Md Mushfiqur Rahman
commited on
Commit
•
c7e4dd5
1
Parent(s):
2efdc5c
Upload with huggingface_hub
Browse files- README.md +15 -0
- all_results.json +20 -0
- config.json +67 -0
- eval_predictions.csv +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- test_predictions.csv +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +649 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- canine
|
5 |
+
- pretrained-on-english-language
|
6 |
+
---
|
7 |
+
|
8 |
+
### How to use
|
9 |
+
|
10 |
+
Here is how to use this model:
|
11 |
+
|
12 |
+
```python
|
13 |
+
from transformers import CanineModel
|
14 |
+
model = CanineModel.from_pretrained('mushfiqur11/<repo name>')
|
15 |
+
```
|
all_results.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 181.82,
|
3 |
+
"eval_accuracy": 0.8567830467257253,
|
4 |
+
"eval_loss": 1.434157371520996,
|
5 |
+
"eval_runtime": 3.1597,
|
6 |
+
"eval_samples": 800,
|
7 |
+
"eval_samples_per_second": 253.186,
|
8 |
+
"eval_steps_per_second": 31.648,
|
9 |
+
"test_accuracy": 0.8472605604349644,
|
10 |
+
"test_loss": 1.4741781949996948,
|
11 |
+
"test_runtime": 3.1659,
|
12 |
+
"test_samples": 800,
|
13 |
+
"test_samples_per_second": 252.694,
|
14 |
+
"test_steps_per_second": 31.587,
|
15 |
+
"train_loss": 0.04628073706757277,
|
16 |
+
"train_runtime": 1159.4207,
|
17 |
+
"train_samples": 1400,
|
18 |
+
"train_samples_per_second": 414.0,
|
19 |
+
"train_steps_per_second": 12.937
|
20 |
+
}
|
config.json
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/canine-s",
|
3 |
+
"architectures": [
|
4 |
+
"CanineForTokenClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 57344,
|
8 |
+
"downsampling_rate": 4,
|
9 |
+
"eos_token_id": 57345,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "ADJ",
|
15 |
+
"1": "ADP",
|
16 |
+
"2": "ADV",
|
17 |
+
"3": "AUX",
|
18 |
+
"4": "CCONJ",
|
19 |
+
"5": "DET",
|
20 |
+
"6": "INTJ",
|
21 |
+
"7": "NOUN",
|
22 |
+
"8": "NUM",
|
23 |
+
"9": "PART",
|
24 |
+
"10": "PRON",
|
25 |
+
"11": "PROPN",
|
26 |
+
"12": "PUNCT",
|
27 |
+
"13": "SCONJ",
|
28 |
+
"14": "SYM",
|
29 |
+
"15": "VERB",
|
30 |
+
"16": "X"
|
31 |
+
},
|
32 |
+
"initializer_range": 0.02,
|
33 |
+
"intermediate_size": 3072,
|
34 |
+
"label2id": {
|
35 |
+
"ADJ": 0,
|
36 |
+
"ADP": 1,
|
37 |
+
"ADV": 2,
|
38 |
+
"AUX": 3,
|
39 |
+
"CCONJ": 4,
|
40 |
+
"DET": 5,
|
41 |
+
"INTJ": 6,
|
42 |
+
"NOUN": 7,
|
43 |
+
"NUM": 8,
|
44 |
+
"PART": 9,
|
45 |
+
"PRON": 10,
|
46 |
+
"PROPN": 11,
|
47 |
+
"PUNCT": 12,
|
48 |
+
"SCONJ": 13,
|
49 |
+
"SYM": 14,
|
50 |
+
"VERB": 15,
|
51 |
+
"X": 16
|
52 |
+
},
|
53 |
+
"layer_norm_eps": 1e-12,
|
54 |
+
"local_transformer_stride": 128,
|
55 |
+
"max_position_embeddings": 16384,
|
56 |
+
"model_type": "canine",
|
57 |
+
"num_attention_heads": 12,
|
58 |
+
"num_hash_buckets": 16384,
|
59 |
+
"num_hash_functions": 8,
|
60 |
+
"num_hidden_layers": 12,
|
61 |
+
"pad_token_id": 0,
|
62 |
+
"torch_dtype": "float32",
|
63 |
+
"transformers_version": "4.17.0",
|
64 |
+
"type_vocab_size": 16,
|
65 |
+
"upsampling_kernel_size": 4,
|
66 |
+
"use_cache": true
|
67 |
+
}
|
eval_predictions.csv
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a5aa416ad637205cbecf655e4a50bd023c88c4aa5a484c4806871c5b8f8b4f42
|
3 |
+
size 528619377
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "sep_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": {"content": "\u0000", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "cls_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "mask_token": {"content": "", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
|
test_predictions.csv
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sep_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "cls_token": {"content": "", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "pad_token": {"content": "\u0000", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "mask_token": {"content": "", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "model_max_length": 2048, "special_tokens_map_file": "/home/mrahma45/.cache/huggingface/transformers/00ffef2919c64d09945b6c06b1c03c4d18bf55f88013390d4443a1eb8c4f8a10.ab71f530366fe02e2834427e7b90198bfd0d573bc4279bfafdb2b95fe2b46dde", "name_or_path": "google/canine-s", "tokenizer_class": "CanineTokenizer"}
|
trainer_state.json
ADDED
@@ -0,0 +1,649 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8567830467257253,
|
3 |
+
"best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-pos-ud-Vietnamese-VTB/checkpoint-5500",
|
4 |
+
"epoch": 181.8181818181818,
|
5 |
+
"global_step": 8000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 2.27,
|
12 |
+
"learning_rate": 8e-05,
|
13 |
+
"loss": 1.7158,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 4.55,
|
18 |
+
"learning_rate": 7.946308724832215e-05,
|
19 |
+
"loss": 0.6087,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 6.82,
|
24 |
+
"learning_rate": 7.89261744966443e-05,
|
25 |
+
"loss": 0.3469,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 9.09,
|
30 |
+
"learning_rate": 7.838926174496645e-05,
|
31 |
+
"loss": 0.2155,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 11.36,
|
36 |
+
"learning_rate": 7.78523489932886e-05,
|
37 |
+
"loss": 0.1352,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 11.36,
|
42 |
+
"eval_accuracy": 0.845839847142609,
|
43 |
+
"eval_loss": 0.7401469945907593,
|
44 |
+
"eval_runtime": 3.1919,
|
45 |
+
"eval_samples_per_second": 250.631,
|
46 |
+
"eval_steps_per_second": 31.329,
|
47 |
+
"step": 500
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 13.64,
|
51 |
+
"learning_rate": 7.731543624161075e-05,
|
52 |
+
"loss": 0.0882,
|
53 |
+
"step": 600
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 15.91,
|
57 |
+
"learning_rate": 7.677852348993288e-05,
|
58 |
+
"loss": 0.0645,
|
59 |
+
"step": 700
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 18.18,
|
63 |
+
"learning_rate": 7.624161073825503e-05,
|
64 |
+
"loss": 0.0481,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 20.45,
|
69 |
+
"learning_rate": 7.570469798657718e-05,
|
70 |
+
"loss": 0.0386,
|
71 |
+
"step": 900
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 22.73,
|
75 |
+
"learning_rate": 7.516778523489933e-05,
|
76 |
+
"loss": 0.0322,
|
77 |
+
"step": 1000
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 22.73,
|
81 |
+
"eval_accuracy": 0.8473163105784263,
|
82 |
+
"eval_loss": 0.9984953999519348,
|
83 |
+
"eval_runtime": 3.1616,
|
84 |
+
"eval_samples_per_second": 253.035,
|
85 |
+
"eval_steps_per_second": 31.629,
|
86 |
+
"step": 1000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 25.0,
|
90 |
+
"learning_rate": 7.463087248322148e-05,
|
91 |
+
"loss": 0.0243,
|
92 |
+
"step": 1100
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 27.27,
|
96 |
+
"learning_rate": 7.409395973154362e-05,
|
97 |
+
"loss": 0.0221,
|
98 |
+
"step": 1200
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 29.55,
|
102 |
+
"learning_rate": 7.355704697986577e-05,
|
103 |
+
"loss": 0.0192,
|
104 |
+
"step": 1300
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 31.82,
|
108 |
+
"learning_rate": 7.302013422818792e-05,
|
109 |
+
"loss": 0.0158,
|
110 |
+
"step": 1400
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 34.09,
|
114 |
+
"learning_rate": 7.248322147651007e-05,
|
115 |
+
"loss": 0.0182,
|
116 |
+
"step": 1500
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 34.09,
|
120 |
+
"eval_accuracy": 0.8484453708528747,
|
121 |
+
"eval_loss": 1.106465220451355,
|
122 |
+
"eval_runtime": 3.1709,
|
123 |
+
"eval_samples_per_second": 252.294,
|
124 |
+
"eval_steps_per_second": 31.537,
|
125 |
+
"step": 1500
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 36.36,
|
129 |
+
"learning_rate": 7.194630872483222e-05,
|
130 |
+
"loss": 0.0173,
|
131 |
+
"step": 1600
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 38.64,
|
135 |
+
"learning_rate": 7.140939597315438e-05,
|
136 |
+
"loss": 0.0136,
|
137 |
+
"step": 1700
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 40.91,
|
141 |
+
"learning_rate": 7.087248322147653e-05,
|
142 |
+
"loss": 0.0116,
|
143 |
+
"step": 1800
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 43.18,
|
147 |
+
"learning_rate": 7.033557046979866e-05,
|
148 |
+
"loss": 0.01,
|
149 |
+
"step": 1900
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 45.45,
|
153 |
+
"learning_rate": 6.979865771812081e-05,
|
154 |
+
"loss": 0.0112,
|
155 |
+
"step": 2000
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 45.45,
|
159 |
+
"eval_accuracy": 0.8446239360778183,
|
160 |
+
"eval_loss": 1.2097970247268677,
|
161 |
+
"eval_runtime": 3.165,
|
162 |
+
"eval_samples_per_second": 252.768,
|
163 |
+
"eval_steps_per_second": 31.596,
|
164 |
+
"step": 2000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 47.73,
|
168 |
+
"learning_rate": 6.926174496644296e-05,
|
169 |
+
"loss": 0.009,
|
170 |
+
"step": 2100
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 50.0,
|
174 |
+
"learning_rate": 6.87248322147651e-05,
|
175 |
+
"loss": 0.0103,
|
176 |
+
"step": 2200
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 52.27,
|
180 |
+
"learning_rate": 6.818791946308725e-05,
|
181 |
+
"loss": 0.0096,
|
182 |
+
"step": 2300
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 54.55,
|
186 |
+
"learning_rate": 6.76510067114094e-05,
|
187 |
+
"loss": 0.008,
|
188 |
+
"step": 2400
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 56.82,
|
192 |
+
"learning_rate": 6.711409395973155e-05,
|
193 |
+
"loss": 0.008,
|
194 |
+
"step": 2500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 56.82,
|
198 |
+
"eval_accuracy": 0.8536564182734063,
|
199 |
+
"eval_loss": 1.282847285270691,
|
200 |
+
"eval_runtime": 3.1784,
|
201 |
+
"eval_samples_per_second": 251.698,
|
202 |
+
"eval_steps_per_second": 31.462,
|
203 |
+
"step": 2500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 59.09,
|
207 |
+
"learning_rate": 6.65771812080537e-05,
|
208 |
+
"loss": 0.0073,
|
209 |
+
"step": 2600
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 61.36,
|
213 |
+
"learning_rate": 6.604026845637585e-05,
|
214 |
+
"loss": 0.008,
|
215 |
+
"step": 2700
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 63.64,
|
219 |
+
"learning_rate": 6.5503355704698e-05,
|
220 |
+
"loss": 0.0076,
|
221 |
+
"step": 2800
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 65.91,
|
225 |
+
"learning_rate": 6.496644295302014e-05,
|
226 |
+
"loss": 0.0072,
|
227 |
+
"step": 2900
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 68.18,
|
231 |
+
"learning_rate": 6.442953020134228e-05,
|
232 |
+
"loss": 0.0064,
|
233 |
+
"step": 3000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 68.18,
|
237 |
+
"eval_accuracy": 0.8514851485148515,
|
238 |
+
"eval_loss": 1.265517234802246,
|
239 |
+
"eval_runtime": 3.1864,
|
240 |
+
"eval_samples_per_second": 251.068,
|
241 |
+
"eval_steps_per_second": 31.384,
|
242 |
+
"step": 3000
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 70.45,
|
246 |
+
"learning_rate": 6.389261744966443e-05,
|
247 |
+
"loss": 0.0054,
|
248 |
+
"step": 3100
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 72.73,
|
252 |
+
"learning_rate": 6.335570469798657e-05,
|
253 |
+
"loss": 0.0047,
|
254 |
+
"step": 3200
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 75.0,
|
258 |
+
"learning_rate": 6.281879194630872e-05,
|
259 |
+
"loss": 0.0064,
|
260 |
+
"step": 3300
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 77.27,
|
264 |
+
"learning_rate": 6.228187919463087e-05,
|
265 |
+
"loss": 0.0057,
|
266 |
+
"step": 3400
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 79.55,
|
270 |
+
"learning_rate": 6.174496644295302e-05,
|
271 |
+
"loss": 0.0064,
|
272 |
+
"step": 3500
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 79.55,
|
276 |
+
"eval_accuracy": 0.8527879103699844,
|
277 |
+
"eval_loss": 1.328222393989563,
|
278 |
+
"eval_runtime": 3.1799,
|
279 |
+
"eval_samples_per_second": 251.578,
|
280 |
+
"eval_steps_per_second": 31.447,
|
281 |
+
"step": 3500
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 81.82,
|
285 |
+
"learning_rate": 6.120805369127517e-05,
|
286 |
+
"loss": 0.0055,
|
287 |
+
"step": 3600
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 84.09,
|
291 |
+
"learning_rate": 6.067114093959732e-05,
|
292 |
+
"loss": 0.0056,
|
293 |
+
"step": 3700
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 86.36,
|
297 |
+
"learning_rate": 6.013422818791947e-05,
|
298 |
+
"loss": 0.0049,
|
299 |
+
"step": 3800
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 88.64,
|
303 |
+
"learning_rate": 5.959731543624162e-05,
|
304 |
+
"loss": 0.0035,
|
305 |
+
"step": 3900
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 90.91,
|
309 |
+
"learning_rate": 5.906040268456377e-05,
|
310 |
+
"loss": 0.0045,
|
311 |
+
"step": 4000
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 90.91,
|
315 |
+
"eval_accuracy": 0.8500955358693764,
|
316 |
+
"eval_loss": 1.3510583639144897,
|
317 |
+
"eval_runtime": 3.1875,
|
318 |
+
"eval_samples_per_second": 250.981,
|
319 |
+
"eval_steps_per_second": 31.373,
|
320 |
+
"step": 4000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 93.18,
|
324 |
+
"learning_rate": 5.8523489932885916e-05,
|
325 |
+
"loss": 0.0038,
|
326 |
+
"step": 4100
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 95.45,
|
330 |
+
"learning_rate": 5.798657718120806e-05,
|
331 |
+
"loss": 0.0033,
|
332 |
+
"step": 4200
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 97.73,
|
336 |
+
"learning_rate": 5.7449664429530206e-05,
|
337 |
+
"loss": 0.004,
|
338 |
+
"step": 4300
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 100.0,
|
342 |
+
"learning_rate": 5.6912751677852354e-05,
|
343 |
+
"loss": 0.0042,
|
344 |
+
"step": 4400
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 102.27,
|
348 |
+
"learning_rate": 5.63758389261745e-05,
|
349 |
+
"loss": 0.0047,
|
350 |
+
"step": 4500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 102.27,
|
354 |
+
"eval_accuracy": 0.8551328817092235,
|
355 |
+
"eval_loss": 1.3701601028442383,
|
356 |
+
"eval_runtime": 3.1715,
|
357 |
+
"eval_samples_per_second": 252.247,
|
358 |
+
"eval_steps_per_second": 31.531,
|
359 |
+
"step": 4500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 104.55,
|
363 |
+
"learning_rate": 5.583892617449665e-05,
|
364 |
+
"loss": 0.004,
|
365 |
+
"step": 4600
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 106.82,
|
369 |
+
"learning_rate": 5.53020134228188e-05,
|
370 |
+
"loss": 0.0031,
|
371 |
+
"step": 4700
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 109.09,
|
375 |
+
"learning_rate": 5.476510067114095e-05,
|
376 |
+
"loss": 0.0047,
|
377 |
+
"step": 4800
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 111.36,
|
381 |
+
"learning_rate": 5.4228187919463095e-05,
|
382 |
+
"loss": 0.0028,
|
383 |
+
"step": 4900
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 113.64,
|
387 |
+
"learning_rate": 5.369127516778524e-05,
|
388 |
+
"loss": 0.0039,
|
389 |
+
"step": 5000
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 113.64,
|
393 |
+
"eval_accuracy": 0.8559145388223033,
|
394 |
+
"eval_loss": 1.4116061925888062,
|
395 |
+
"eval_runtime": 3.1845,
|
396 |
+
"eval_samples_per_second": 251.218,
|
397 |
+
"eval_steps_per_second": 31.402,
|
398 |
+
"step": 5000
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 115.91,
|
402 |
+
"learning_rate": 5.315436241610739e-05,
|
403 |
+
"loss": 0.004,
|
404 |
+
"step": 5100
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 118.18,
|
408 |
+
"learning_rate": 5.261744966442954e-05,
|
409 |
+
"loss": 0.0034,
|
410 |
+
"step": 5200
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 120.45,
|
414 |
+
"learning_rate": 5.208053691275168e-05,
|
415 |
+
"loss": 0.0027,
|
416 |
+
"step": 5300
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 122.73,
|
420 |
+
"learning_rate": 5.154362416107383e-05,
|
421 |
+
"loss": 0.0023,
|
422 |
+
"step": 5400
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 125.0,
|
426 |
+
"learning_rate": 5.100671140939598e-05,
|
427 |
+
"loss": 0.0029,
|
428 |
+
"step": 5500
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 125.0,
|
432 |
+
"eval_accuracy": 0.8567830467257253,
|
433 |
+
"eval_loss": 1.434157371520996,
|
434 |
+
"eval_runtime": 3.1462,
|
435 |
+
"eval_samples_per_second": 254.277,
|
436 |
+
"eval_steps_per_second": 31.785,
|
437 |
+
"step": 5500
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 127.27,
|
441 |
+
"learning_rate": 5.0469798657718126e-05,
|
442 |
+
"loss": 0.0027,
|
443 |
+
"step": 5600
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 129.55,
|
447 |
+
"learning_rate": 4.9932885906040274e-05,
|
448 |
+
"loss": 0.0033,
|
449 |
+
"step": 5700
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 131.82,
|
453 |
+
"learning_rate": 4.939597315436242e-05,
|
454 |
+
"loss": 0.0029,
|
455 |
+
"step": 5800
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 134.09,
|
459 |
+
"learning_rate": 4.885906040268457e-05,
|
460 |
+
"loss": 0.0023,
|
461 |
+
"step": 5900
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 136.36,
|
465 |
+
"learning_rate": 4.832214765100672e-05,
|
466 |
+
"loss": 0.0033,
|
467 |
+
"step": 6000
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 136.36,
|
471 |
+
"eval_accuracy": 0.854003821434775,
|
472 |
+
"eval_loss": 1.4268622398376465,
|
473 |
+
"eval_runtime": 3.1765,
|
474 |
+
"eval_samples_per_second": 251.849,
|
475 |
+
"eval_steps_per_second": 31.481,
|
476 |
+
"step": 6000
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 138.64,
|
480 |
+
"learning_rate": 4.779060402684564e-05,
|
481 |
+
"loss": 0.0031,
|
482 |
+
"step": 6100
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 140.91,
|
486 |
+
"learning_rate": 4.725369127516779e-05,
|
487 |
+
"loss": 0.0031,
|
488 |
+
"step": 6200
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 143.18,
|
492 |
+
"learning_rate": 4.6716778523489936e-05,
|
493 |
+
"loss": 0.0031,
|
494 |
+
"step": 6300
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 145.45,
|
498 |
+
"learning_rate": 4.6179865771812084e-05,
|
499 |
+
"loss": 0.0024,
|
500 |
+
"step": 6400
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 147.73,
|
504 |
+
"learning_rate": 4.564295302013423e-05,
|
505 |
+
"loss": 0.0033,
|
506 |
+
"step": 6500
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 147.73,
|
510 |
+
"eval_accuracy": 0.853048462741011,
|
511 |
+
"eval_loss": 1.4698809385299683,
|
512 |
+
"eval_runtime": 3.1835,
|
513 |
+
"eval_samples_per_second": 251.299,
|
514 |
+
"eval_steps_per_second": 31.412,
|
515 |
+
"step": 6500
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 150.0,
|
519 |
+
"learning_rate": 4.510604026845638e-05,
|
520 |
+
"loss": 0.0023,
|
521 |
+
"step": 6600
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 152.27,
|
525 |
+
"learning_rate": 4.456912751677853e-05,
|
526 |
+
"loss": 0.0022,
|
527 |
+
"step": 6700
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 154.55,
|
531 |
+
"learning_rate": 4.403221476510068e-05,
|
532 |
+
"loss": 0.0026,
|
533 |
+
"step": 6800
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 156.82,
|
537 |
+
"learning_rate": 4.3495302013422825e-05,
|
538 |
+
"loss": 0.002,
|
539 |
+
"step": 6900
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 159.09,
|
543 |
+
"learning_rate": 4.295838926174497e-05,
|
544 |
+
"loss": 0.0019,
|
545 |
+
"step": 7000
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 159.09,
|
549 |
+
"eval_accuracy": 0.8554802848705924,
|
550 |
+
"eval_loss": 1.4916952848434448,
|
551 |
+
"eval_runtime": 3.1771,
|
552 |
+
"eval_samples_per_second": 251.799,
|
553 |
+
"eval_steps_per_second": 31.475,
|
554 |
+
"step": 7000
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 161.36,
|
558 |
+
"learning_rate": 4.242147651006712e-05,
|
559 |
+
"loss": 0.0032,
|
560 |
+
"step": 7100
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 163.64,
|
564 |
+
"learning_rate": 4.188456375838927e-05,
|
565 |
+
"loss": 0.002,
|
566 |
+
"step": 7200
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 165.91,
|
570 |
+
"learning_rate": 4.134765100671141e-05,
|
571 |
+
"loss": 0.0022,
|
572 |
+
"step": 7300
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 168.18,
|
576 |
+
"learning_rate": 4.081073825503356e-05,
|
577 |
+
"loss": 0.0016,
|
578 |
+
"step": 7400
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 170.45,
|
582 |
+
"learning_rate": 4.027382550335571e-05,
|
583 |
+
"loss": 0.0018,
|
584 |
+
"step": 7500
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 170.45,
|
588 |
+
"eval_accuracy": 0.854872329338197,
|
589 |
+
"eval_loss": 1.4624574184417725,
|
590 |
+
"eval_runtime": 3.1705,
|
591 |
+
"eval_samples_per_second": 252.327,
|
592 |
+
"eval_steps_per_second": 31.541,
|
593 |
+
"step": 7500
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 172.73,
|
597 |
+
"learning_rate": 3.9736912751677856e-05,
|
598 |
+
"loss": 0.0017,
|
599 |
+
"step": 7600
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 175.0,
|
603 |
+
"learning_rate": 3.9200000000000004e-05,
|
604 |
+
"loss": 0.0014,
|
605 |
+
"step": 7700
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 177.27,
|
609 |
+
"learning_rate": 3.866308724832215e-05,
|
610 |
+
"loss": 0.0009,
|
611 |
+
"step": 7800
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 179.55,
|
615 |
+
"learning_rate": 3.81261744966443e-05,
|
616 |
+
"loss": 0.0014,
|
617 |
+
"step": 7900
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 181.82,
|
621 |
+
"learning_rate": 3.758926174496645e-05,
|
622 |
+
"loss": 0.001,
|
623 |
+
"step": 8000
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 181.82,
|
627 |
+
"eval_accuracy": 0.8567830467257253,
|
628 |
+
"eval_loss": 1.5747320652008057,
|
629 |
+
"eval_runtime": 3.1536,
|
630 |
+
"eval_samples_per_second": 253.674,
|
631 |
+
"eval_steps_per_second": 31.709,
|
632 |
+
"step": 8000
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 181.82,
|
636 |
+
"step": 8000,
|
637 |
+
"total_flos": 4.180403182626816e+16,
|
638 |
+
"train_loss": 0.04628073706757277,
|
639 |
+
"train_runtime": 1159.4207,
|
640 |
+
"train_samples_per_second": 414.0,
|
641 |
+
"train_steps_per_second": 12.937
|
642 |
+
}
|
643 |
+
],
|
644 |
+
"max_steps": 15000,
|
645 |
+
"num_train_epochs": 341,
|
646 |
+
"total_flos": 4.180403182626816e+16,
|
647 |
+
"trial_name": null,
|
648 |
+
"trial_params": null
|
649 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6f1df1695bf4ed5c88dfb74b2d1449d1a21de5f5d1fd7364b1d8cf0b90d04361
|
3 |
+
size 3387
|