is the state-dict where optimization parameters are kept.
314
- param_names is a List[str] while each str is name for a parameter
315
- in batched set of parameters "param".
316
- """
317
- assert len(tuples) >= 1
318
- clipping_scale = group["clipping_scale"]
319
- (first_p, first_state, _) = tuples[0]
320
- step = first_state["step"]
321
- if clipping_scale is None or step == 0:
322
- # no clipping. return early on step == 0 because the other
323
- # parameters' state won't have been initialized yet.
324
- return 1.0
325
- clipping_update_period = group["clipping_update_period"]
326
-
327
- tot_sumsq = torch.tensor(0.0, device=first_p.device)
328
- for (p, state, param_names) in tuples:
329
- grad = p.grad
330
- if grad.is_sparse:
331
- raise RuntimeError(
332
- "ScaledAdam optimizer does not support sparse gradients")
333
- if p.numel() == p.shape[0]: # a batch of scalars
334
- tot_sumsq += (grad**2).sum() # sum() to change shape [1] to []
335
- else:
336
- tot_sumsq += ((grad * state["param_rms"])**2).sum()
337
-
338
- tot_norm = tot_sumsq.sqrt()
339
- if "model_norms" not in first_state:
340
- first_state["model_norms"] = torch.zeros(
341
- clipping_update_period, device=p.device)
342
- first_state["model_norms"][step % clipping_update_period] = tot_norm
343
-
344
- if step % clipping_update_period == 0:
345
- # Print some stats.
346
- # We don't reach here if step == 0 because we would have returned
347
- # above.
348
- sorted_norms = first_state["model_norms"].sort()[0].to("cpu")
349
- quartiles = []
350
- for n in range(0, 5):
351
- index = min(
352
- clipping_update_period - 1,
353
- (clipping_update_period // 4) * n, )
354
- quartiles.append(sorted_norms[index].item())
355
-
356
- median = quartiles[2]
357
- threshold = clipping_scale * median
358
- first_state["model_norm_threshold"] = threshold
359
- percent_clipped = (first_state["num_clipped"] * 100.0 /
360
- clipping_update_period
361
- if "num_clipped" in first_state else 0.0)
362
- first_state["num_clipped"] = 0
363
- quartiles = " ".join(["%.3e" % x for x in quartiles])
364
- logging.info(
365
- f"Clipping_scale={clipping_scale}, grad-norm quartiles {quartiles}, "
366
- f"threshold={threshold:.3e}, percent-clipped={percent_clipped:.1f}"
367
- )
368
-
369
- if step < clipping_update_period:
370
- return 1.0 # We have not yet estimated a norm to clip to.
371
- else:
372
- try:
373
- model_norm_threshold = first_state["model_norm_threshold"]
374
- except KeyError:
375
- logging.info(
376
- "Warning: model_norm_threshold not in state: possibly "
377
- "you changed config when restarting, adding clipping_scale option?"
378
- )
379
- return 1.0
380
- ans = min(1.0, (model_norm_threshold / (tot_norm + 1.0e-20)).item())
381
- if ans < 1.0:
382
- first_state["num_clipped"] += 1
383
- if ans < 0.1:
384
- logging.warn(
385
- f"Scaling gradients by {ans}, model_norm_threshold={model_norm_threshold}"
386
- )
387
- if self.show_dominant_parameters:
388
- assert p.shape[0] == len(param_names)
389
- self._show_gradient_dominating_parameter(tuples, tot_sumsq)
390
- return ans
391
-
392
- def _show_gradient_dominating_parameter(
393
- self, tuples: List[Tuple[Tensor, dict, List[str]]],
394
- tot_sumsq: Tensor):
395
- """
396
- Show information of parameter wihch dominanting tot_sumsq.
397
-
398
- Args:
399
- tuples: a list of tuples of (param, state, param_names)
400
- where param is a batched set of parameters,
401
- with a .grad (1st dim is batch dim)
402
- and state is the state-dict where optimization parameters are kept.
403
- param_names is a List[str] while each str is name for a parameter
404
- in batched set of parameters "param".
405
- tot_sumsq: sumsq of all parameters. Though it's could be calculated
406
- from tuples, we still pass it to save some time.
407
- """
408
- all_sumsq_orig = {}
409
- for (p, state, batch_param_names) in tuples:
410
- # p is a stacked batch parameters.
411
- batch_grad = p.grad
412
- if p.numel() == p.shape[0]: # a batch of scalars
413
- batch_sumsq_orig = batch_grad**2
414
- # Dummpy values used by following `zip` statement.
415
- batch_rms_orig = torch.ones(p.shape[0])
416
- else:
417
- batch_rms_orig = state["param_rms"]
418
- batch_sumsq_orig = ((batch_grad * batch_rms_orig)**2).sum(
419
- dim=list(range(1, batch_grad.ndim)))
420
-
421
- for name, sumsq_orig, rms, grad in zip(batch_param_names,
422
- batch_sumsq_orig,
423
- batch_rms_orig, batch_grad):
424
-
425
- proportion_orig = sumsq_orig / tot_sumsq
426
- all_sumsq_orig[name] = (proportion_orig, sumsq_orig, rms, grad)
427
-
428
- assert torch.isclose(
429
- sum([value[0] for value in all_sumsq_orig.values()]).cpu(),
430
- torch.tensor(1.0), )
431
- sorted_by_proportion = {
432
- k: v
433
- for k, v in sorted(
434
- all_sumsq_orig.items(),
435
- key=lambda item: item[1][0],
436
- reverse=True, )
437
- }
438
- dominant_param_name = next(iter(sorted_by_proportion))
439
- (dominant_proportion, dominant_sumsq, dominant_rms,
440
- dominant_grad, ) = sorted_by_proportion[dominant_param_name]
441
- logging.info(f"Parameter Dominanting tot_sumsq {dominant_param_name}"
442
- f" with proportion {dominant_proportion:.2f},"
443
- f" where dominant_sumsq=(grad_sumsq*orig_rms_sq)"
444
- f"={dominant_sumsq:.3e},"
445
- f" grad_sumsq = {(dominant_grad**2).sum():.3e},"
446
- f" orig_rms_sq={(dominant_rms**2).item():.3e}")
447
-
448
- def _step_one_batch(self,
449
- group: dict,
450
- p: Tensor,
451
- state: dict,
452
- clipping_scale: float):
453
- """
454
- Do the step for one parameter, which is actually going to be a batch of
455
- `real` parameters, with dim 0 as the batch dim.
456
- Args:
457
- group: dict to look up configuration values
458
- p: parameter to update (actually multiple parameters stacked together
459
- as a batch)
460
- state: state-dict for p, to look up the optimizer state
461
- """
462
- lr = group["lr"]
463
- size_update_period = group["size_update_period"]
464
- beta1 = group["betas"][0]
465
-
466
- grad = p.grad
467
- if clipping_scale != 1.0:
468
- grad = grad * clipping_scale
469
- step = state["step"]
470
- delta = state["delta"]
471
-
472
- delta.mul_(beta1)
473
- batch_size = p.shape[0]
474
- numel = p.numel() // batch_size
475
- if numel > 1:
476
- # Update the size/scale of p, and set param_rms
477
- scale_grads = state["scale_grads"]
478
- scale_grads[step % size_update_period] = (p * grad).sum(
479
- dim=list(range(1, p.ndim)), keepdim=True)
480
- if step % size_update_period == size_update_period - 1:
481
- param_rms = state["param_rms"] # shape: (batch_size, 1, 1, ..)
482
- param_rms.copy_((p**2)
483
- .mean(dim=list(range(1, p.ndim)), keepdim=True)
484
- .sqrt())
485
- if step > 0:
486
- # self._size_update() learns the overall scale on the
487
- # parameter, by shrinking or expanding it.
488
- self._size_update(group, scale_grads, p, state)
489
-
490
- if numel == 1:
491
- # For parameters with 1 element we just use regular Adam.
492
- # Updates delta.
493
- self._step_scalar(group, p, state)
494
- else:
495
- self._step(group, p, state)
496
-
497
- state["step"] = step + 1
498
-
499
- def _size_update(self,
500
- group: dict,
501
- scale_grads: Tensor,
502
- p: Tensor,
503
- state: dict) -> None:
504
- """
505
- Called only where p.numel() > 1, this updates the scale of the parameter.
506
- If we imagine: p = underlying_param * scale.exp(), and we are doing
507
- gradient descent on underlying param and on scale, this function does the update
508
- on `scale`.
509
-
510
- Args:
511
- group: dict to look up configuration values
512
- scale_grads: a tensor of shape (size_update_period, batch_size, 1, 1,...) containing
513
- grads w.r.t. the scales.
514
- p: The parameter to update
515
- state: The state-dict of p
516
- """
517
-
518
- param_rms = state["param_rms"]
519
- beta1, beta2 = group["betas"]
520
- size_lr = group["lr"] * group["scalar_lr_scale"]
521
- param_min_rms = group["param_min_rms"]
522
- param_max_rms = group["param_max_rms"]
523
- eps = group["eps"]
524
- step = state["step"]
525
- batch_size = p.shape[0]
526
-
527
- size_update_period = scale_grads.shape[0]
528
- # correct beta2 for the size update period: we will have
529
- # faster decay at this level.
530
- beta2_corr = beta2**size_update_period
531
-
532
- scale_exp_avg_sq = state[
533
- "scale_exp_avg_sq"] # shape: (batch_size, 1, 1, ..)
534
- scale_exp_avg_sq.mul_(beta2_corr).add_(
535
- (scale_grads**2).mean(dim=0), # mean over dim `size_update_period`
536
- alpha=1 - beta2_corr, ) # shape is (batch_size, 1, 1, ...)
537
-
538
- # The 1st time we reach here is when size_step == 1.
539
- size_step = (step + 1) // size_update_period
540
- bias_correction2 = 1 - beta2_corr**size_step
541
- # we don't bother with bias_correction1; this will help prevent divergence
542
- # at the start of training.
543
-
544
- denom = scale_exp_avg_sq.sqrt() + eps
545
-
546
- scale_step = (-size_lr * (bias_correction2**0.5) *
547
- scale_grads.sum(dim=0) / denom)
548
-
549
- is_too_small = param_rms < param_min_rms
550
- is_too_large = param_rms > param_max_rms
551
-
552
- # when the param gets too small, just don't shrink it any further.
553
- scale_step.masked_fill_(is_too_small, 0.0)
554
- # when it gets too large, stop it from getting any larger.
555
- scale_step.masked_fill_(is_too_large, -size_lr * size_update_period)
556
- delta = state["delta"]
557
- # the factor of (1-beta1) relates to momentum.
558
- delta.add_(p * scale_step, alpha=(1 - beta1))
559
-
560
- def _step(self, group: dict, p: Tensor, state: dict):
561
- """
562
- This function does the core update of self.step(), in the case where the members of
563
- the batch have more than 1 element.
564
-
565
- Args:
566
- group: A dict which will be used to look up configuration values
567
- p: The parameter to be updated
568
- grad: The grad of p
569
- state: The state-dict corresponding to parameter p
570
-
571
- This function modifies p.
572
- """
573
- grad = p.grad
574
- lr = group["lr"]
575
- beta1, beta2 = group["betas"]
576
- eps = group["eps"]
577
- param_min_rms = group["param_min_rms"]
578
- step = state["step"]
579
-
580
- exp_avg_sq = state["exp_avg_sq"]
581
- exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=(1 - beta2))
582
-
583
- this_step = state["step"] - (state["zero_step"]
584
- if "zero_step" in state else 0)
585
- bias_correction2 = 1 - beta2**(this_step + 1)
586
- if bias_correction2 < 0.99:
587
- # note: not in-place.
588
- exp_avg_sq = exp_avg_sq * (1.0 / bias_correction2)
589
-
590
- denom = exp_avg_sq.sqrt()
591
- denom += eps
592
- grad = grad / denom
593
-
594
- alpha = -lr * (1 - beta1) * state["param_rms"].clamp(min=param_min_rms)
595
-
596
- delta = state["delta"]
597
- delta.add_(grad * alpha)
598
- p.add_(delta)
599
-
600
- def _step_scalar(self, group: dict, p: Tensor, state: dict):
601
- """
602
- A simplified form of the core update for scalar tensors, where we cannot get a good
603
- estimate of the parameter rms.
604
- """
605
- beta1, beta2 = group["betas"]
606
- scalar_max = group["scalar_max"]
607
- eps = group["eps"]
608
- lr = group["lr"] * group["scalar_lr_scale"]
609
- grad = p.grad
610
-
611
- exp_avg_sq = state["exp_avg_sq"] # shape: (batch_size,)
612
- exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
613
-
614
- # bias_correction2 is like in Adam. Don't bother with bias_correction1;
615
- # slower update at the start will help stability anyway.
616
- bias_correction2 = 1 - beta2**(state["step"] + 1)
617
- denom = (exp_avg_sq / bias_correction2).sqrt() + eps
618
-
619
- delta = state["delta"]
620
- delta.add_(grad / denom, alpha=-lr * (1 - beta1))
621
- p.clamp_(min=-scalar_max, max=scalar_max)
622
- p.add_(delta)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/patched_mha_with_cache.py DELETED
@@ -1,465 +0,0 @@
1
- from torch.nn.functional import *
2
- from torch.nn.functional import (
3
- _mha_shape_check,
4
- _canonical_mask,
5
- _none_or_dtype,
6
- _in_projection_packed,
7
- )
8
- from torch.nn import functional as F
9
- import torch
10
- # Tensor = torch.Tensor
11
- # from typing import Callable, List, Optional, Tuple, Union
12
-
13
-
14
- def multi_head_attention_forward_patched(
15
- query: Tensor,
16
- key: Tensor,
17
- value: Tensor,
18
- embed_dim_to_check: int,
19
- num_heads: int,
20
- in_proj_weight: Optional[Tensor],
21
- in_proj_bias: Optional[Tensor],
22
- bias_k: Optional[Tensor],
23
- bias_v: Optional[Tensor],
24
- add_zero_attn: bool,
25
- dropout_p: float,
26
- out_proj_weight: Tensor,
27
- out_proj_bias: Optional[Tensor],
28
- training: bool = True,
29
- key_padding_mask: Optional[Tensor] = None,
30
- need_weights: bool = True,
31
- attn_mask: Optional[Tensor] = None,
32
- use_separate_proj_weight: bool = False,
33
- q_proj_weight: Optional[Tensor] = None,
34
- k_proj_weight: Optional[Tensor] = None,
35
- v_proj_weight: Optional[Tensor] = None,
36
- static_k: Optional[Tensor] = None,
37
- static_v: Optional[Tensor] = None,
38
- average_attn_weights: bool = True,
39
- is_causal: bool = False,
40
- cache=None,
41
- ) -> Tuple[Tensor, Optional[Tensor]]:
42
- r"""
43
- Args:
44
- query, key, value: map a query and a set of key-value pairs to an output.
45
- See "Attention Is All You Need" for more details.
46
- embed_dim_to_check: total dimension of the model.
47
- num_heads: parallel attention heads.
48
- in_proj_weight, in_proj_bias: input projection weight and bias.
49
- bias_k, bias_v: bias of the key and value sequences to be added at dim=0.
50
- add_zero_attn: add a new batch of zeros to the key and
51
- value sequences at dim=1.
52
- dropout_p: probability of an element to be zeroed.
53
- out_proj_weight, out_proj_bias: the output projection weight and bias.
54
- training: apply dropout if is ``True``.
55
- key_padding_mask: if provided, specified padding elements in the key will
56
- be ignored by the attention. This is an binary mask. When the value is True,
57
- the corresponding value on the attention layer will be filled with -inf.
58
- need_weights: output attn_output_weights.
59
- Default: `True`
60
- Note: `needs_weight` defaults to `True`, but should be set to `False`
61
- For best performance when attention weights are not nedeeded.
62
- *Setting needs_weights to `True`
63
- leads to a significant performance degradation.*
64
- attn_mask: 2D or 3D mask that prevents attention to certain positions. A 2D mask will be broadcasted for all
65
- the batches while a 3D mask allows to specify a different mask for the entries of each batch.
66
- is_causal: If specified, applies a causal mask as attention mask, and ignores
67
- attn_mask for computing scaled dot product attention.
68
- Default: ``False``.
69
- .. warning::
70
- is_causal is provides a hint that the attn_mask is the
71
- causal mask.Providing incorrect hints can result in
72
- incorrect execution, including forward and backward
73
- compatibility.
74
- use_separate_proj_weight: the function accept the proj. weights for query, key,
75
- and value in different forms. If false, in_proj_weight will be used, which is
76
- a combination of q_proj_weight, k_proj_weight, v_proj_weight.
77
- q_proj_weight, k_proj_weight, v_proj_weight, in_proj_bias: input projection weight and bias.
78
- static_k, static_v: static key and value used for attention operators.
79
- average_attn_weights: If true, indicates that the returned ``attn_weights`` should be averaged across heads.
80
- Otherwise, ``attn_weights`` are provided separately per head. Note that this flag only has an effect
81
- when ``need_weights=True.``. Default: True
82
-
83
-
84
- Shape:
85
- Inputs:
86
- - query: :math:`(L, E)` or :math:`(L, N, E)` where L is the target sequence length, N is the batch size, E is
87
- the embedding dimension.
88
- - key: :math:`(S, E)` or :math:`(S, N, E)`, where S is the source sequence length, N is the batch size, E is
89
- the embedding dimension.
90
- - value: :math:`(S, E)` or :math:`(S, N, E)` where S is the source sequence length, N is the batch size, E is
91
- the embedding dimension.
92
- - key_padding_mask: :math:`(S)` or :math:`(N, S)` where N is the batch size, S is the source sequence length.
93
- If a FloatTensor is provided, it will be directly added to the value.
94
- If a BoolTensor is provided, the positions with the
95
- value of ``True`` will be ignored while the position with the value of ``False`` will be unchanged.
96
- - attn_mask: 2D mask :math:`(L, S)` where L is the target sequence length, S is the source sequence length.
97
- 3D mask :math:`(N*num_heads, L, S)` where N is the batch size, L is the target sequence length,
98
- S is the source sequence length. attn_mask ensures that position i is allowed to attend the unmasked
99
- positions. If a BoolTensor is provided, positions with ``True``
100
- are not allowed to attend while ``False`` values will be unchanged. If a FloatTensor
101
- is provided, it will be added to the attention weight.
102
- - static_k: :math:`(N*num_heads, S, E/num_heads)`, where S is the source sequence length,
103
- N is the batch size, E is the embedding dimension. E/num_heads is the head dimension.
104
- - static_v: :math:`(N*num_heads, S, E/num_heads)`, where S is the source sequence length,
105
- N is the batch size, E is the embedding dimension. E/num_heads is the head dimension.
106
-
107
- Outputs:
108
- - attn_output: :math:`(L, E)` or :math:`(L, N, E)` where L is the target sequence length, N is the batch size,
109
- E is the embedding dimension.
110
- - attn_output_weights: Only returned when ``need_weights=True``. If ``average_attn_weights=True``, returns
111
- attention weights averaged across heads of shape :math:`(L, S)` when input is unbatched or
112
- :math:`(N, L, S)`, where :math:`N` is the batch size, :math:`L` is the target sequence length, and
113
- :math:`S` is the source sequence length. If ``average_attn_weights=False``, returns attention weights per
114
- head of shape :math:`(num_heads, L, S)` when input is unbatched or :math:`(N, num_heads, L, S)`.
115
- """
116
- tens_ops = (
117
- query,
118
- key,
119
- value,
120
- in_proj_weight,
121
- in_proj_bias,
122
- bias_k,
123
- bias_v,
124
- out_proj_weight,
125
- out_proj_bias,
126
- )
127
- if has_torch_function(tens_ops):
128
- return handle_torch_function(
129
- multi_head_attention_forward,
130
- tens_ops,
131
- query,
132
- key,
133
- value,
134
- embed_dim_to_check,
135
- num_heads,
136
- in_proj_weight,
137
- in_proj_bias,
138
- bias_k,
139
- bias_v,
140
- add_zero_attn,
141
- dropout_p,
142
- out_proj_weight,
143
- out_proj_bias,
144
- training=training,
145
- key_padding_mask=key_padding_mask,
146
- need_weights=need_weights,
147
- attn_mask=attn_mask,
148
- is_causal=is_causal,
149
- use_separate_proj_weight=use_separate_proj_weight,
150
- q_proj_weight=q_proj_weight,
151
- k_proj_weight=k_proj_weight,
152
- v_proj_weight=v_proj_weight,
153
- static_k=static_k,
154
- static_v=static_v,
155
- average_attn_weights=average_attn_weights,
156
- cache=cache,
157
- )
158
-
159
- is_batched = _mha_shape_check(
160
- query, key, value, key_padding_mask, attn_mask, num_heads
161
- )
162
-
163
- # For unbatched input, we unsqueeze at the expected batch-dim to pretend that the input
164
- # is batched, run the computation and before returning squeeze the
165
- # batch dimension so that the output doesn't carry this temporary batch dimension.
166
- if not is_batched:
167
- # unsqueeze if the input is unbatched
168
- query = query.unsqueeze(1)
169
- key = key.unsqueeze(1)
170
- value = value.unsqueeze(1)
171
- if key_padding_mask is not None:
172
- key_padding_mask = key_padding_mask.unsqueeze(0)
173
-
174
- # set up shape vars
175
- tgt_len, bsz, embed_dim = query.shape
176
- src_len, _, _ = key.shape
177
-
178
- key_padding_mask = _canonical_mask(
179
- mask=key_padding_mask,
180
- mask_name="key_padding_mask",
181
- other_type=_none_or_dtype(attn_mask),
182
- other_name="attn_mask",
183
- target_type=query.dtype,
184
- )
185
-
186
- if is_causal and attn_mask is None:
187
- raise RuntimeError(
188
- "Need attn_mask if specifying the is_causal hint. "
189
- "You may use the Transformer module method "
190
- "`generate_square_subsequent_mask` to create this mask."
191
- )
192
-
193
- if is_causal and key_padding_mask is None and not need_weights:
194
- # when we have a kpm or need weights, we need attn_mask
195
- # Otherwise, we use the is_causal hint go as is_causal
196
- # indicator to SDPA.
197
- attn_mask = None
198
- else:
199
- attn_mask = _canonical_mask(
200
- mask=attn_mask,
201
- mask_name="attn_mask",
202
- other_type=None,
203
- other_name="",
204
- target_type=query.dtype,
205
- check_other=False,
206
- )
207
-
208
- if key_padding_mask is not None:
209
- # We have the attn_mask, and use that to merge kpm into it.
210
- # Turn off use of is_causal hint, as the merged mask is no
211
- # longer causal.
212
- is_causal = False
213
-
214
- assert (
215
- embed_dim == embed_dim_to_check
216
- ), f"was expecting embedding dimension of {embed_dim_to_check}, but got {embed_dim}"
217
- if isinstance(embed_dim, torch.Tensor):
218
- # embed_dim can be a tensor when JIT tracing
219
- head_dim = embed_dim.div(num_heads, rounding_mode="trunc")
220
- else:
221
- head_dim = embed_dim // num_heads
222
- assert (
223
- head_dim * num_heads == embed_dim
224
- ), f"embed_dim {embed_dim} not divisible by num_heads {num_heads}"
225
- if use_separate_proj_weight:
226
- # allow MHA to have different embedding dimensions when separate projection weights are used
227
- assert (
228
- key.shape[:2] == value.shape[:2]
229
- ), f"key's sequence and batch dims {key.shape[:2]} do not match value's {value.shape[:2]}"
230
- else:
231
- assert (
232
- key.shape == value.shape
233
- ), f"key shape {key.shape} does not match value shape {value.shape}"
234
-
235
- #
236
- # compute in-projection
237
- #
238
- if not use_separate_proj_weight:
239
- assert (
240
- in_proj_weight is not None
241
- ), "use_separate_proj_weight is False but in_proj_weight is None"
242
- q, k, v = _in_projection_packed(query, key, value, in_proj_weight, in_proj_bias)
243
- else:
244
- assert (
245
- q_proj_weight is not None
246
- ), "use_separate_proj_weight is True but q_proj_weight is None"
247
- assert (
248
- k_proj_weight is not None
249
- ), "use_separate_proj_weight is True but k_proj_weight is None"
250
- assert (
251
- v_proj_weight is not None
252
- ), "use_separate_proj_weight is True but v_proj_weight is None"
253
- if in_proj_bias is None:
254
- b_q = b_k = b_v = None
255
- else:
256
- b_q, b_k, b_v = in_proj_bias.chunk(3)
257
- q, k, v = _in_projection(
258
- query,
259
- key,
260
- value,
261
- q_proj_weight,
262
- k_proj_weight,
263
- v_proj_weight,
264
- b_q,
265
- b_k,
266
- b_v,
267
- )
268
- if cache != None:
269
- if cache["first_infer"] == 1:
270
- cache["k"][cache["stage"]] = k
271
- # print(0,cache["k"].shape)
272
- cache["v"][cache["stage"]] = v
273
- else: ###12个layer每个都要留自己的cache_kv
274
- # print(1,cache["k"].shape)
275
- cache["k"][cache["stage"]] = torch.cat(
276
- [cache["k"][cache["stage"]], k], 0
277
- ) ##本来时序是1,但是proj的时候可能transpose了所以时序到0维了
278
- cache["v"][cache["stage"]] = torch.cat([cache["v"][cache["stage"]], v], 0)
279
- # print(2, cache["k"].shape)
280
- src_len = cache["k"][cache["stage"]].shape[0]
281
- k = cache["k"][cache["stage"]]
282
- v = cache["v"][cache["stage"]]
283
- # if attn_mask is not None:
284
- # attn_mask=attn_mask[-1:,]
285
- # print(attn_mask.shape,attn_mask)
286
- cache["stage"] = (cache["stage"] + 1) % cache["all_stage"]
287
- # print(2333,cache)
288
- # prep attention mask
289
-
290
- attn_mask = _canonical_mask(
291
- mask=attn_mask,
292
- mask_name="attn_mask",
293
- other_type=None,
294
- other_name="",
295
- target_type=q.dtype,
296
- check_other=False,
297
- )
298
-
299
- if attn_mask is not None:
300
- # ensure attn_mask's dim is 3
301
- if attn_mask.dim() == 2:
302
- correct_2d_size = (tgt_len, src_len)
303
- if attn_mask.shape != correct_2d_size:
304
- raise RuntimeError(
305
- f"The shape of the 2D attn_mask is {attn_mask.shape}, but should be {correct_2d_size}."
306
- )
307
- attn_mask = attn_mask.unsqueeze(0)
308
- elif attn_mask.dim() == 3:
309
- correct_3d_size = (bsz * num_heads, tgt_len, src_len)
310
- if attn_mask.shape != correct_3d_size:
311
- raise RuntimeError(
312
- f"The shape of the 3D attn_mask is {attn_mask.shape}, but should be {correct_3d_size}."
313
- )
314
- else:
315
- raise RuntimeError(
316
- f"attn_mask's dimension {attn_mask.dim()} is not supported"
317
- )
318
-
319
- # add bias along batch dimension (currently second)
320
- if bias_k is not None and bias_v is not None:
321
- assert static_k is None, "bias cannot be added to static key."
322
- assert static_v is None, "bias cannot be added to static value."
323
- k = torch.cat([k, bias_k.repeat(1, bsz, 1)])
324
- v = torch.cat([v, bias_v.repeat(1, bsz, 1)])
325
- if attn_mask is not None:
326
- attn_mask = pad(attn_mask, (0, 1))
327
- if key_padding_mask is not None:
328
- key_padding_mask = pad(key_padding_mask, (0, 1))
329
- else:
330
- assert bias_k is None
331
- assert bias_v is None
332
-
333
- #
334
- # reshape q, k, v for multihead attention and make em batch first
335
- #
336
- q = q.view(tgt_len, bsz * num_heads, head_dim).transpose(0, 1)
337
- if static_k is None:
338
- k = k.view(k.shape[0], bsz * num_heads, head_dim).transpose(0, 1)
339
- else:
340
- # TODO finish disentangling control flow so we don't do in-projections when statics are passed
341
- assert (
342
- static_k.size(0) == bsz * num_heads
343
- ), f"expecting static_k.size(0) of {bsz * num_heads}, but got {static_k.size(0)}"
344
- assert (
345
- static_k.size(2) == head_dim
346
- ), f"expecting static_k.size(2) of {head_dim}, but got {static_k.size(2)}"
347
- k = static_k
348
- if static_v is None:
349
- v = v.view(v.shape[0], bsz * num_heads, head_dim).transpose(0, 1)
350
- else:
351
- # TODO finish disentangling control flow so we don't do in-projections when statics are passed
352
- assert (
353
- static_v.size(0) == bsz * num_heads
354
- ), f"expecting static_v.size(0) of {bsz * num_heads}, but got {static_v.size(0)}"
355
- assert (
356
- static_v.size(2) == head_dim
357
- ), f"expecting static_v.size(2) of {head_dim}, but got {static_v.size(2)}"
358
- v = static_v
359
-
360
- # add zero attention along batch dimension (now first)
361
- if add_zero_attn:
362
- zero_attn_shape = (bsz * num_heads, 1, head_dim)
363
- k = torch.cat(
364
- [k, torch.zeros(zero_attn_shape, dtype=k.dtype, device=k.device)], dim=1
365
- )
366
- v = torch.cat(
367
- [v, torch.zeros(zero_attn_shape, dtype=v.dtype, device=v.device)], dim=1
368
- )
369
- if attn_mask is not None:
370
- attn_mask = pad(attn_mask, (0, 1))
371
- if key_padding_mask is not None:
372
- key_padding_mask = pad(key_padding_mask, (0, 1))
373
-
374
- # update source sequence length after adjustments
375
- src_len = k.size(1)
376
-
377
- # merge key padding and attention masks
378
- if key_padding_mask is not None:
379
- assert key_padding_mask.shape == (
380
- bsz,
381
- src_len,
382
- ), f"expecting key_padding_mask shape of {(bsz, src_len)}, but got {key_padding_mask.shape}"
383
- key_padding_mask = (
384
- key_padding_mask.view(bsz, 1, 1, src_len)
385
- .expand(-1, num_heads, -1, -1)
386
- .reshape(bsz * num_heads, 1, src_len)
387
- )
388
- if attn_mask is None:
389
- attn_mask = key_padding_mask
390
- else:
391
- attn_mask = attn_mask + key_padding_mask
392
-
393
- # adjust dropout probability
394
- if not training:
395
- dropout_p = 0.0
396
-
397
- #
398
- # (deep breath) calculate attention and out projection
399
- #
400
-
401
- if need_weights:
402
- B, Nt, E = q.shape
403
- q_scaled = q / math.sqrt(E)
404
-
405
- assert not (
406
- is_causal and attn_mask is None
407
- ), "FIXME: is_causal not implemented for need_weights"
408
-
409
- if attn_mask is not None:
410
- attn_output_weights = torch.baddbmm(
411
- attn_mask, q_scaled, k.transpose(-2, -1)
412
- )
413
- else:
414
- attn_output_weights = torch.bmm(q_scaled, k.transpose(-2, -1))
415
- attn_output_weights = softmax(attn_output_weights, dim=-1)
416
- if dropout_p > 0.0:
417
- attn_output_weights = dropout(attn_output_weights, p=dropout_p)
418
-
419
- attn_output = torch.bmm(attn_output_weights, v)
420
-
421
- attn_output = (
422
- attn_output.transpose(0, 1).contiguous().view(tgt_len * bsz, embed_dim)
423
- )
424
- attn_output = linear(attn_output, out_proj_weight, out_proj_bias)
425
- attn_output = attn_output.view(tgt_len, bsz, attn_output.size(1))
426
-
427
- # optionally average attention weights over heads
428
- attn_output_weights = attn_output_weights.view(bsz, num_heads, tgt_len, src_len)
429
- if average_attn_weights:
430
- attn_output_weights = attn_output_weights.mean(dim=1)
431
-
432
- if not is_batched:
433
- # squeeze the output if input was unbatched
434
- attn_output = attn_output.squeeze(1)
435
- attn_output_weights = attn_output_weights.squeeze(0)
436
- return attn_output, attn_output_weights
437
- else:
438
- # attn_mask can be either (L,S) or (N*num_heads, L, S)
439
- # if attn_mask's shape is (1, L, S) we need to unsqueeze to (1, 1, L, S)
440
- # in order to match the input for SDPA of (N, num_heads, L, S)
441
- if attn_mask is not None:
442
- if attn_mask.size(0) == 1 and attn_mask.dim() == 3:
443
- attn_mask = attn_mask.unsqueeze(0)
444
- else:
445
- attn_mask = attn_mask.view(bsz, num_heads, -1, src_len)
446
-
447
- q = q.view(bsz, num_heads, tgt_len, head_dim)
448
- k = k.view(bsz, num_heads, src_len, head_dim)
449
- v = v.view(bsz, num_heads, src_len, head_dim)
450
-
451
- # with torch.backends.cuda.sdp_kernel(enable_flash=True, enable_math=True, enable_mem_efficient=True):
452
- attn_output = scaled_dot_product_attention(
453
- q, k, v, attn_mask, dropout_p, is_causal
454
- )
455
-
456
- attn_output = (
457
- attn_output.permute(2, 0, 1, 3).contiguous().view(bsz * tgt_len, embed_dim)
458
- )
459
-
460
- attn_output = linear(attn_output, out_proj_weight, out_proj_bias)
461
- attn_output = attn_output.view(tgt_len, bsz, attn_output.size(1))
462
- if not is_batched:
463
- # squeeze the output if input was unbatched
464
- attn_output = attn_output.squeeze(1)
465
- return attn_output, None
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/patched_mha_with_cache_onnx.py DELETED
@@ -1,92 +0,0 @@
1
- from torch.nn.functional import *
2
- from torch.nn.functional import (
3
- _mha_shape_check,
4
- _canonical_mask,
5
- _none_or_dtype,
6
- _in_projection_packed,
7
- )
8
-
9
- def multi_head_attention_forward_patched(
10
- query,
11
- key,
12
- value,
13
- embed_dim_to_check: int,
14
- num_heads: int,
15
- in_proj_weight,
16
- in_proj_bias: Optional[Tensor],
17
- bias_k: Optional[Tensor],
18
- bias_v: Optional[Tensor],
19
- add_zero_attn: bool,
20
- dropout_p: float,
21
- out_proj_weight: Tensor,
22
- out_proj_bias: Optional[Tensor],
23
- training: bool = True,
24
- key_padding_mask: Optional[Tensor] = None,
25
- need_weights: bool = True,
26
- attn_mask: Optional[Tensor] = None,
27
- use_separate_proj_weight: bool = False,
28
- q_proj_weight: Optional[Tensor] = None,
29
- k_proj_weight: Optional[Tensor] = None,
30
- v_proj_weight: Optional[Tensor] = None,
31
- static_k: Optional[Tensor] = None,
32
- static_v: Optional[Tensor] = None,
33
- average_attn_weights: bool = True,
34
- is_causal: bool = False,
35
- cache=None,
36
- ) -> Tuple[Tensor, Optional[Tensor]]:
37
-
38
- # set up shape vars
39
- _, _, embed_dim = query.shape
40
- attn_mask = _canonical_mask(
41
- mask=attn_mask,
42
- mask_name="attn_mask",
43
- other_type=None,
44
- other_name="",
45
- target_type=query.dtype,
46
- check_other=False,
47
- )
48
- head_dim = embed_dim // num_heads
49
-
50
- proj_qkv = linear(query, in_proj_weight, in_proj_bias)
51
- proj_qkv = proj_qkv.unflatten(-1, (3, query.size(-1))).unsqueeze(0).transpose(0, -2).squeeze(-2).contiguous()
52
- q, k, v = proj_qkv[0], proj_qkv[1], proj_qkv[2]
53
-
54
- if cache["first_infer"] == 1:
55
- cache["k"][cache["stage"]] = k
56
- cache["v"][cache["stage"]] = v
57
- else:
58
- cache["k"][cache["stage"]] = torch.cat([cache["k"][cache["stage"]][:-1], k], 0)
59
- cache["v"][cache["stage"]] = torch.cat([cache["v"][cache["stage"]][:-1], v], 0)
60
- k = cache["k"][cache["stage"]]
61
- v = cache["v"][cache["stage"]]
62
- cache["stage"] = (cache["stage"] + 1) % cache["all_stage"]
63
-
64
- attn_mask = _canonical_mask(
65
- mask=attn_mask,
66
- mask_name="attn_mask",
67
- other_type=None,
68
- other_name="",
69
- target_type=q.dtype,
70
- check_other=False,
71
- )
72
- attn_mask = attn_mask.unsqueeze(0)
73
-
74
- q = q.view(-1, num_heads, head_dim).transpose(0, 1)
75
- k = k.view(-1, num_heads, head_dim).transpose(0, 1)
76
- v = v.view(-1, num_heads, head_dim).transpose(0, 1)
77
-
78
- dropout_p = 0.0
79
- attn_mask = attn_mask.unsqueeze(0)
80
- q = q.view(num_heads, -1, head_dim).unsqueeze(0)
81
- k = k.view(num_heads, -1, head_dim).unsqueeze(0)
82
- v = v.view(num_heads, -1, head_dim).unsqueeze(0)
83
- attn_output = scaled_dot_product_attention(
84
- q, k, v, attn_mask, dropout_p, is_causal
85
- )
86
- attn_output = (
87
- attn_output.permute(2, 0, 1, 3).contiguous().view(-1, embed_dim)
88
- )
89
- attn_output = linear(attn_output, out_proj_weight, out_proj_bias)
90
- attn_output = attn_output.view(-1, 1, attn_output.size(1))
91
-
92
- return attn_output
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/scaling.py DELETED
@@ -1,335 +0,0 @@
1
- # Copyright 2022 Xiaomi Corp. (authors: Daniel Povey)
2
- #
3
- # See ../../../../LICENSE for clarification regarding multiple authors
4
- #
5
- # Licensed under the Apache License, Version 2.0 (the "License");
6
- # you may not use this file except in compliance with the License.
7
- # You may obtain a copy of the License at
8
- #
9
- # http://www.apache.org/licenses/LICENSE-2.0
10
- #
11
- # Unless required by applicable law or agreed to in writing, software
12
- # distributed under the License is distributed on an "AS IS" BASIS,
13
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
- # See the License for the specific language governing permissions and
15
- # limitations under the License.
16
- import logging
17
- import math
18
- import random
19
- from typing import Optional
20
- from typing import Tuple
21
- from typing import Union
22
-
23
- import torch
24
- import torch.nn as nn
25
- from torch import Tensor
26
-
27
-
28
- class DoubleSwishFunction(torch.autograd.Function):
29
- """
30
- double_swish(x) = x * torch.sigmoid(x-1)
31
- This is a definition, originally motivated by its close numerical
32
- similarity to swish(swish(x)), where swish(x) = x * sigmoid(x).
33
-
34
- Memory-efficient derivative computation:
35
- double_swish(x) = x * s, where s(x) = torch.sigmoid(x-1)
36
- double_swish'(x) = d/dx double_swish(x) = x * s'(x) + x' * s(x) = x * s'(x) + s(x).
37
- Now, s'(x) = s(x) * (1-s(x)).
38
- double_swish'(x) = x * s'(x) + s(x).
39
- = x * s(x) * (1-s(x)) + s(x).
40
- = double_swish(x) * (1-s(x)) + s(x)
41
- ... so we just need to remember s(x) but not x itself.
42
- """
43
-
44
- @staticmethod
45
- def forward(ctx, x: Tensor) -> Tensor:
46
- requires_grad = x.requires_grad
47
- x_dtype = x.dtype
48
- if x.dtype == torch.float16:
49
- x = x.to(torch.float32)
50
-
51
- s = torch.sigmoid(x - 1.0)
52
- y = x * s
53
-
54
- if requires_grad:
55
- deriv = y * (1 - s) + s
56
- # notes on derivative of x * sigmoid(x - 1):
57
- # https://www.wolframalpha.com/input?i=d%2Fdx+%28x+*+sigmoid%28x-1%29%29
58
- # min \simeq -0.043638. Take floor as -0.043637 so it's a lower bund
59
- # max \simeq 1.1990. Take ceil to be 1.2 so it's an upper bound.
60
- # the combination of "+ torch.rand_like(deriv)" and casting to torch.uint8 (which
61
- # floors), should be expectation-preserving.
62
- floor = -0.043637
63
- ceil = 1.2
64
- d_scaled = (deriv - floor) * (255.0 / (ceil - floor)) + torch.rand_like(
65
- deriv
66
- )
67
- if __name__ == "__main__":
68
- # for self-testing only.
69
- assert d_scaled.min() >= 0.0
70
- assert d_scaled.max() < 256.0
71
- d_int = d_scaled.to(torch.uint8)
72
- ctx.save_for_backward(d_int)
73
- if x.dtype == torch.float16 or torch.is_autocast_enabled():
74
- y = y.to(torch.float16)
75
- return y
76
-
77
- @staticmethod
78
- def backward(ctx, y_grad: Tensor) -> Tensor:
79
- (d,) = ctx.saved_tensors
80
- # the same constants as used in forward pass.
81
- floor = -0.043637
82
- ceil = 1.2
83
- d = d * ((ceil - floor) / 255.0) + floor
84
- return y_grad * d
85
-
86
-
87
- class DoubleSwish(torch.nn.Module):
88
- def forward(self, x: Tensor) -> Tensor:
89
- """Return double-swish activation function which is an approximation to Swish(Swish(x)),
90
- that we approximate closely with x * sigmoid(x-1).
91
- """
92
- if torch.jit.is_scripting() or torch.jit.is_tracing():
93
- return x * torch.sigmoid(x - 1.0)
94
- return DoubleSwishFunction.apply(x)
95
-
96
-
97
- class ActivationBalancerFunction(torch.autograd.Function):
98
- @staticmethod
99
- def forward(
100
- ctx,
101
- x: Tensor,
102
- scale_factor: Tensor,
103
- sign_factor: Optional[Tensor],
104
- channel_dim: int,
105
- ) -> Tensor:
106
- if channel_dim < 0:
107
- channel_dim += x.ndim
108
- ctx.channel_dim = channel_dim
109
- xgt0 = x > 0
110
- if sign_factor is None:
111
- ctx.save_for_backward(xgt0, scale_factor)
112
- else:
113
- ctx.save_for_backward(xgt0, scale_factor, sign_factor)
114
- return x
115
-
116
- @staticmethod
117
- def backward(ctx, x_grad: Tensor) -> Tuple[Tensor, None, None, None]:
118
- if len(ctx.saved_tensors) == 3:
119
- xgt0, scale_factor, sign_factor = ctx.saved_tensors
120
- for _ in range(ctx.channel_dim, x_grad.ndim - 1):
121
- scale_factor = scale_factor.unsqueeze(-1)
122
- sign_factor = sign_factor.unsqueeze(-1)
123
- factor = sign_factor + scale_factor * (xgt0.to(x_grad.dtype) - 0.5)
124
- else:
125
- xgt0, scale_factor = ctx.saved_tensors
126
- for _ in range(ctx.channel_dim, x_grad.ndim - 1):
127
- scale_factor = scale_factor.unsqueeze(-1)
128
- factor = scale_factor * (xgt0.to(x_grad.dtype) - 0.5)
129
- neg_delta_grad = x_grad.abs() * factor
130
- return (
131
- x_grad - neg_delta_grad,
132
- None,
133
- None,
134
- None,
135
- )
136
-
137
-
138
- def _compute_scale_factor(
139
- x: Tensor,
140
- channel_dim: int,
141
- min_abs: float,
142
- max_abs: float,
143
- gain_factor: float,
144
- max_factor: float,
145
- ) -> Tensor:
146
- if channel_dim < 0:
147
- channel_dim += x.ndim
148
- sum_dims = [d for d in range(x.ndim) if d != channel_dim]
149
- x_abs_mean = torch.mean(x.abs(), dim=sum_dims).to(torch.float32)
150
-
151
- if min_abs == 0.0:
152
- below_threshold = 0.0
153
- else:
154
- # below_threshold is 0 if x_abs_mean > min_abs, can be at most max_factor if
155
- # x_abs)_mean , min_abs.
156
- below_threshold = ((min_abs - x_abs_mean) * (gain_factor / min_abs)).clamp(
157
- min=0, max=max_factor
158
- )
159
-
160
- above_threshold = ((x_abs_mean - max_abs) * (gain_factor / max_abs)).clamp(
161
- min=0, max=max_factor
162
- )
163
-
164
- return below_threshold - above_threshold
165
-
166
-
167
- def _compute_sign_factor(
168
- x: Tensor,
169
- channel_dim: int,
170
- min_positive: float,
171
- max_positive: float,
172
- gain_factor: float,
173
- max_factor: float,
174
- ) -> Tensor:
175
- if channel_dim < 0:
176
- channel_dim += x.ndim
177
- sum_dims = [d for d in range(x.ndim) if d != channel_dim]
178
- proportion_positive = torch.mean((x > 0).to(torch.float32), dim=sum_dims)
179
- if min_positive == 0.0:
180
- factor1 = 0.0
181
- else:
182
- # 0 if proportion_positive >= min_positive, else can be
183
- # as large as max_factor.
184
- factor1 = (
185
- (min_positive - proportion_positive) * (gain_factor / min_positive)
186
- ).clamp_(min=0, max=max_factor)
187
-
188
- if max_positive == 1.0:
189
- factor2 = 0.0
190
- else:
191
- # 0 if self.proportion_positive <= max_positive, else can be
192
- # as large as -max_factor.
193
- factor2 = (
194
- (proportion_positive - max_positive) * (gain_factor / (1.0 - max_positive))
195
- ).clamp_(min=0, max=max_factor)
196
- sign_factor = factor1 - factor2
197
- # require min_positive != 0 or max_positive != 1:
198
- assert not isinstance(sign_factor, float)
199
- return sign_factor
200
-
201
-
202
- class ActivationBalancer(torch.nn.Module):
203
- """
204
- Modifies the backpropped derivatives of a function to try to encourage, for
205
- each channel, that it is positive at least a proportion `threshold` of the
206
- time. It does this by multiplying negative derivative values by up to
207
- (1+max_factor), and positive derivative values by up to (1-max_factor),
208
- interpolated from 1 at the threshold to those extremal values when none
209
- of the inputs are positive.
210
-
211
- Args:
212
- num_channels: the number of channels
213
- channel_dim: the dimension/axis corresponding to the channel, e.g.
214
- -1, 0, 1, 2; will be interpreted as an offset from x.ndim if negative.
215
- min_positive: the minimum, per channel, of the proportion of the time
216
- that (x > 0), below which we start to modify the derivatives.
217
- max_positive: the maximum, per channel, of the proportion of the time
218
- that (x > 0), above which we start to modify the derivatives.
219
- max_factor: the maximum factor by which we modify the derivatives for
220
- either the sign constraint or the magnitude constraint;
221
- e.g. with max_factor=0.02, the the derivatives would be multiplied by
222
- values in the range [0.98..1.02].
223
- sign_gain_factor: determines the 'gain' with which we increase the
224
- change in gradient once the constraints on min_positive and max_positive
225
- are violated.
226
- scale_gain_factor: determines the 'gain' with which we increase the
227
- change in gradient once the constraints on min_abs and max_abs
228
- are violated.
229
- min_abs: the minimum average-absolute-value difference from the mean
230
- value per channel, which we allow, before we start to modify
231
- the derivatives to prevent this.
232
- max_abs: the maximum average-absolute-value difference from the mean
233
- value per channel, which we allow, before we start to modify
234
- the derivatives to prevent this.
235
- min_prob: determines the minimum probability with which we modify the
236
- gradients for the {min,max}_positive and {min,max}_abs constraints,
237
- on each forward(). This is done randomly to prevent all layers
238
- from doing it at the same time. Early in training we may use
239
- higher probabilities than this; it will decay to this value.
240
- """
241
-
242
- def __init__(
243
- self,
244
- num_channels: int,
245
- channel_dim: int,
246
- min_positive: float = 0.05,
247
- max_positive: float = 0.95,
248
- max_factor: float = 0.04,
249
- sign_gain_factor: float = 0.01,
250
- scale_gain_factor: float = 0.02,
251
- min_abs: float = 0.2,
252
- max_abs: float = 100.0,
253
- min_prob: float = 0.1,
254
- ):
255
- super(ActivationBalancer, self).__init__()
256
- self.num_channels = num_channels
257
- self.channel_dim = channel_dim
258
- self.min_positive = min_positive
259
- self.max_positive = max_positive
260
- self.max_factor = max_factor
261
- self.min_abs = min_abs
262
- self.max_abs = max_abs
263
- self.min_prob = min_prob
264
- self.sign_gain_factor = sign_gain_factor
265
- self.scale_gain_factor = scale_gain_factor
266
-
267
- # count measures how many times the forward() function has been called.
268
- # We occasionally sync this to a tensor called `count`, that exists to
269
- # make sure it is synced to disk when we load and save the model.
270
- self.cpu_count = 0
271
- self.register_buffer("count", torch.tensor(0, dtype=torch.int64))
272
-
273
- def forward(self, x: Tensor) -> Tensor:
274
- if torch.jit.is_scripting() or not x.requires_grad or torch.jit.is_tracing():
275
- return _no_op(x)
276
-
277
- count = self.cpu_count
278
- self.cpu_count += 1
279
-
280
- if random.random() < 0.01:
281
- # Occasionally sync self.cpu_count with self.count.
282
- # count affects the decay of 'prob'. don't do this on every iter,
283
- # because syncing with the GPU is slow.
284
- self.cpu_count = max(self.cpu_count, self.count.item())
285
- self.count.fill_(self.cpu_count)
286
-
287
- # the prob of doing some work exponentially decreases from 0.5 till it hits
288
- # a floor at min_prob (==0.1, by default)
289
- prob = max(self.min_prob, 0.5 ** (1 + (count / 4000.0)))
290
-
291
- if random.random() < prob:
292
- sign_gain_factor = 0.5
293
- if self.min_positive != 0.0 or self.max_positive != 1.0:
294
- sign_factor = _compute_sign_factor(
295
- x,
296
- self.channel_dim,
297
- self.min_positive,
298
- self.max_positive,
299
- gain_factor=self.sign_gain_factor / prob,
300
- max_factor=self.max_factor,
301
- )
302
- else:
303
- sign_factor = None
304
-
305
- scale_factor = _compute_scale_factor(
306
- x.detach(),
307
- self.channel_dim,
308
- min_abs=self.min_abs,
309
- max_abs=self.max_abs,
310
- gain_factor=self.scale_gain_factor / prob,
311
- max_factor=self.max_factor,
312
- )
313
- return ActivationBalancerFunction.apply(
314
- x,
315
- scale_factor,
316
- sign_factor,
317
- self.channel_dim,
318
- )
319
- else:
320
- return _no_op(x)
321
-
322
-
323
- def BalancedDoubleSwish(
324
- d_model, channel_dim=-1, max_abs=10.0, min_prob=0.25
325
- ) -> nn.Sequential:
326
- """
327
- ActivationBalancer -> DoubleSwish
328
- """
329
- balancer = ActivationBalancer(
330
- d_model, channel_dim=channel_dim, max_abs=max_abs, min_prob=min_prob
331
- )
332
- return nn.Sequential(
333
- balancer,
334
- DoubleSwish(),
335
- )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/transformer.py DELETED
@@ -1,378 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/transformer.py
2
- import copy
3
- import numbers
4
- from functools import partial
5
- from typing import Any
6
- from typing import Callable
7
- from typing import List
8
- from typing import Optional
9
- from typing import Tuple
10
- from typing import Union
11
-
12
- import torch
13
- from AR.modules.activation import MultiheadAttention
14
- from AR.modules.scaling import BalancedDoubleSwish
15
- from torch import nn
16
- from torch import Tensor
17
- from torch.nn import functional as F
18
-
19
- _shape_t = Union[int, List[int], torch.Size]
20
-
21
-
22
- class LayerNorm(nn.Module):
23
- __constants__ = ["normalized_shape", "eps", "elementwise_affine"]
24
- normalized_shape: Tuple[int, ...]
25
- eps: float
26
- elementwise_affine: bool
27
-
28
- def __init__(
29
- self,
30
- normalized_shape: _shape_t,
31
- eps: float = 1e-5,
32
- elementwise_affine: bool = True,
33
- device=None,
34
- dtype=None,
35
- ) -> None:
36
- factory_kwargs = {"device": device, "dtype": dtype}
37
- super(LayerNorm, self).__init__()
38
- if isinstance(normalized_shape, numbers.Integral):
39
- # mypy error: incompatible types in assignment
40
- normalized_shape = (normalized_shape,) # type: ignore[assignment]
41
- self.normalized_shape = tuple(normalized_shape) # type: ignore[arg-type]
42
- self.eps = eps
43
- self.elementwise_affine = elementwise_affine
44
- if self.elementwise_affine:
45
- self.weight = nn.Parameter(
46
- torch.empty(self.normalized_shape, **factory_kwargs)
47
- )
48
- self.bias = nn.Parameter(
49
- torch.empty(self.normalized_shape, **factory_kwargs)
50
- )
51
- else:
52
- self.register_parameter("weight", None)
53
- self.register_parameter("bias", None)
54
-
55
- self.reset_parameters()
56
-
57
- def reset_parameters(self) -> None:
58
- if self.elementwise_affine:
59
- nn.init.ones_(self.weight)
60
- nn.init.zeros_(self.bias)
61
-
62
- def forward(self, input: Tensor, embedding: Any = None) -> Tensor:
63
- if isinstance(input, tuple):
64
- input, embedding = input
65
- return (
66
- F.layer_norm(
67
- input,
68
- self.normalized_shape,
69
- self.weight,
70
- self.bias,
71
- self.eps,
72
- ),
73
- embedding,
74
- )
75
-
76
- assert embedding is None
77
- return F.layer_norm(
78
- input, self.normalized_shape, self.weight, self.bias, self.eps
79
- )
80
-
81
- def extra_repr(self) -> str:
82
- return (
83
- "{normalized_shape}, eps={eps}, "
84
- "elementwise_affine={elementwise_affine}".format(**self.__dict__)
85
- )
86
-
87
-
88
- class IdentityNorm(nn.Module):
89
- def __init__(
90
- self,
91
- d_model: int,
92
- eps: float = 1e-5,
93
- device=None,
94
- dtype=None,
95
- ) -> None:
96
- super(IdentityNorm, self).__init__()
97
-
98
- def forward(self, input: Tensor, embedding: Any = None) -> Tensor:
99
- if isinstance(input, tuple):
100
- return input
101
-
102
- assert embedding is None
103
- return input
104
-
105
-
106
- class TransformerEncoder(nn.Module):
107
- r"""TransformerEncoder is a stack of N encoder layers. Users can build the
108
- BERT(https://arxiv.org/abs/1810.04805) model with corresponding parameters.
109
-
110
- Args:
111
- encoder_layer: an instance of the TransformerEncoderLayer() class (required).
112
- num_layers: the number of sub-encoder-layers in the encoder (required).
113
- norm: the layer normalization component (optional).
114
- enable_nested_tensor: if True, input will automatically convert to nested tensor
115
- (and convert back on output). This will improve the overall performance of
116
- TransformerEncoder when padding rate is high. Default: ``True`` (enabled).
117
-
118
- Examples::
119
- >>> encoder_layer = TransformerEncoderLayer(d_model=512, nhead=8)
120
- >>> transformer_encoder = TransformerEncoder(encoder_layer, num_layers=6)
121
- >>> src = torch.rand(10, 32, 512)
122
- >>> out = transformer_encoder(src)
123
- """
124
- __constants__ = ["norm"]
125
-
126
- def __init__(self, encoder_layer, num_layers, norm=None):
127
- super(TransformerEncoder, self).__init__()
128
- self.layers = _get_clones(encoder_layer, num_layers)
129
- self.num_layers = num_layers
130
- self.norm = norm
131
-
132
- def forward(
133
- self,
134
- src: Tensor,
135
- mask: Optional[Tensor] = None,
136
- src_key_padding_mask: Optional[Tensor] = None,
137
- return_layer_states: bool = False,
138
- cache=None,
139
- ) -> Tensor:
140
- r"""Pass the input through the encoder layers in turn.
141
-
142
- Args:
143
- src: the sequence to the encoder (required).
144
- mask: the mask for the src sequence (optional).
145
- src_key_padding_mask: the mask for the src keys per batch (optional).
146
- return_layer_states: return layers' state (optional).
147
-
148
- Shape:
149
- see the docs in Transformer class.
150
- """
151
- if return_layer_states:
152
- layer_states = [] # layers' output
153
- output = src
154
- for mod in self.layers:
155
- output = mod(
156
- output,
157
- src_mask=mask,
158
- src_key_padding_mask=src_key_padding_mask,
159
- cache=cache,
160
- )
161
- layer_states.append(output[0])
162
-
163
- if self.norm is not None:
164
- output = self.norm(output)
165
-
166
- return layer_states, output
167
-
168
- output = src
169
- for mod in self.layers:
170
- output = mod(
171
- output,
172
- src_mask=mask,
173
- src_key_padding_mask=src_key_padding_mask,
174
- cache=cache,
175
- )
176
-
177
- if self.norm is not None:
178
- output = self.norm(output)
179
-
180
- return output
181
-
182
-
183
- class TransformerEncoderLayer(nn.Module):
184
- __constants__ = ["batch_first", "norm_first"]
185
-
186
- def __init__(
187
- self,
188
- d_model: int,
189
- nhead: int,
190
- dim_feedforward: int = 2048,
191
- dropout: float = 0.1,
192
- activation: Union[str, Callable[[Tensor], Tensor]] = F.relu,
193
- batch_first: bool = False,
194
- norm_first: bool = False,
195
- device=None,
196
- dtype=None,
197
- linear1_self_attention_cls: nn.Module = nn.Linear,
198
- linear2_self_attention_cls: nn.Module = nn.Linear,
199
- linear1_feedforward_cls: nn.Module = nn.Linear,
200
- linear2_feedforward_cls: nn.Module = nn.Linear,
201
- layer_norm_cls: nn.Module = LayerNorm,
202
- layer_norm_eps: float = 1e-5,
203
- adaptive_layer_norm=False,
204
- ) -> None:
205
- factory_kwargs = {"device": device, "dtype": dtype}
206
- super(TransformerEncoderLayer, self).__init__()
207
- # print(233333333333,d_model,nhead)
208
- # import os
209
- # os._exit(2333333)
210
- self.self_attn = MultiheadAttention(
211
- d_model, # 512 16
212
- nhead,
213
- dropout=dropout,
214
- batch_first=batch_first,
215
- linear1_cls=linear1_self_attention_cls,
216
- linear2_cls=linear2_self_attention_cls,
217
- **factory_kwargs,
218
- )
219
-
220
- # Implementation of Feedforward model
221
- self.linear1 = linear1_feedforward_cls(
222
- d_model, dim_feedforward, **factory_kwargs
223
- )
224
- self.dropout = nn.Dropout(dropout)
225
- self.linear2 = linear2_feedforward_cls(
226
- dim_feedforward, d_model, **factory_kwargs
227
- )
228
-
229
- self.norm_first = norm_first
230
- self.dropout1 = nn.Dropout(dropout)
231
- self.dropout2 = nn.Dropout(dropout)
232
-
233
- # Legacy string support for activation function.
234
- if isinstance(activation, str):
235
- activation = _get_activation_fn(activation)
236
- elif isinstance(activation, partial):
237
- activation = activation(d_model)
238
- elif activation == BalancedDoubleSwish:
239
- activation = BalancedDoubleSwish(d_model)
240
-
241
- # # We can't test self.activation in forward() in TorchScript,
242
- # # so stash some information about it instead.
243
- # if activation is F.relu or isinstance(activation, torch.nn.ReLU):
244
- # self.activation_relu_or_gelu = 1
245
- # elif activation is F.gelu or isinstance(activation, torch.nn.GELU):
246
- # self.activation_relu_or_gelu = 2
247
- # else:
248
- # self.activation_relu_or_gelu = 0
249
- self.activation = activation
250
-
251
- norm1 = layer_norm_cls(d_model, eps=layer_norm_eps, **factory_kwargs)
252
- if layer_norm_cls == IdentityNorm:
253
- norm2 = BalancedBasicNorm(d_model, eps=layer_norm_eps, **factory_kwargs)
254
- else:
255
- norm2 = layer_norm_cls(d_model, eps=layer_norm_eps, **factory_kwargs)
256
-
257
- if adaptive_layer_norm:
258
- self.norm1 = AdaptiveLayerNorm(d_model, norm1)
259
- self.norm2 = AdaptiveLayerNorm(d_model, norm2)
260
- else:
261
- self.norm1 = norm1
262
- self.norm2 = norm2
263
-
264
- def __setstate__(self, state):
265
- super(TransformerEncoderLayer, self).__setstate__(state)
266
- if not hasattr(self, "activation"):
267
- self.activation = F.relu
268
-
269
- def forward(
270
- self,
271
- src: Tensor,
272
- src_mask: Optional[Tensor] = None,
273
- src_key_padding_mask: Optional[Tensor] = None,
274
- cache=None,
275
- ) -> Tensor:
276
- r"""Pass the input through the encoder layer.
277
-
278
- Args:
279
- src: the sequence to the encoder layer (required).
280
- src_mask: the mask for the src sequence (optional).
281
- src_key_padding_mask: the mask for the src keys per batch (optional).
282
-
283
- Shape:
284
- see the docs in Transformer class.
285
- """
286
- x, stage_embedding = src, None
287
- is_src_tuple = False
288
- if isinstance(src, tuple):
289
- x, stage_embedding = src
290
- is_src_tuple = True
291
-
292
- if src_key_padding_mask is not None:
293
- _skpm_dtype = src_key_padding_mask.dtype
294
- if _skpm_dtype != torch.bool and not torch.is_floating_point(
295
- src_key_padding_mask
296
- ):
297
- raise AssertionError(
298
- "only bool and floating types of key_padding_mask are supported"
299
- )
300
-
301
- if self.norm_first:
302
- x = x + self._sa_block(
303
- self.norm1(x, stage_embedding),
304
- src_mask,
305
- src_key_padding_mask,
306
- cache=cache,
307
- )
308
- x = x + self._ff_block(self.norm2(x, stage_embedding))
309
- else:
310
- x = self.norm1(
311
- x + self._sa_block(x, src_mask, src_key_padding_mask, cache=cache),
312
- stage_embedding,
313
- )
314
- x = self.norm2(x + self._ff_block(x), stage_embedding)
315
-
316
- if is_src_tuple:
317
- return (x, stage_embedding)
318
- return x
319
-
320
- # self-attention block
321
- def _sa_block(
322
- self,
323
- x: Tensor,
324
- attn_mask: Optional[Tensor],
325
- key_padding_mask: Optional[Tensor],
326
- cache=None,
327
- ) -> Tensor:
328
- # print(x.shape,attn_mask.shape,key_padding_mask)
329
- # torch.Size([1, 188, 512]) torch.Size([188, 188]) None
330
- # import os
331
- # os._exit(23333)
332
- x = self.self_attn(
333
- x,
334
- x,
335
- x,
336
- attn_mask=attn_mask,
337
- key_padding_mask=key_padding_mask,
338
- need_weights=False,
339
- cache=cache,
340
- )[0]
341
- return self.dropout1(x)
342
-
343
- # feed forward block
344
- def _ff_block(self, x: Tensor) -> Tensor:
345
- x = self.linear2(self.dropout(self.activation(self.linear1(x))))
346
- return self.dropout2(x)
347
-
348
-
349
- class AdaptiveLayerNorm(nn.Module):
350
- r"""Adaptive Layer Normalization"""
351
-
352
- def __init__(self, d_model, norm) -> None:
353
- super(AdaptiveLayerNorm, self).__init__()
354
- self.project_layer = nn.Linear(d_model, 2 * d_model)
355
- self.norm = norm
356
- self.d_model = d_model
357
- self.eps = self.norm.eps
358
-
359
- def forward(self, input: Tensor, embedding: Tensor = None) -> Tensor:
360
- if isinstance(input, tuple):
361
- input, embedding = input
362
- weight, bias = torch.split(
363
- self.project_layer(embedding),
364
- split_size_or_sections=self.d_model,
365
- dim=-1,
366
- )
367
- return (weight * self.norm(input) + bias, embedding)
368
-
369
- weight, bias = torch.split(
370
- self.project_layer(embedding),
371
- split_size_or_sections=self.d_model,
372
- dim=-1,
373
- )
374
- return weight * self.norm(input) + bias
375
-
376
-
377
- def _get_clones(module, N):
378
- return nn.ModuleList([copy.deepcopy(module) for i in range(N)])
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/modules/transformer_onnx.py DELETED
@@ -1,292 +0,0 @@
1
- # modified from https://github.com/lifeiteng/vall-e/blob/main/valle/modules/transformer.py
2
- import copy
3
- import numbers
4
- from functools import partial
5
- from typing import Any
6
- from typing import Callable
7
- from typing import List
8
- from typing import Optional
9
- from typing import Tuple
10
- from typing import Union
11
-
12
- import torch
13
- from AR.modules.activation_onnx import MultiheadAttention
14
- from AR.modules.scaling import BalancedDoubleSwish
15
- from torch import nn
16
- from torch import Tensor
17
- from torch.nn import functional as F
18
-
19
- _shape_t = Union[int, List[int], torch.Size]
20
-
21
-
22
- class LayerNorm(nn.Module):
23
- __constants__ = ["normalized_shape", "eps", "elementwise_affine"]
24
- normalized_shape: Tuple[int, ...]
25
- eps: float
26
- elementwise_affine: bool
27
-
28
- def __init__(
29
- self,
30
- normalized_shape: _shape_t,
31
- eps: float = 1e-5,
32
- elementwise_affine: bool = True,
33
- device=None,
34
- dtype=None,
35
- ) -> None:
36
- factory_kwargs = {"device": device, "dtype": dtype}
37
- super(LayerNorm, self).__init__()
38
- if isinstance(normalized_shape, numbers.Integral):
39
- # mypy error: incompatible types in assignment
40
- normalized_shape = (normalized_shape,) # type: ignore[assignment]
41
- self.normalized_shape = tuple(normalized_shape) # type: ignore[arg-type]
42
- self.eps = eps
43
- self.elementwise_affine = elementwise_affine
44
- if self.elementwise_affine:
45
- self.weight = nn.Parameter(
46
- torch.empty(self.normalized_shape, **factory_kwargs)
47
- )
48
- self.bias = nn.Parameter(
49
- torch.empty(self.normalized_shape, **factory_kwargs)
50
- )
51
- else:
52
- self.register_parameter("weight", None)
53
- self.register_parameter("bias", None)
54
-
55
- self.reset_parameters()
56
-
57
- def reset_parameters(self) -> None:
58
- if self.elementwise_affine:
59
- nn.init.ones_(self.weight)
60
- nn.init.zeros_(self.bias)
61
-
62
- def forward(self, input: Tensor, embedding: Any = None) -> Tensor:
63
- if isinstance(input, tuple):
64
- input, embedding = input
65
- return (
66
- F.layer_norm(
67
- input,
68
- self.normalized_shape,
69
- self.weight,
70
- self.bias,
71
- self.eps,
72
- ),
73
- embedding,
74
- )
75
-
76
- assert embedding is None
77
- return F.layer_norm(
78
- input, self.normalized_shape, self.weight, self.bias, self.eps
79
- )
80
-
81
- def extra_repr(self) -> str:
82
- return (
83
- "{normalized_shape}, eps={eps}, "
84
- "elementwise_affine={elementwise_affine}".format(**self.__dict__)
85
- )
86
-
87
-
88
- class IdentityNorm(nn.Module):
89
- def __init__(
90
- self,
91
- d_model: int,
92
- eps: float = 1e-5,
93
- device=None,
94
- dtype=None,
95
- ) -> None:
96
- super(IdentityNorm, self).__init__()
97
-
98
- def forward(self, input: Tensor, embedding: Any = None) -> Tensor:
99
- if isinstance(input, tuple):
100
- return input
101
-
102
- assert embedding is None
103
- return input
104
-
105
-
106
- class TransformerEncoder(nn.Module):
107
- r"""TransformerEncoder is a stack of N encoder layers. Users can build the
108
- BERT(https://arxiv.org/abs/1810.04805) model with corresponding parameters.
109
-
110
- Args:
111
- encoder_layer: an instance of the TransformerEncoderLayer() class (required).
112
- num_layers: the number of sub-encoder-layers in the encoder (required).
113
- norm: the layer normalization component (optional).
114
- enable_nested_tensor: if True, input will automatically convert to nested tensor
115
- (and convert back on output). This will improve the overall performance of
116
- TransformerEncoder when padding rate is high. Default: ``True`` (enabled).
117
-
118
- Examples::
119
- >>> encoder_layer = TransformerEncoderLayer(d_model=512, nhead=8)
120
- >>> transformer_encoder = TransformerEncoder(encoder_layer, num_layers=6)
121
- >>> src = torch.rand(10, 32, 512)
122
- >>> out = transformer_encoder(src)
123
- """
124
- __constants__ = ["norm"]
125
-
126
- def __init__(self, encoder_layer, num_layers, norm=None):
127
- super(TransformerEncoder, self).__init__()
128
- self.layers = _get_clones(encoder_layer, num_layers)
129
- self.num_layers = num_layers
130
- self.norm = norm
131
-
132
- def forward(
133
- self,
134
- src: Tensor,
135
- mask: Optional[Tensor] = None,
136
- src_key_padding_mask: Optional[Tensor] = None,
137
- return_layer_states: bool = False,
138
- cache=None,
139
- ) -> Tensor:
140
- output = src
141
- for mod in self.layers:
142
- output = mod(
143
- output,
144
- src_mask=mask,
145
- src_key_padding_mask=src_key_padding_mask,
146
- cache=cache,
147
- )
148
-
149
- if self.norm is not None:
150
- output = self.norm(output)
151
-
152
- return output
153
-
154
-
155
- class TransformerEncoderLayer(nn.Module):
156
- __constants__ = ["batch_first", "norm_first"]
157
- def __init__(
158
- self,
159
- d_model: int,
160
- nhead: int,
161
- dim_feedforward: int = 2048,
162
- dropout: float = 0.1,
163
- activation: Union[str, Callable[[Tensor], Tensor]] = F.relu,
164
- batch_first: bool = False,
165
- norm_first: bool = False,
166
- device=None,
167
- dtype=None,
168
- linear1_self_attention_cls: nn.Module = nn.Linear,
169
- linear2_self_attention_cls: nn.Module = nn.Linear,
170
- linear1_feedforward_cls: nn.Module = nn.Linear,
171
- linear2_feedforward_cls: nn.Module = nn.Linear,
172
- layer_norm_cls: nn.Module = LayerNorm,
173
- layer_norm_eps: float = 1e-5,
174
- adaptive_layer_norm=False,
175
- ) -> None:
176
- factory_kwargs = {"device": device, "dtype": dtype}
177
- super(TransformerEncoderLayer, self).__init__()
178
- self.self_attn = MultiheadAttention(
179
- d_model, # 512 16
180
- nhead,
181
- dropout=dropout,
182
- batch_first=batch_first,
183
- linear1_cls=linear1_self_attention_cls,
184
- linear2_cls=linear2_self_attention_cls,
185
- **factory_kwargs,
186
- )
187
- self.linear1 = linear1_feedforward_cls(
188
- d_model, dim_feedforward, **factory_kwargs
189
- )
190
- self.dropout = nn.Dropout(dropout)
191
- self.linear2 = linear2_feedforward_cls(
192
- dim_feedforward, d_model, **factory_kwargs
193
- )
194
- self.norm_first = norm_first
195
- self.dropout1 = nn.Dropout(dropout)
196
- self.dropout2 = nn.Dropout(dropout)
197
- if isinstance(activation, str):
198
- activation = _get_activation_fn(activation)
199
- elif isinstance(activation, partial):
200
- activation = activation(d_model)
201
- elif activation == BalancedDoubleSwish:
202
- activation = BalancedDoubleSwish(d_model)
203
- self.activation = activation
204
-
205
- norm1 = layer_norm_cls(d_model, eps=layer_norm_eps, **factory_kwargs)
206
- if layer_norm_cls == IdentityNorm:
207
- norm2 = BalancedBasicNorm(d_model, eps=layer_norm_eps, **factory_kwargs)
208
- else:
209
- norm2 = layer_norm_cls(d_model, eps=layer_norm_eps, **factory_kwargs)
210
-
211
- if adaptive_layer_norm:
212
- self.norm1 = AdaptiveLayerNorm(d_model, norm1)
213
- self.norm2 = AdaptiveLayerNorm(d_model, norm2)
214
- else:
215
- self.norm1 = norm1
216
- self.norm2 = norm2
217
-
218
- def __setstate__(self, state):
219
- super(TransformerEncoderLayer, self).__setstate__(state)
220
- if not hasattr(self, "activation"):
221
- self.activation = F.relu
222
-
223
- def forward(
224
- self,
225
- src: Tensor,
226
- src_mask: Optional[Tensor] = None,
227
- src_key_padding_mask: Optional[Tensor] = None,
228
- cache=None,
229
- ) -> Tensor:
230
- x = src
231
- stage_embedding = None
232
- x = self.norm1(
233
- x + self._sa_block(x, src_mask, src_key_padding_mask, cache=cache),
234
- stage_embedding,
235
- )
236
- x = self.norm2(x + self._ff_block(x), stage_embedding)
237
-
238
- return x
239
-
240
- def _sa_block(
241
- self,
242
- x: Tensor,
243
- attn_mask: Optional[Tensor],
244
- key_padding_mask: Optional[Tensor],
245
- cache=None,
246
- ) -> Tensor:
247
- x = self.self_attn(
248
- x,
249
- x,
250
- x,
251
- attn_mask=attn_mask,
252
- key_padding_mask=key_padding_mask,
253
- need_weights=False,
254
- cache=cache,
255
- )
256
- return self.dropout1(x)
257
-
258
- def _ff_block(self, x: Tensor) -> Tensor:
259
- x = self.linear2(self.dropout(self.activation(self.linear1(x))))
260
- return self.dropout2(x)
261
-
262
-
263
- class AdaptiveLayerNorm(nn.Module):
264
- r"""Adaptive Layer Normalization"""
265
-
266
- def __init__(self, d_model, norm) -> None:
267
- super(AdaptiveLayerNorm, self).__init__()
268
- self.project_layer = nn.Linear(d_model, 2 * d_model)
269
- self.norm = norm
270
- self.d_model = d_model
271
- self.eps = self.norm.eps
272
-
273
- def forward(self, input: Tensor, embedding: Tensor = None) -> Tensor:
274
- if isinstance(input, tuple):
275
- input, embedding = input
276
- weight, bias = torch.split(
277
- self.project_layer(embedding),
278
- split_size_or_sections=self.d_model,
279
- dim=-1,
280
- )
281
- return (weight * self.norm(input) + bias, embedding)
282
-
283
- weight, bias = torch.split(
284
- self.project_layer(embedding),
285
- split_size_or_sections=self.d_model,
286
- dim=-1,
287
- )
288
- return weight * self.norm(input) + bias
289
-
290
-
291
- def _get_clones(module, N):
292
- return nn.ModuleList([copy.deepcopy(module) for i in range(N)])
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/text_processing/__init__.py DELETED
File without changes
AR/text_processing/phonemizer.py DELETED
@@ -1,79 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/text_processing/phonemizer.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- import itertools
4
- import re
5
- from typing import Dict
6
- from typing import List
7
-
8
- import regex
9
- from gruut import sentences
10
- from gruut.const import Sentence
11
- from gruut.const import Word
12
- from AR.text_processing.symbols import SYMBOL_TO_ID
13
-
14
-
15
- class GruutPhonemizer:
16
- def __init__(self, language: str):
17
- self._phonemizer = sentences
18
- self.lang = language
19
- self.symbol_to_id = SYMBOL_TO_ID
20
- self._special_cases_dict: Dict[str] = {
21
- r"\.\.\.": "... ",
22
- ";": "; ",
23
- ":": ": ",
24
- ",": ", ",
25
- r"\.": ". ",
26
- "!": "! ",
27
- r"\?": "? ",
28
- "—": "—",
29
- "…": "… ",
30
- "«": "«",
31
- "»": "»",
32
- }
33
- self._punctuation_regexp: str = (
34
- rf"([{''.join(self._special_cases_dict.keys())}])"
35
- )
36
-
37
- def _normalize_punctuation(self, text: str) -> str:
38
- text = regex.sub(rf"\pZ+{self._punctuation_regexp}", r"\1", text)
39
- text = regex.sub(rf"{self._punctuation_regexp}(\pL)", r"\1 \2", text)
40
- text = regex.sub(r"\pZ+", r" ", text)
41
- return text.strip()
42
-
43
- def _convert_punctuation(self, word: Word) -> str:
44
- if not word.phonemes:
45
- return ""
46
- if word.phonemes[0] in ["‖", "|"]:
47
- return word.text.strip()
48
-
49
- phonemes = "".join(word.phonemes)
50
- # remove modifier characters ˈˌː with regex
51
- phonemes = re.sub(r"[ˈˌː͡]", "", phonemes)
52
- return phonemes.strip()
53
-
54
- def phonemize(self, text: str, espeak: bool = False) -> str:
55
- text_to_phonemize: str = self._normalize_punctuation(text)
56
- sents: List[Sentence] = [
57
- sent
58
- for sent in self._phonemizer(text_to_phonemize, lang="en-us", espeak=espeak)
59
- ]
60
- words: List[str] = [
61
- self._convert_punctuation(word) for word in itertools.chain(*sents)
62
- ]
63
- return " ".join(words)
64
-
65
- def transform(self, phonemes):
66
- # convert phonemes to ids
67
- # dictionary is in symbols.py
68
- return [self.symbol_to_id[p] for p in phonemes if p in self.symbol_to_id.keys()]
69
-
70
-
71
- if __name__ == "__main__":
72
- phonemizer = GruutPhonemizer("en-us")
73
- # text -> IPA
74
- phonemes = phonemizer.phonemize("Hello, wor-ld ?")
75
- print("phonemes:", phonemes)
76
- print("len(phonemes):", len(phonemes))
77
- phoneme_ids = phonemizer.transform(phonemes)
78
- print("phoneme_ids:", phoneme_ids)
79
- print("len(phoneme_ids):", len(phoneme_ids))
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/text_processing/symbols.py DELETED
@@ -1,10 +0,0 @@
1
- # modified from https://github.com/yangdongchao/SoundStorm/blob/master/soundstorm/s1/AR/text_processing/symbols.py
2
- # reference: https://github.com/lifeiteng/vall-e
3
- PAD = "_"
4
- PUNCTUATION = ';:,.!?¡¿—…"«»“” '
5
- LETTERS = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
6
- IPA_LETTERS = "ɑɐɒæɓʙβɔɕçɗɖðʤəɘɚɛɜɝɞɟʄɡɠɢʛɦɧħɥʜɨɪʝɭɬɫɮʟɱɯɰŋɳɲɴøɵɸθœɶʘɹɺɾɻʀʁɽʂʃʈʧʉʊʋⱱʌɣɤʍχʎʏʑʐʒʔʡʕʢǀǁǂǃˈˌːˑʼʴʰʱʲʷˠˤ˞↓↑→↗↘'̩'ᵻ"
7
- SYMBOLS = [PAD] + list(PUNCTUATION) + list(LETTERS) + list(IPA_LETTERS)
8
- SPACE_ID = SYMBOLS.index(" ")
9
- SYMBOL_TO_ID = {s: i for i, s in enumerate(SYMBOLS)}
10
- ID_TO_SYMBOL = {i: s for i, s in enumerate(SYMBOLS)}
 
 
 
 
 
 
 
 
 
 
 
AR/utils/__init__.py DELETED
@@ -1,37 +0,0 @@
1
- import re
2
-
3
-
4
- def str2bool(str):
5
- return True if str.lower() == 'true' else False
6
-
7
-
8
- def get_newest_ckpt(string_list):
9
- # 定义一个正则表达式模式,用于匹配字符串中的数字
10
- pattern = r'epoch=(\d+)-step=(\d+)\.ckpt'
11
-
12
- # 使用正则表达式提取每个字符串中的数字信息,并创建一个包含元组的列表
13
- extracted_info = []
14
- for string in string_list:
15
- match = re.match(pattern, string)
16
- if match:
17
- epoch = int(match.group(1))
18
- step = int(match.group(2))
19
- extracted_info.append((epoch, step, string))
20
- # 按照 epoch 后面的数字和 step 后面的数字进行排序
21
- sorted_info = sorted(
22
- extracted_info, key=lambda x: (x[0], x[1]), reverse=True)
23
- # 获取最新的 ckpt 文件名
24
- newest_ckpt = sorted_info[0][2]
25
- return newest_ckpt
26
-
27
-
28
- # 文本存在且不为空时 return True
29
- def check_txt_file(file_path):
30
- try:
31
- with open(file_path, 'r') as file:
32
- text = file.readline().strip()
33
- assert text.strip() != ''
34
- return text
35
- except Exception:
36
- return False
37
- return False
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/utils/initialize.py DELETED
@@ -1,38 +0,0 @@
1
- #!/usr/bin/env python3
2
- """Initialize modules for espnet2 neural networks."""
3
- import torch
4
- from typeguard import check_argument_types
5
-
6
-
7
- def initialize(model: torch.nn.Module, init: str):
8
- """Initialize weights of a neural network module.
9
-
10
- Parameters are initialized using the given method or distribution.
11
-
12
- Custom initialization routines can be implemented into submodules
13
- as function `espnet_initialization_fn` within the custom module.
14
-
15
- Args:
16
- model: Target.
17
- init: Method of initialization.
18
- """
19
- assert check_argument_types()
20
- print("init with", init)
21
-
22
- # weight init
23
- for p in model.parameters():
24
- if p.dim() > 1:
25
- if init == "xavier_uniform":
26
- torch.nn.init.xavier_uniform_(p.data)
27
- elif init == "xavier_normal":
28
- torch.nn.init.xavier_normal_(p.data)
29
- elif init == "kaiming_uniform":
30
- torch.nn.init.kaiming_uniform_(p.data, nonlinearity="relu")
31
- elif init == "kaiming_normal":
32
- torch.nn.init.kaiming_normal_(p.data, nonlinearity="relu")
33
- else:
34
- raise ValueError("Unknown initialization: " + init)
35
- # bias init
36
- for name, p in model.named_parameters():
37
- if ".bias" in name and p.dim() == 1:
38
- p.data.zero_()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
AR/utils/io.py DELETED
@@ -1,34 +0,0 @@
1
- import sys
2
-
3
- import torch
4
- import yaml
5
-
6
-
7
- def load_yaml_config(path):
8
- with open(path) as f:
9
- config = yaml.full_load(f)
10
- return config
11
-
12
-
13
- def save_config_to_yaml(config, path):
14
- assert path.endswith(".yaml")
15
- with open(path, "w") as f:
16
- f.write(yaml.dump(config))
17
- f.close()
18
-
19
-
20
- def write_args(args, path):
21
- args_dict = dict(
22
- (name, getattr(args, name)) for name in dir(args) if not name.startswith("_")
23
- )
24
- with open(path, "a") as args_file:
25
- args_file.write("==> torch version: {}\n".format(torch.__version__))
26
- args_file.write(
27
- "==> cudnn version: {}\n".format(torch.backends.cudnn.version())
28
- )
29
- args_file.write("==> Cmd:\n")
30
- args_file.write(str(sys.argv))
31
- args_file.write("\n==> args:\n")
32
- for k, v in sorted(args_dict.items()):
33
- args_file.write(" %s: %s\n" % (str(k), str(v)))
34
- args_file.close()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
README.md DELETED
@@ -1,12 +0,0 @@
1
- ---
2
- title: Gpt Sovits
3
- emoji: 📉
4
- colorFrom: red
5
- colorTo: pink
6
- sdk: gradio
7
- sdk_version: 3.38.0
8
- app_file: app.py
9
- pinned: false
10
- ---
11
-
12
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
__pycache__/utils.cpython-39.pyc DELETED
Binary file (11.3 kB)
 
app.py DELETED
@@ -1,678 +0,0 @@
1
- '''
2
- 按中英混合识别
3
- 按日英混合识别
4
- 多语种启动切分识别语种
5
- 全部按中文识别
6
- 全部按英文识别
7
- 全部按日文识别
8
- '''
9
- import os, re, logging
10
- import LangSegment
11
- logging.getLogger("markdown_it").setLevel(logging.ERROR)
12
- logging.getLogger("urllib3").setLevel(logging.ERROR)
13
- logging.getLogger("httpcore").setLevel(logging.ERROR)
14
- logging.getLogger("httpx").setLevel(logging.ERROR)
15
- logging.getLogger("asyncio").setLevel(logging.ERROR)
16
- logging.getLogger("charset_normalizer").setLevel(logging.ERROR)
17
- logging.getLogger("torchaudio._extension").setLevel(logging.ERROR)
18
- import pdb
19
-
20
- if os.path.exists("./gweight.txt"):
21
- with open("./gweight.txt", 'r', encoding="utf-8") as file:
22
- gweight_data = file.read()
23
- gpt_path = os.environ.get(
24
- "gpt_path", gweight_data)
25
- else:
26
- gpt_path = os.environ.get(
27
- "gpt_path", "GPT_SoVITS/pretrained_models/s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt")
28
-
29
- if os.path.exists("./sweight.txt"):
30
- with open("./sweight.txt", 'r', encoding="utf-8") as file:
31
- sweight_data = file.read()
32
- sovits_path = os.environ.get("sovits_path", sweight_data)
33
- else:
34
- sovits_path = os.environ.get("sovits_path", "GPT_SoVITS/pretrained_models/s2G488k.pth")
35
- # gpt_path = os.environ.get(
36
- # "gpt_path", "pretrained_models/s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt"
37
- # )
38
- # sovits_path = os.environ.get("sovits_path", "pretrained_models/s2G488k.pth")
39
- cnhubert_base_path = os.environ.get(
40
- "cnhubert_base_path", "GPT_SoVITS/pretrained_models/chinese-hubert-base"
41
- )
42
- bert_path = os.environ.get(
43
- "bert_path", "GPT_SoVITS/pretrained_models/chinese-roberta-wwm-ext-large"
44
- )
45
- infer_ttswebui = os.environ.get("infer_ttswebui", 9872)
46
- infer_ttswebui = int(infer_ttswebui)
47
- is_share = os.environ.get("is_share", "False")
48
- is_share = eval(is_share)
49
- if "_CUDA_VISIBLE_DEVICES" in os.environ:
50
- os.environ["CUDA_VISIBLE_DEVICES"] = os.environ["_CUDA_VISIBLE_DEVICES"]
51
- is_half = eval(os.environ.get("is_half", "True"))
52
- import gradio as gr
53
- from transformers import AutoModelForMaskedLM, AutoTokenizer
54
- import numpy as np
55
- import librosa, torch
56
- from feature_extractor import cnhubert
57
-
58
- cnhubert.cnhubert_base_path = cnhubert_base_path
59
-
60
- from module.models import SynthesizerTrn
61
- from AR.models.t2s_lightning_module import Text2SemanticLightningModule
62
- from text import cleaned_text_to_sequence
63
- from text.cleaner import clean_text
64
- from time import time as ttime
65
- from module.mel_processing import spectrogram_torch
66
- from my_utils import load_audio
67
- from tools.i18n.i18n import I18nAuto
68
-
69
- i18n = I18nAuto()
70
-
71
- os.environ['PYTORCH_ENABLE_MPS_FALLBACK'] = '1' # 确保直接启动推理UI时也能够设置。
72
-
73
- if torch.cuda.is_available():
74
- device = "cuda"
75
- elif torch.backends.mps.is_available():
76
- device = "mps"
77
- else:
78
- device = "cpu"
79
-
80
- tokenizer = AutoTokenizer.from_pretrained(bert_path)
81
- bert_model = AutoModelForMaskedLM.from_pretrained(bert_path)
82
- if is_half == True:
83
- bert_model = bert_model.half().to(device)
84
- else:
85
- bert_model = bert_model.to(device)
86
-
87
-
88
- def get_bert_feature(text, word2ph):
89
- with torch.no_grad():
90
- inputs = tokenizer(text, return_tensors="pt")
91
- for i in inputs:
92
- inputs[i] = inputs[i].to(device)
93
- res = bert_model(**inputs, output_hidden_states=True)
94
- res = torch.cat(res["hidden_states"][-3:-2], -1)[0].cpu()[1:-1]
95
- assert len(word2ph) == len(text)
96
- phone_level_feature = []
97
- for i in range(len(word2ph)):
98
- repeat_feature = res[i].repeat(word2ph[i], 1)
99
- phone_level_feature.append(repeat_feature)
100
- phone_level_feature = torch.cat(phone_level_feature, dim=0)
101
- return phone_level_feature.T
102
-
103
-
104
- class DictToAttrRecursive(dict):
105
- def __init__(self, input_dict):
106
- super().__init__(input_dict)
107
- for key, value in input_dict.items():
108
- if isinstance(value, dict):
109
- value = DictToAttrRecursive(value)
110
- self[key] = value
111
- setattr(self, key, value)
112
-
113
- def __getattr__(self, item):
114
- try:
115
- return self[item]
116
- except KeyError:
117
- raise AttributeError(f"Attribute {item} not found")
118
-
119
- def __setattr__(self, key, value):
120
- if isinstance(value, dict):
121
- value = DictToAttrRecursive(value)
122
- super(DictToAttrRecursive, self).__setitem__(key, value)
123
- super().__setattr__(key, value)
124
-
125
- def __delattr__(self, item):
126
- try:
127
- del self[item]
128
- except KeyError:
129
- raise AttributeError(f"Attribute {item} not found")
130
-
131
-
132
- ssl_model = cnhubert.get_model()
133
- if is_half == True:
134
- ssl_model = ssl_model.half().to(device)
135
- else:
136
- ssl_model = ssl_model.to(device)
137
-
138
-
139
- def change_sovits_weights(sovits_path):
140
- global vq_model, hps
141
- dict_s2 = torch.load(sovits_path, map_location="cpu")
142
- hps = dict_s2["config"]
143
- hps = DictToAttrRecursive(hps)
144
- hps.model.semantic_frame_rate = "25hz"
145
- vq_model = SynthesizerTrn(
146
- hps.data.filter_length // 2 + 1,
147
- hps.train.segment_size // hps.data.hop_length,
148
- n_speakers=hps.data.n_speakers,
149
- **hps.model
150
- )
151
- if ("pretrained" not in sovits_path):
152
- del vq_model.enc_q
153
- if is_half == True:
154
- vq_model = vq_model.half().to(device)
155
- else:
156
- vq_model = vq_model.to(device)
157
- vq_model.eval()
158
- print(vq_model.load_state_dict(dict_s2["weight"], strict=False))
159
- with open("./sweight.txt", "w", encoding="utf-8") as f:
160
- f.write(sovits_path)
161
-
162
-
163
- change_sovits_weights(sovits_path)
164
-
165
-
166
- def change_gpt_weights(gpt_path):
167
- global hz, max_sec, t2s_model, config
168
- hz = 50
169
- dict_s1 = torch.load(gpt_path, map_location="cpu")
170
- config = dict_s1["config"]
171
- max_sec = config["data"]["max_sec"]
172
- t2s_model = Text2SemanticLightningModule(config, "****", is_train=False)
173
- t2s_model.load_state_dict(dict_s1["weight"])
174
- if is_half == True:
175
- t2s_model = t2s_model.half()
176
- t2s_model = t2s_model.to(device)
177
- t2s_model.eval()
178
- total = sum([param.nelement() for param in t2s_model.parameters()])
179
- print("Number of parameter: %.2fM" % (total / 1e6))
180
- with open("./gweight.txt", "w", encoding="utf-8") as f: f.write(gpt_path)
181
-
182
-
183
- change_gpt_weights(gpt_path)
184
-
185
-
186
- def get_spepc(hps, filename):
187
- audio = load_audio(filename, int(hps.data.sampling_rate))
188
- audio = torch.FloatTensor(audio)
189
- audio_norm = audio
190
- audio_norm = audio_norm.unsqueeze(0)
191
- spec = spectrogram_torch(
192
- audio_norm,
193
- hps.data.filter_length,
194
- hps.data.sampling_rate,
195
- hps.data.hop_length,
196
- hps.data.win_length,
197
- center=False,
198
- )
199
- return spec
200
-
201
-
202
- dict_language = {
203
- i18n("中文"): "all_zh",#全部按中文识别
204
- i18n("英文"): "en",#全部按英文识别#######不变
205
- i18n("日文"): "all_ja",#全部按日文识别
206
- i18n("中英混合"): "zh",#按中英混合识别####不变
207
- i18n("日英混合"): "ja",#按日英混合识别####不变
208
- i18n("多语种混合"): "auto",#多语种启动切分识别语种
209
- }
210
-
211
-
212
- def splite_en_inf(sentence, language):
213
- pattern = re.compile(r'[a-zA-Z ]+')
214
- textlist = []
215
- langlist = []
216
- pos = 0
217
- for match in pattern.finditer(sentence):
218
- start, end = match.span()
219
- if start > pos:
220
- textlist.append(sentence[pos:start])
221
- langlist.append(language)
222
- textlist.append(sentence[start:end])
223
- langlist.append("en")
224
- pos = end
225
- if pos < len(sentence):
226
- textlist.append(sentence[pos:])
227
- langlist.append(language)
228
- # Merge punctuation into previous word
229
- for i in range(len(textlist)-1, 0, -1):
230
- if re.match(r'^[\W_]+$', textlist[i]):
231
- textlist[i-1] += textlist[i]
232
- del textlist[i]
233
- del langlist[i]
234
- # Merge consecutive words with the same language tag
235
- i = 0
236
- while i < len(langlist) - 1:
237
- if langlist[i] == langlist[i+1]:
238
- textlist[i] += textlist[i+1]
239
- del textlist[i+1]
240
- del langlist[i+1]
241
- else:
242
- i += 1
243
-
244
- return textlist, langlist
245
-
246
-
247
- def clean_text_inf(text, language):
248
- formattext = ""
249
- language = language.replace("all_","")
250
- for tmp in LangSegment.getTexts(text):
251
- if language == "ja":
252
- if tmp["lang"] == language or tmp["lang"] == "zh":
253
- formattext += tmp["text"] + " "
254
- continue
255
- if tmp["lang"] == language:
256
- formattext += tmp["text"] + " "
257
- while " " in formattext:
258
- formattext = formattext.replace(" ", " ")
259
- phones, word2ph, norm_text = clean_text(formattext, language)
260
- phones = cleaned_text_to_sequence(phones)
261
- return phones, word2ph, norm_text
262
-
263
- dtype=torch.float16 if is_half == True else torch.float32
264
- def get_bert_inf(phones, word2ph, norm_text, language):
265
- language=language.replace("all_","")
266
- if language == "zh":
267
- bert = get_bert_feature(norm_text, word2ph).to(device)#.to(dtype)
268
- else:
269
- bert = torch.zeros(
270
- (1024, len(phones)),
271
- dtype=torch.float16 if is_half == True else torch.float32,
272
- ).to(device)
273
-
274
- return bert
275
-
276
-
277
- def nonen_clean_text_inf(text, language):
278
- if(language!="auto"):
279
- textlist, langlist = splite_en_inf(text, language)
280
- else:
281
- textlist=[]
282
- langlist=[]
283
- for tmp in LangSegment.getTexts(text):
284
- langlist.append(tmp["lang"])
285
- textlist.append(tmp["text"])
286
- phones_list = []
287
- word2ph_list = []
288
- norm_text_list = []
289
- for i in range(len(textlist)):
290
- lang = langlist[i]
291
- phones, word2ph, norm_text = clean_text_inf(textlist[i], lang)
292
- phones_list.append(phones)
293
- if lang == "zh":
294
- word2ph_list.append(word2ph)
295
- norm_text_list.append(norm_text)
296
- print(word2ph_list)
297
- phones = sum(phones_list, [])
298
- word2ph = sum(word2ph_list, [])
299
- norm_text = ' '.join(norm_text_list)
300
-
301
- return phones, word2ph, norm_text
302
-
303
-
304
- def nonen_get_bert_inf(text, language):
305
- if(language!="auto"):
306
- textlist, langlist = splite_en_inf(text, language)
307
- else:
308
- textlist=[]
309
- langlist=[]
310
- for tmp in LangSegment.getTexts(text):
311
- langlist.append(tmp["lang"])
312
- textlist.append(tmp["text"])
313
- print(textlist)
314
- print(langlist)
315
- bert_list = []
316
- for i in range(len(textlist)):
317
- lang = langlist[i]
318
- phones, word2ph, norm_text = clean_text_inf(textlist[i], lang)
319
- bert = get_bert_inf(phones, word2ph, norm_text, lang)
320
- bert_list.append(bert)
321
- bert = torch.cat(bert_list, dim=1)
322
-
323
- return bert
324
-
325
-
326
- splits = {",", "。", "?", "!", ",", ".", "?", "!", "~", ":", ":", "—", "…", }
327
-
328
-
329
- def get_first(text):
330
- pattern = "[" + "".join(re.escape(sep) for sep in splits) + "]"
331
- text = re.split(pattern, text)[0].strip()
332
- return text
333
-
334
-
335
- def get_cleaned_text_final(text,language):
336
- if language in {"en","all_zh","all_ja"}:
337
- phones, word2ph, norm_text = clean_text_inf(text, language)
338
- elif language in {"zh", "ja","auto"}:
339
- phones, word2ph, norm_text = nonen_clean_text_inf(text, language)
340
- return phones, word2ph, norm_text
341
-
342
- def get_bert_final(phones, word2ph, text,language,device):
343
- if language == "en":
344
- bert = get_bert_inf(phones, word2ph, text, language)
345
- elif language in {"zh", "ja","auto"}:
346
- bert = nonen_get_bert_inf(text, language)
347
- elif language == "all_zh":
348
- bert = get_bert_feature(text, word2ph).to(device)
349
- else:
350
- bert = torch.zeros((1024, len(phones))).to(device)
351
- return bert
352
-
353
- def merge_short_text_in_array(texts, threshold):
354
- if (len(texts)) < 2:
355
- return texts
356
- result = []
357
- text = ""
358
- for ele in texts:
359
- text += ele
360
- if len(text) >= threshold:
361
- result.append(text)
362
- text = ""
363
- if (len(text) > 0):
364
- if len(result) == 0:
365
- result.append(text)
366
- else:
367
- result[len(result) - 1] += text
368
- return result
369
-
370
- def get_tts_wav(ref_wav_path, prompt_text, prompt_language, text, text_language, how_to_cut=i18n("不切"), top_k=20, top_p=0.6, temperature=0.6, ref_free = False):
371
- if prompt_text is None or len(prompt_text) == 0:
372
- ref_free = True
373
- t0 = ttime()
374
- prompt_language = dict_language[prompt_language]
375
- text_language = dict_language[text_language]
376
- if not ref_free:
377
- prompt_text = prompt_text.strip("\n")
378
- if (prompt_text[-1] not in splits): prompt_text += "。" if prompt_language != "en" else "."
379
- print(i18n("实际输入的参考文本:"), prompt_text)
380
- text = text.strip("\n")
381
- if (text[0] not in splits and len(get_first(text)) < 4): text = "。" + text if text_language != "en" else "." + text
382
-
383
- print(i18n("实际输入的目标文本:"), text)
384
- zero_wav = np.zeros(
385
- int(hps.data.sampling_rate * 0.3),
386
- dtype=np.float16 if is_half == True else np.float32,
387
- )
388
- with torch.no_grad():
389
- wav16k, sr = librosa.load(ref_wav_path, sr=16000)
390
- if (wav16k.shape[0] > 160000 or wav16k.shape[0] < 48000):
391
- raise OSError(i18n("参考音频在3~10秒范围外,请更换!"))
392
- wav16k = torch.from_numpy(wav16k)
393
- zero_wav_torch = torch.from_numpy(zero_wav)
394
- if is_half == True:
395
- wav16k = wav16k.half().to(device)
396
- zero_wav_torch = zero_wav_torch.half().to(device)
397
- else:
398
- wav16k = wav16k.to(device)
399
- zero_wav_torch = zero_wav_torch.to(device)
400
- wav16k = torch.cat([wav16k, zero_wav_torch])
401
- ssl_content = ssl_model.model(wav16k.unsqueeze(0))[
402
- "last_hidden_state"
403
- ].transpose(
404
- 1, 2
405
- ) # .float()
406
- codes = vq_model.extract_latent(ssl_content)
407
-
408
- prompt_semantic = codes[0, 0]
409
- t1 = ttime()
410
-
411
- if (how_to_cut == i18n("凑四句一切")):
412
- text = cut1(text)
413
- elif (how_to_cut == i18n("凑50字一切")):
414
- text = cut2(text)
415
- elif (how_to_cut == i18n("按中文句号。切")):
416
- text = cut3(text)
417
- elif (how_to_cut == i18n("按英文句号.切")):
418
- text = cut4(text)
419
- elif (how_to_cut == i18n("按标点符号切")):
420
- text = cut5(text)
421
- while "\n\n" in text:
422
- text = text.replace("\n\n", "\n")
423
- print(i18n("实际输入的目标文本(切句后):"), text)
424
- texts = text.split("\n")
425
- texts = merge_short_text_in_array(texts, 5)
426
- audio_opt = []
427
- if not ref_free:
428
- phones1, word2ph1, norm_text1=get_cleaned_text_final(prompt_text, prompt_language)
429
- print("前端处理后的参考文本:%s"%norm_text1)
430
- bert1=get_bert_final(phones1, word2ph1, norm_text1,prompt_language,device).to(dtype)
431
-
432
- for text in texts:
433
- # 解决输入目标文本的空行导致报错的问题
434
- if (len(text.strip()) == 0):
435
- continue
436
- if (text[-1] not in splits): text += "。" if text_language != "en" else "."
437
- print(i18n("实际输入的目标文本(每句):"), text)
438
- phones2, word2ph2, norm_text2 = get_cleaned_text_final(text, text_language)
439
- print(i18n("前端处理后的文本(每句):"), norm_text2)
440
- bert2 = get_bert_final(phones2, word2ph2, norm_text2, text_language, device).to(dtype)
441
- if not ref_free:
442
- bert = torch.cat([bert1, bert2], 1)
443
- all_phoneme_ids = torch.LongTensor(phones1+phones2).to(device).unsqueeze(0)
444
- else:
445
- bert = bert2
446
- all_phoneme_ids = torch.LongTensor(phones2).to(device).unsqueeze(0)
447
-
448
- bert = bert.to(device).unsqueeze(0)
449
- all_phoneme_len = torch.tensor([all_phoneme_ids.shape[-1]]).to(device)
450
- prompt = prompt_semantic.unsqueeze(0).to(device)
451
- t2 = ttime()
452
- with torch.no_grad():
453
- # pred_semantic = t2s_model.model.infer(
454
- pred_semantic, idx = t2s_model.model.infer_panel(
455
- all_phoneme_ids,
456
- all_phoneme_len,
457
- None if ref_free else prompt,
458
- bert,
459
- # prompt_phone_len=ph_offset,
460
- top_k=top_k,
461
- top_p=top_p,
462
- temperature=temperature,
463
- early_stop_num=hz * max_sec,
464
- )
465
- t3 = ttime()
466
- # print(pred_semantic.shape,idx)
467
- pred_semantic = pred_semantic[:, -idx:].unsqueeze(
468
- 0
469
- ) # .unsqueeze(0)#mq要多unsqueeze一次
470
- refer = get_spepc(hps, ref_wav_path) # .to(device)
471
- if is_half == True:
472
- refer = refer.half().to(device)
473
- else:
474
- refer = refer.to(device)
475
- # audio = vq_model.decode(pred_semantic, all_phoneme_ids, refer).detach().cpu().numpy()[0, 0]
476
- audio = (
477
- vq_model.decode(
478
- pred_semantic, torch.LongTensor(phones2).to(device).unsqueeze(0), refer
479
- )
480
- .detach()
481
- .cpu()
482
- .numpy()[0, 0]
483
- ) ###试试重建不带上prompt部分
484
- max_audio=np.abs(audio).max()#简单防止16bit爆音
485
- if max_audio>1:audio/=max_audio
486
- audio_opt.append(audio)
487
- audio_opt.append(zero_wav)
488
- t4 = ttime()
489
- print("%.3f\t%.3f\t%.3f\t%.3f" % (t1 - t0, t2 - t1, t3 - t2, t4 - t3))
490
- yield hps.data.sampling_rate, (np.concatenate(audio_opt, 0) * 32768).astype(
491
- np.int16
492
- )
493
-
494
-
495
- def split(todo_text):
496
- todo_text = todo_text.replace("……", "。").replace("——", ",")
497
- if todo_text[-1] not in splits:
498
- todo_text += "。"
499
- i_split_head = i_split_tail = 0
500
- len_text = len(todo_text)
501
- todo_texts = []
502
- while 1:
503
- if i_split_head >= len_text:
504
- break # 结尾一定有标点,所以直接跳出即可,最后一段在上次已加入
505
- if todo_text[i_split_head] in splits:
506
- i_split_head += 1
507
- todo_texts.append(todo_text[i_split_tail:i_split_head])
508
- i_split_tail = i_split_head
509
- else:
510
- i_split_head += 1
511
- return todo_texts
512
-
513
-
514
- def cut1(inp):
515
- inp = inp.strip("\n")
516
- inps = split(inp)
517
- split_idx = list(range(0, len(inps), 4))
518
- split_idx[-1] = None
519
- if len(split_idx) > 1:
520
- opts = []
521
- for idx in range(len(split_idx) - 1):
522
- opts.append("".join(inps[split_idx[idx]: split_idx[idx + 1]]))
523
- else:
524
- opts = [inp]
525
- return "\n".join(opts)
526
-
527
-
528
- def cut2(inp):
529
- inp = inp.strip("\n")
530
- inps = split(inp)
531
- if len(inps) < 2:
532
- return inp
533
- opts = []
534
- summ = 0
535
- tmp_str = ""
536
- for i in range(len(inps)):
537
- summ += len(inps[i])
538
- tmp_str += inps[i]
539
- if summ > 50:
540
- summ = 0
541
- opts.append(tmp_str)
542
- tmp_str = ""
543
- if tmp_str != "":
544
- opts.append(tmp_str)
545
- # print(opts)
546
- if len(opts) > 1 and len(opts[-1]) < 50: ##如果最后一个太短了,和前一个合一起
547
- opts[-2] = opts[-2] + opts[-1]
548
- opts = opts[:-1]
549
- return "\n".join(opts)
550
-
551
-
552
- def cut3(inp):
553
- inp = inp.strip("\n")
554
- return "\n".join(["%s" % item for item in inp.strip("。").split("。")])
555
-
556
-
557
- def cut4(inp):
558
- inp = inp.strip("\n")
559
- return "\n".join(["%s" % item for item in inp.strip(".").split(".")])
560
-
561
-
562
- # contributed by https://github.com/AI-Hobbyist/GPT-SoVITS/blob/main/GPT_SoVITS/inference_webui.py
563
- def cut5(inp):
564
- # if not re.search(r'[^\w\s]', inp[-1]):
565
- # inp += '。'
566
- inp = inp.strip("\n")
567
- punds = r'[,.;?!、,。?!;:]'
568
- items = re.split(f'({punds})', inp)
569
- items = ["".join(group) for group in zip(items[::2], items[1::2])]
570
- opt = "\n".join(items)
571
- return opt
572
-
573
-
574
- def custom_sort_key(s):
575
- # 使用正则表达式提取字符串中的数字部分和非数字部分
576
- parts = re.split('(\d+)', s)
577
- # 将数字部分转换为整数,非数字部分保持不变
578
- parts = [int(part) if part.isdigit() else part for part in parts]
579
- return parts
580
-
581
-
582
- def change_choices():
583
- SoVITS_names, GPT_names = get_weights_names()
584
- return {"choices": sorted(SoVITS_names, key=custom_sort_key), "__type__": "update"}, {"choices": sorted(GPT_names, key=custom_sort_key), "__type__": "update"}
585
-
586
-
587
- pretrained_sovits_name = "GPT_SoVITS/pretrained_models/s2G488k.pth"
588
- pretrained_gpt_name = "GPT_SoVITS/pretrained_models/s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt"
589
- SoVITS_weight_root = "SoVITS_weights"
590
- GPT_weight_root = "GPT_weights"
591
- os.makedirs(SoVITS_weight_root, exist_ok=True)
592
- os.makedirs(GPT_weight_root, exist_ok=True)
593
-
594
-
595
- def get_weights_names():
596
- SoVITS_names = [pretrained_sovits_name]
597
- for name in os.listdir(SoVITS_weight_root):
598
- if name.endswith(".pth"): SoVITS_names.append("%s/%s" % (SoVITS_weight_root, name))
599
- GPT_names = [pretrained_gpt_name]
600
- for name in os.listdir(GPT_weight_root):
601
- if name.endswith(".ckpt"): GPT_names.append("%s/%s" % (GPT_weight_root, name))
602
- return SoVITS_names, GPT_names
603
-
604
-
605
- SoVITS_names, GPT_names = get_weights_names()
606
-
607
- with gr.Blocks(title="GPT-SoVITS WebUI") as app:
608
- gr.Markdown(
609
- value=i18n("本软件以MIT协议开源, 作者不对软件具备任何控制力, 使用软件者、传播软件导出的声音者自负全责. <br>如不认可该条款, 则不能使用或引用软件包内任何代码和文件. 详见根目录<b>LICENSE</b>.")
610
- )
611
- with gr.Group():
612
- gr.Markdown(value=i18n("模型切换"))
613
- with gr.Row():
614
- GPT_dropdown = gr.Dropdown(label=i18n("GPT模型列表"), choices=sorted(GPT_names, key=custom_sort_key), value=gpt_path, interactive=True)
615
- SoVITS_dropdown = gr.Dropdown(label=i18n("SoVITS模型列表"), choices=sorted(SoVITS_names, key=custom_sort_key), value=sovits_path, interactive=True)
616
- refresh_button = gr.Button(i18n("刷新模型路径"), variant="primary")
617
- refresh_button.click(fn=change_choices, inputs=[], outputs=[SoVITS_dropdown, GPT_dropdown])
618
- SoVITS_dropdown.change(change_sovits_weights, [SoVITS_dropdown], [])
619
- GPT_dropdown.change(change_gpt_weights, [GPT_dropdown], [])
620
- gr.Markdown(value=i18n("*请上传并填写参考信息"))
621
- with gr.Row():
622
- inp_ref = gr.Audio(label=i18n("请上传3~10秒内参考音频,超过会报错!"), type="filepath")
623
- with gr.Column():
624
- ref_text_free = gr.Checkbox(label=i18n("开启无参考文本模式。不填参考文本亦相当于开启。"), value=False, interactive=True, show_label=True)
625
- gr.Markdown(i18n("使用无参考文本模式时建议使用微调的GPT"))
626
- prompt_text = gr.Textbox(label=i18n("参考音频的文本"), value="")
627
- prompt_language = gr.Dropdown(
628
- label=i18n("参考音频的语种"), choices=[i18n("中文"), i18n("英文"), i18n("日文"), i18n("中英混合"), i18n("日英混合"), i18n("多语种混合")], value=i18n("中文")
629
- )
630
- gr.Markdown(value=i18n("*请填写需要合成的目标文本。中英混合选中文,日英混合选日文,中日混合暂不支持,非目标语言文本自动遗弃。"))
631
- with gr.Row():
632
- text = gr.Textbox(label=i18n("需要合成的文本"), value="")
633
- text_language = gr.Dropdown(
634
- label=i18n("需要合成的语种"), choices=[i18n("中文"), i18n("英文"), i18n("日文"), i18n("中英混合"), i18n("日英混合"), i18n("多语种混合")], value=i18n("中文")
635
- )
636
- how_to_cut = gr.Radio(
637
- label=i18n("怎么切"),
638
- choices=[i18n("不切"), i18n("凑四句一切"), i18n("凑50字一切"), i18n("按中文句号。切"), i18n("按英文句号.切"), i18n("按标点符号切"), ],
639
- value=i18n("凑四句一切"),
640
- interactive=True,
641
- )
642
- with gr.Row():
643
- gr.Markdown("gpt采样参数(无参考文本时不要太低):")
644
- top_k = gr.Slider(minimum=1,maximum=100,step=1,label=i18n("top_k"),value=5,interactive=True)
645
- top_p = gr.Slider(minimum=0,maximum=1,step=0.05,label=i18n("top_p"),value=1,interactive=True)
646
- temperature = gr.Slider(minimum=0,maximum=1,step=0.05,label=i18n("temperature"),value=1,interactive=True)
647
- inference_button = gr.Button(i18n("合成语音"), variant="primary")
648
- output = gr.Audio(label=i18n("输出的语音"))
649
-
650
- inference_button.click(
651
- get_tts_wav,
652
- [inp_ref, prompt_text, prompt_language, text, text_language, how_to_cut, top_k, top_p, temperature, ref_text_free],
653
- [output], api_name="GetVoice"
654
- )
655
-
656
- gr.Markdown(value=i18n("文本切分工具。太长的文本合成出来效果不一定好,所以太长建议先切。合成会根据文本的换行分开合成再拼起来。"))
657
- with gr.Row():
658
- text_inp = gr.Textbox(label=i18n("需要合成的切分前文本"), value="")
659
- button1 = gr.Button(i18n("凑四句一切"), variant="primary")
660
- button2 = gr.Button(i18n("凑50字一切"), variant="primary")
661
- button3 = gr.Button(i18n("按中文句号。切"), variant="primary")
662
- button4 = gr.Button(i18n("按英文句号.切"), variant="primary")
663
- button5 = gr.Button(i18n("按标点符号切"), variant="primary")
664
- text_opt = gr.Textbox(label=i18n("切分后文本"), value="")
665
- button1.click(cut1, [text_inp], [text_opt])
666
- button2.click(cut2, [text_inp], [text_opt])
667
- button3.click(cut3, [text_inp], [text_opt])
668
- button4.click(cut4, [text_inp], [text_opt])
669
- button5.click(cut5, [text_inp], [text_opt])
670
- gr.Markdown(value=i18n("后续将支持转音素、手工修改音素、语音合成分步执行。"))
671
-
672
- app.queue(concurrency_count=511, max_size=1022).launch(
673
- server_name="0.0.0.0",
674
- inbrowser=True,
675
- share=is_share,
676
- server_port=infer_ttswebui,
677
- quiet=True,
678
- )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s1.yaml DELETED
@@ -1,31 +0,0 @@
1
- train:
2
- seed: 1234
3
- epochs: 300
4
- batch_size: 8
5
- gradient_accumulation: 4
6
- save_every_n_epoch: 1
7
- precision: 16
8
- gradient_clip: 1.0
9
- optimizer:
10
- lr: 0.01
11
- lr_init: 0.00001
12
- lr_end: 0.0001
13
- warmup_steps: 2000
14
- decay_steps: 40000
15
- data:
16
- max_eval_sample: 8
17
- max_sec: 54
18
- num_workers: 1
19
- pad_val: 1024 # same with EOS in model
20
- model:
21
- vocab_size: 1025
22
- phoneme_vocab_size: 512
23
- embedding_dim: 512
24
- hidden_dim: 512
25
- head: 16
26
- linear_units: 2048
27
- n_layer: 12
28
- dropout: 0
29
- EOS: 1024
30
- inference:
31
- top_k: 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s1big.yaml DELETED
@@ -1,31 +0,0 @@
1
- train:
2
- seed: 1234
3
- epochs: 300
4
- batch_size: 8
5
- gradient_accumulation: 4
6
- save_every_n_epoch: 1
7
- precision: 16-mixed
8
- gradient_clip: 1.0
9
- optimizer:
10
- lr: 0.01
11
- lr_init: 0.00001
12
- lr_end: 0.0001
13
- warmup_steps: 2000
14
- decay_steps: 40000
15
- data:
16
- max_eval_sample: 8
17
- max_sec: 54
18
- num_workers: 1
19
- pad_val: 1024 # same with EOS in model
20
- model:
21
- vocab_size: 1025
22
- phoneme_vocab_size: 512
23
- embedding_dim: 1024
24
- hidden_dim: 1024
25
- head: 16
26
- linear_units: 2048
27
- n_layer: 16
28
- dropout: 0
29
- EOS: 1024
30
- inference:
31
- top_k: 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s1big2.yaml DELETED
@@ -1,31 +0,0 @@
1
- train:
2
- seed: 1234
3
- epochs: 300
4
- batch_size: 12
5
- gradient_accumulation: 4
6
- save_every_n_epoch: 1
7
- precision: 16-mixed
8
- gradient_clip: 1.0
9
- optimizer:
10
- lr: 0.01
11
- lr_init: 0.00001
12
- lr_end: 0.0001
13
- warmup_steps: 2000
14
- decay_steps: 40000
15
- data:
16
- max_eval_sample: 8
17
- max_sec: 54
18
- num_workers: 1
19
- pad_val: 1024 # same with EOS in model
20
- model:
21
- vocab_size: 1025
22
- phoneme_vocab_size: 512
23
- embedding_dim: 1024
24
- hidden_dim: 1024
25
- head: 16
26
- linear_units: 2048
27
- n_layer: 6
28
- dropout: 0
29
- EOS: 1024
30
- inference:
31
- top_k: 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s1longer.yaml DELETED
@@ -1,31 +0,0 @@
1
- train:
2
- seed: 1234
3
- epochs: 20
4
- batch_size: 8
5
- save_every_n_epoch: 1
6
- precision: 16-mixed
7
- gradient_clip: 1.0
8
- optimizer:
9
- lr: 0.01
10
- lr_init: 0.00001
11
- lr_end: 0.0001
12
- warmup_steps: 2000
13
- decay_steps: 40000
14
- data:
15
- max_eval_sample: 8
16
- max_sec: 54
17
- num_workers: 4
18
- pad_val: 1024 # same with EOS in model
19
- model:
20
- vocab_size: 1025
21
- phoneme_vocab_size: 512
22
- embedding_dim: 512
23
- hidden_dim: 512
24
- head: 16
25
- linear_units: 2048
26
- n_layer: 24
27
- dropout: 0
28
- EOS: 1024
29
- random_bert: 0
30
- inference:
31
- top_k: 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s1mq.yaml DELETED
@@ -1,77 +0,0 @@
1
- train:
2
- seed: 1234
3
- epochs: 100
4
- batch_size: 6
5
- gradient_accumulation: 4
6
- save_every_n_epoch: 1
7
- precision: 32
8
- gradient_clip: 1.0
9
- optimizer:
10
- lr: 0.01
11
- lr_init: 0.00001
12
- lr_end: 0.0001
13
- warmup_steps: 2000
14
- decay_steps: 40000
15
- data:
16
- max_eval_sample: 8
17
- max_sec: 40
18
- num_workers: 1
19
- pad_val: 1024 # same with EOS in model
20
- model:
21
- saving_path: "ckpt/"
22
- resume_checkpoint: null
23
- vocoder_config_path: "quantizer/new_ckpt/config.json"
24
- vocoder_ckpt_path: "quantizer/new_ckpt/g_00600000"
25
- datadir: "/home/liweiche/GigaSpeech/wavs"
26
- metapath: "/home/liweiche/GigaSpeech/train2.json"
27
- val_metapath: "/home/liweiche/GigaSpeech/dev2.json"
28
- sampledir: "logs/"
29
- pretrained_path: null
30
- lr: 0.0001
31
- batch_size: 200.0
32
- train_bucket_size: 8192
33
- training_step: 800000
34
- optim_flat_percent: 0.0
35
- warmup_step: 50
36
- adam_beta1: 0.9
37
- adam_beta2: 0.98
38
- ffd_size: 3072
39
- hidden_size: 768
40
- enc_nlayers: 6
41
- dec_nlayers: 6
42
- nheads: 12
43
- ar_layer: 4
44
- ar_ffd_size: 1024
45
- ar_hidden_size: 256
46
- ar_nheads: 4
47
- aligner_softmax_temp: 1.0
48
- layer_norm_eps: 0.00001
49
- speaker_embed_dropout: 0.05
50
- label_smoothing: 0.0
51
- val_check_interval: 5000
52
- check_val_every_n_epoch: 1
53
- precision: "fp16"
54
- nworkers: 16
55
- distributed: true
56
- accelerator: "ddp"
57
- version: null
58
- accumulate_grad_batches: 1
59
- use_repetition_token: true
60
- use_repetition_gating: false
61
- repetition_penalty: 1.0
62
- sampling_temperature: 1.0
63
- top_k: -1
64
- min_top_k: 3
65
- top_p: 0.8
66
- sample_num: 4
67
- length_penalty_max_length: 15000
68
- length_penalty_max_prob: 0.95
69
- max_input_length: 2048
70
- max_output_length: 2000
71
- sample_rate: 16000
72
- n_codes: 1024
73
- n_cluster_groups: 1
74
- phone_context_window: 4
75
- phoneset_size: 1000
76
- inference:
77
- top_k: 5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
configs/s2.json DELETED
@@ -1,90 +0,0 @@
1
- {
2
- "train": {
3
- "log_interval": 100,
4
- "eval_interval": 500,
5
- "seed": 1234,
6
- "epochs": 100,
7
- "learning_rate": 0.0001,
8
- "betas": [
9
- 0.8,
10
- 0.99
11
- ],
12
- "eps": 1e-09,
13
- "batch_size": 32,
14
- "fp16_run": true,
15
- "lr_decay": 0.999875,
16
- "segment_size": 20480,
17
- "init_lr_ratio": 1,
18
- "warmup_epochs": 0,
19
- "c_mel": 45,
20
- "c_kl": 1.0,
21
- "text_low_lr_rate": 0.4
22
- },
23
- "data": {
24
- "max_wav_value": 32768.0,
25
- "sampling_rate": 32000,
26
- "filter_length": 2048,
27
- "hop_length": 640,
28
- "win_length": 2048,
29
- "n_mel_channels": 128,
30
- "mel_fmin": 0.0,
31
- "mel_fmax": null,
32
- "add_blank": true,
33
- "n_speakers": 300,
34
- "cleaned_text": true
35
- },
36
- "model": {
37
- "inter_channels": 192,
38
- "hidden_channels": 192,
39
- "filter_channels": 768,
40
- "n_heads": 2,
41
- "n_layers": 6,
42
- "kernel_size": 3,
43
- "p_dropout": 0.1,
44
- "resblock": "1",
45
- "resblock_kernel_sizes": [
46
- 3,
47
- 7,
48
- 11
49
- ],
50
- "resblock_dilation_sizes": [
51
- [
52
- 1,
53
- 3,
54
- 5
55
- ],
56
- [
57
- 1,
58
- 3,
59
- 5
60
- ],
61
- [
62
- 1,
63
- 3,
64
- 5
65
- ]
66
- ],
67
- "upsample_rates": [
68
- 10,
69
- 8,
70
- 2,
71
- 2,
72
- 2
73
- ],
74
- "upsample_initial_channel": 512,
75
- "upsample_kernel_sizes": [
76
- 16,
77
- 16,
78
- 8,
79
- 2,
80
- 2
81
- ],
82
- "n_layers_q": 3,
83
- "use_spectral_norm": false,
84
- "gin_channels": 512,
85
- "semantic_frame_rate": "25hz",
86
- "freeze_quantizer": true
87
- },
88
- "s2_ckpt_dir": "logs/s2/big2k1",
89
- "content_module": "cnhubert"
90
- }