File size: 7,588 Bytes
85d3b29
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1a7d583
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
85d3b29
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1a7d583
 
85d3b29
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
import os
import glob
import json
import torch
import argparse
import numpy as np
from scipy.io.wavfile import read


def load_checkpoint(checkpoint_path, model, optimizer=None, load_opt=1):
    assert os.path.isfile(checkpoint_path)
    checkpoint_dict = torch.load(checkpoint_path, map_location="cpu")

    saved_state_dict = checkpoint_dict["model"]
    if hasattr(model, "module"):
        state_dict = model.module.state_dict()
    else:
        state_dict = model.state_dict()
    new_state_dict = {}
    for k, v in state_dict.items():
        try:
            new_state_dict[k] = saved_state_dict[k]
            if saved_state_dict[k].shape != state_dict[k].shape:
                print(
                    "shape-%s-mismatch|need-%s|get-%s",
                    k,
                    state_dict[k].shape,
                    saved_state_dict[k].shape,
                )
                raise KeyError
        except:
            print("%s is not in the checkpoint", k)
            new_state_dict[k] = v
    if hasattr(model, "module"):
        model.module.load_state_dict(new_state_dict, strict=False)
    else:
        model.load_state_dict(new_state_dict, strict=False)

    iteration = checkpoint_dict["iteration"]
    learning_rate = checkpoint_dict["learning_rate"]
    if optimizer is not None and load_opt == 1:
        optimizer.load_state_dict(checkpoint_dict["optimizer"])
    print(f"Loaded checkpoint '{checkpoint_path}' (epoch {iteration})")
    return model, optimizer, learning_rate, iteration


def save_checkpoint(model, optimizer, learning_rate, iteration, checkpoint_path):
    print(f"Saved model '{checkpoint_path}' (epoch {iteration})")
    if hasattr(model, "module"):
        state_dict = model.module.state_dict()
    else:
        state_dict = model.state_dict()
    torch.save(
        {
            "model": state_dict,
            "iteration": iteration,
            "optimizer": optimizer.state_dict(),
            "learning_rate": learning_rate,
        },
        checkpoint_path,
    )


def summarize(
    writer,
    global_step,
    scalars={},
    histograms={},
    images={},
    audios={},
    audio_sampling_rate=22050,
):
    for k, v in scalars.items():
        writer.add_scalar(k, v, global_step)
    for k, v in histograms.items():
        writer.add_histogram(k, v, global_step)
    for k, v in images.items():
        writer.add_image(k, v, global_step, dataformats="HWC")
    for k, v in audios.items():
        writer.add_audio(k, v, global_step, audio_sampling_rate)


def latest_checkpoint_path(dir_path, regex="G_*.pth"):
    f_list = glob.glob(os.path.join(dir_path, regex))
    f_list.sort(key=lambda f: int("".join(filter(str.isdigit, f))))
    x = f_list[-1]
    return x


def plot_spectrogram_to_numpy(spectrogram):
    import matplotlib.pylab as plt
    import numpy as np

    fig, ax = plt.subplots(figsize=(10, 2))
    im = ax.imshow(spectrogram, aspect="auto", origin="lower", interpolation="none")
    plt.colorbar(im, ax=ax)
    plt.xlabel("Frames")
    plt.ylabel("Channels")
    plt.tight_layout()

    fig.canvas.draw()
    data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep="")
    data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))
    plt.close()
    return data


def load_wav_to_torch(full_path):
    sampling_rate, data = read(full_path)
    return torch.FloatTensor(data.astype(np.float32)), sampling_rate


def load_filepaths_and_text(filename, split="|"):
    with open(filename, encoding="utf-8") as f:
        filepaths_and_text = [line.strip().split(split) for line in f]
    return filepaths_and_text


def get_hparams():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "-se",
        "--save_every_epoch",
        type=int,
        required=True,
        help="checkpoint save frequency (epoch)",
    )
    parser.add_argument(
        "-te", "--total_epoch", type=int, required=True, help="total_epoch"
    )
    parser.add_argument(
        "-pg", "--pretrainG", type=str, default="", help="Pretrained Discriminator path"
    )
    parser.add_argument(
        "-pd", "--pretrainD", type=str, default="", help="Pretrained Generator path"
    )
    parser.add_argument("-g", "--gpus", type=str, default="0", help="split by -")
    parser.add_argument(
        "-bs", "--batch_size", type=int, required=True, help="batch size"
    )
    parser.add_argument(
        "-e", "--experiment_dir", type=str, required=True, help="experiment dir"
    )
    parser.add_argument(
        "-sr", "--sample_rate", type=str, required=True, help="sample rate, 32k/40k/48k"
    )
    parser.add_argument(
        "-sw",
        "--save_every_weights",
        type=str,
        default="0",
        help="save the extracted model in weights directory when saving checkpoints",
    )
    parser.add_argument(
        "-v", "--version", type=str, required=True, help="model version"
    )
    parser.add_argument(
        "-f0",
        "--if_f0",
        type=int,
        required=True,
        help="use f0 as one of the inputs of the model, 1 or 0",
    )
    parser.add_argument(
        "-l",
        "--if_latest",
        type=int,
        required=True,
        help="if only save the latest G/D pth file, 1 or 0",
    )
    parser.add_argument(
        "-c",
        "--if_cache_data_in_gpu",
        type=int,
        required=True,
        help="if caching the dataset in GPU memory, 1 or 0",
    )

    parser.add_argument(
        "-od",
        "--overtraining_detector",
        type=int,
        required=True,
        help="Detect overtraining or not, 1 or 0",
    )
    parser.add_argument(
        "-ot",
        "--overtraining_threshold",
        type=int,
        default=50,
        help="overtraining_threshold",
    )

    args = parser.parse_args()
    name = args.experiment_dir
    experiment_dir = os.path.join("./logs", args.experiment_dir)
    config_save_path = os.path.join(experiment_dir, "config.json")
    with open(config_save_path, "r") as f:
        config = json.load(f)
    hparams = HParams(**config)
    hparams.model_dir = hparams.experiment_dir = experiment_dir
    hparams.save_every_epoch = args.save_every_epoch
    hparams.name = name
    hparams.total_epoch = args.total_epoch
    hparams.pretrainG = args.pretrainG
    hparams.pretrainD = args.pretrainD
    hparams.version = args.version
    hparams.gpus = args.gpus
    hparams.train.batch_size = args.batch_size
    hparams.sample_rate = args.sample_rate
    hparams.if_f0 = args.if_f0
    hparams.if_latest = args.if_latest
    hparams.save_every_weights = args.save_every_weights
    hparams.if_cache_data_in_gpu = args.if_cache_data_in_gpu
    hparams.data.training_files = f"{experiment_dir}/filelist.txt"
    hparams.overtraining_detector = args.overtraining_detector
    hparams.overtraining_threshold = args.overtraining_threshold
    return hparams


class HParams:
    def __init__(self, **kwargs):
        for k, v in kwargs.items():
            if type(v) == dict:
                v = HParams(**v)
            self[k] = v

    def keys(self):
        return self.__dict__.keys()

    def items(self):
        return self.__dict__.items()

    def values(self):
        return self.__dict__.values()

    def __len__(self):
        return len(self.__dict__)

    def __getitem__(self, key):
        return getattr(self, key)

    def __setitem__(self, key, value):
        return setattr(self, key, value)

    def __contains__(self, key):
        return key in self.__dict__

    def __repr__(self):
        return self.__dict__.__repr__()