alen commited on
Initial commit
Browse files- checkpoint-522/config.json +61 -0
- checkpoint-522/generation_config.json +240 -0
- checkpoint-522/model.safetensors +3 -0
- checkpoint-522/optimizer.pt +3 -0
- checkpoint-522/preprocessor_config.json +14 -0
- checkpoint-522/rng_state.pth +3 -0
- checkpoint-522/scheduler.pt +3 -0
- checkpoint-522/trainer_state.json +173 -0
- checkpoint-522/training_args.bin +3 -0
- config.json +61 -0
- model.safetensors +3 -0
- preprocessor_config.json +14 -0
- training_args.bin +3 -0
checkpoint-522/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"apply_spec_augment": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"WhisperForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"begin_suppress_tokens": null,
|
| 11 |
+
"bos_token_id": 50257,
|
| 12 |
+
"classifier_proj_size": 256,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 24,
|
| 18 |
+
"decoder_start_token_id": 50258,
|
| 19 |
+
"dropout": 0.0,
|
| 20 |
+
"encoder_attention_heads": 16,
|
| 21 |
+
"encoder_ffn_dim": 4096,
|
| 22 |
+
"encoder_layerdrop": 0.0,
|
| 23 |
+
"encoder_layers": 24,
|
| 24 |
+
"eos_token_id": 50257,
|
| 25 |
+
"forced_decoder_ids": [
|
| 26 |
+
[
|
| 27 |
+
1,
|
| 28 |
+
50259
|
| 29 |
+
],
|
| 30 |
+
[
|
| 31 |
+
2,
|
| 32 |
+
50359
|
| 33 |
+
],
|
| 34 |
+
[
|
| 35 |
+
3,
|
| 36 |
+
50363
|
| 37 |
+
]
|
| 38 |
+
],
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"mask_feature_length": 10,
|
| 42 |
+
"mask_feature_min_masks": 0,
|
| 43 |
+
"mask_feature_prob": 0.0,
|
| 44 |
+
"mask_time_length": 10,
|
| 45 |
+
"mask_time_min_masks": 2,
|
| 46 |
+
"mask_time_prob": 0.05,
|
| 47 |
+
"max_length": null,
|
| 48 |
+
"max_source_positions": 1500,
|
| 49 |
+
"max_target_positions": 448,
|
| 50 |
+
"median_filter_width": 7,
|
| 51 |
+
"model_type": "whisper",
|
| 52 |
+
"num_hidden_layers": 24,
|
| 53 |
+
"num_mel_bins": 80,
|
| 54 |
+
"pad_token_id": 50257,
|
| 55 |
+
"scale_embedding": false,
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.48.0.dev0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"use_weighted_layer_sum": false,
|
| 60 |
+
"vocab_size": 51865
|
| 61 |
+
}
|
checkpoint-522/generation_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alignment_heads": [
|
| 3 |
+
[
|
| 4 |
+
13,
|
| 5 |
+
15
|
| 6 |
+
],
|
| 7 |
+
[
|
| 8 |
+
15,
|
| 9 |
+
4
|
| 10 |
+
],
|
| 11 |
+
[
|
| 12 |
+
15,
|
| 13 |
+
15
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
16,
|
| 17 |
+
1
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
20,
|
| 21 |
+
0
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
23,
|
| 25 |
+
4
|
| 26 |
+
]
|
| 27 |
+
],
|
| 28 |
+
"begin_suppress_tokens": [
|
| 29 |
+
220,
|
| 30 |
+
50257
|
| 31 |
+
],
|
| 32 |
+
"bos_token_id": 50257,
|
| 33 |
+
"decoder_start_token_id": 50258,
|
| 34 |
+
"eos_token_id": 50257,
|
| 35 |
+
"is_multilingual": true,
|
| 36 |
+
"lang_to_id": {
|
| 37 |
+
"<|af|>": 50327,
|
| 38 |
+
"<|am|>": 50334,
|
| 39 |
+
"<|ar|>": 50272,
|
| 40 |
+
"<|as|>": 50350,
|
| 41 |
+
"<|az|>": 50304,
|
| 42 |
+
"<|ba|>": 50355,
|
| 43 |
+
"<|be|>": 50330,
|
| 44 |
+
"<|bg|>": 50292,
|
| 45 |
+
"<|bn|>": 50302,
|
| 46 |
+
"<|bo|>": 50347,
|
| 47 |
+
"<|br|>": 50309,
|
| 48 |
+
"<|bs|>": 50315,
|
| 49 |
+
"<|ca|>": 50270,
|
| 50 |
+
"<|cs|>": 50283,
|
| 51 |
+
"<|cy|>": 50297,
|
| 52 |
+
"<|da|>": 50285,
|
| 53 |
+
"<|de|>": 50261,
|
| 54 |
+
"<|el|>": 50281,
|
| 55 |
+
"<|en|>": 50259,
|
| 56 |
+
"<|es|>": 50262,
|
| 57 |
+
"<|et|>": 50307,
|
| 58 |
+
"<|eu|>": 50310,
|
| 59 |
+
"<|fa|>": 50300,
|
| 60 |
+
"<|fi|>": 50277,
|
| 61 |
+
"<|fo|>": 50338,
|
| 62 |
+
"<|fr|>": 50265,
|
| 63 |
+
"<|gl|>": 50319,
|
| 64 |
+
"<|gu|>": 50333,
|
| 65 |
+
"<|haw|>": 50352,
|
| 66 |
+
"<|ha|>": 50354,
|
| 67 |
+
"<|he|>": 50279,
|
| 68 |
+
"<|hi|>": 50276,
|
| 69 |
+
"<|hr|>": 50291,
|
| 70 |
+
"<|ht|>": 50339,
|
| 71 |
+
"<|hu|>": 50286,
|
| 72 |
+
"<|hy|>": 50312,
|
| 73 |
+
"<|id|>": 50275,
|
| 74 |
+
"<|is|>": 50311,
|
| 75 |
+
"<|it|>": 50274,
|
| 76 |
+
"<|ja|>": 50266,
|
| 77 |
+
"<|jw|>": 50356,
|
| 78 |
+
"<|ka|>": 50329,
|
| 79 |
+
"<|kk|>": 50316,
|
| 80 |
+
"<|km|>": 50323,
|
| 81 |
+
"<|kn|>": 50306,
|
| 82 |
+
"<|ko|>": 50264,
|
| 83 |
+
"<|la|>": 50294,
|
| 84 |
+
"<|lb|>": 50345,
|
| 85 |
+
"<|ln|>": 50353,
|
| 86 |
+
"<|lo|>": 50336,
|
| 87 |
+
"<|lt|>": 50293,
|
| 88 |
+
"<|lv|>": 50301,
|
| 89 |
+
"<|mg|>": 50349,
|
| 90 |
+
"<|mi|>": 50295,
|
| 91 |
+
"<|mk|>": 50308,
|
| 92 |
+
"<|ml|>": 50296,
|
| 93 |
+
"<|mn|>": 50314,
|
| 94 |
+
"<|mr|>": 50320,
|
| 95 |
+
"<|ms|>": 50282,
|
| 96 |
+
"<|mt|>": 50343,
|
| 97 |
+
"<|my|>": 50346,
|
| 98 |
+
"<|ne|>": 50313,
|
| 99 |
+
"<|nl|>": 50271,
|
| 100 |
+
"<|nn|>": 50342,
|
| 101 |
+
"<|no|>": 50288,
|
| 102 |
+
"<|oc|>": 50328,
|
| 103 |
+
"<|pa|>": 50321,
|
| 104 |
+
"<|pl|>": 50269,
|
| 105 |
+
"<|ps|>": 50340,
|
| 106 |
+
"<|pt|>": 50267,
|
| 107 |
+
"<|ro|>": 50284,
|
| 108 |
+
"<|ru|>": 50263,
|
| 109 |
+
"<|sa|>": 50344,
|
| 110 |
+
"<|sd|>": 50332,
|
| 111 |
+
"<|si|>": 50322,
|
| 112 |
+
"<|sk|>": 50298,
|
| 113 |
+
"<|sl|>": 50305,
|
| 114 |
+
"<|sn|>": 50324,
|
| 115 |
+
"<|so|>": 50326,
|
| 116 |
+
"<|sq|>": 50317,
|
| 117 |
+
"<|sr|>": 50303,
|
| 118 |
+
"<|su|>": 50357,
|
| 119 |
+
"<|sv|>": 50273,
|
| 120 |
+
"<|sw|>": 50318,
|
| 121 |
+
"<|ta|>": 50287,
|
| 122 |
+
"<|te|>": 50299,
|
| 123 |
+
"<|tg|>": 50331,
|
| 124 |
+
"<|th|>": 50289,
|
| 125 |
+
"<|tk|>": 50341,
|
| 126 |
+
"<|tl|>": 50348,
|
| 127 |
+
"<|tr|>": 50268,
|
| 128 |
+
"<|tt|>": 50351,
|
| 129 |
+
"<|uk|>": 50280,
|
| 130 |
+
"<|ur|>": 50290,
|
| 131 |
+
"<|uz|>": 50337,
|
| 132 |
+
"<|vi|>": 50278,
|
| 133 |
+
"<|yi|>": 50335,
|
| 134 |
+
"<|yo|>": 50325,
|
| 135 |
+
"<|zh|>": 50260
|
| 136 |
+
},
|
| 137 |
+
"language": "vi",
|
| 138 |
+
"max_initial_timestamp_index": 50,
|
| 139 |
+
"max_length": 448,
|
| 140 |
+
"no_timestamps_token_id": 50363,
|
| 141 |
+
"pad_token_id": 50257,
|
| 142 |
+
"prev_sot_token_id": 50361,
|
| 143 |
+
"return_timestamps": true,
|
| 144 |
+
"suppress_tokens": [
|
| 145 |
+
1,
|
| 146 |
+
2,
|
| 147 |
+
7,
|
| 148 |
+
8,
|
| 149 |
+
9,
|
| 150 |
+
10,
|
| 151 |
+
14,
|
| 152 |
+
25,
|
| 153 |
+
26,
|
| 154 |
+
27,
|
| 155 |
+
28,
|
| 156 |
+
29,
|
| 157 |
+
31,
|
| 158 |
+
58,
|
| 159 |
+
59,
|
| 160 |
+
60,
|
| 161 |
+
61,
|
| 162 |
+
62,
|
| 163 |
+
63,
|
| 164 |
+
90,
|
| 165 |
+
91,
|
| 166 |
+
92,
|
| 167 |
+
93,
|
| 168 |
+
359,
|
| 169 |
+
503,
|
| 170 |
+
522,
|
| 171 |
+
542,
|
| 172 |
+
873,
|
| 173 |
+
893,
|
| 174 |
+
902,
|
| 175 |
+
918,
|
| 176 |
+
922,
|
| 177 |
+
931,
|
| 178 |
+
1350,
|
| 179 |
+
1853,
|
| 180 |
+
1982,
|
| 181 |
+
2460,
|
| 182 |
+
2627,
|
| 183 |
+
3246,
|
| 184 |
+
3253,
|
| 185 |
+
3268,
|
| 186 |
+
3536,
|
| 187 |
+
3846,
|
| 188 |
+
3961,
|
| 189 |
+
4183,
|
| 190 |
+
4667,
|
| 191 |
+
6585,
|
| 192 |
+
6647,
|
| 193 |
+
7273,
|
| 194 |
+
9061,
|
| 195 |
+
9383,
|
| 196 |
+
10428,
|
| 197 |
+
10929,
|
| 198 |
+
11938,
|
| 199 |
+
12033,
|
| 200 |
+
12331,
|
| 201 |
+
12562,
|
| 202 |
+
13793,
|
| 203 |
+
14157,
|
| 204 |
+
14635,
|
| 205 |
+
15265,
|
| 206 |
+
15618,
|
| 207 |
+
16553,
|
| 208 |
+
16604,
|
| 209 |
+
18362,
|
| 210 |
+
18956,
|
| 211 |
+
20075,
|
| 212 |
+
21675,
|
| 213 |
+
22520,
|
| 214 |
+
26130,
|
| 215 |
+
26161,
|
| 216 |
+
26435,
|
| 217 |
+
28279,
|
| 218 |
+
29464,
|
| 219 |
+
31650,
|
| 220 |
+
32302,
|
| 221 |
+
32470,
|
| 222 |
+
36865,
|
| 223 |
+
42863,
|
| 224 |
+
47425,
|
| 225 |
+
49870,
|
| 226 |
+
50254,
|
| 227 |
+
50258,
|
| 228 |
+
50358,
|
| 229 |
+
50359,
|
| 230 |
+
50360,
|
| 231 |
+
50361,
|
| 232 |
+
50362
|
| 233 |
+
],
|
| 234 |
+
"task": "transcribe",
|
| 235 |
+
"task_to_id": {
|
| 236 |
+
"transcribe": 50359,
|
| 237 |
+
"translate": 50358
|
| 238 |
+
},
|
| 239 |
+
"transformers_version": "4.48.0.dev0"
|
| 240 |
+
}
|
checkpoint-522/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:454814e215d69c26398b5553e1d6d418787af98fedfaf313c3d7f4d485d2b514
|
| 3 |
+
size 3055544304
|
checkpoint-522/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c9f7b5110d2056a3b16635460c54f79e77bd96bd43b039765cf82c090f184f3
|
| 3 |
+
size 6099375168
|
checkpoint-522/preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 4 |
+
"feature_size": 80,
|
| 5 |
+
"hop_length": 160,
|
| 6 |
+
"n_fft": 400,
|
| 7 |
+
"n_samples": 480000,
|
| 8 |
+
"nb_max_frames": 3000,
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"padding_value": 0.0,
|
| 11 |
+
"processor_class": "WhisperProcessor",
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
checkpoint-522/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97a92c080b361020f6586b63a2b355444949392da4fcf0a32a71639585ffb480
|
| 3 |
+
size 14244
|
checkpoint-522/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cd2a2eba5b08abdeae10bde55bfe1b715ea99f6aabce35857abbb48d7514033
|
| 3 |
+
size 1064
|
checkpoint-522/trainer_state.json
ADDED
|
@@ -0,0 +1,173 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 522,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14367816091954022,
|
| 13 |
+
"grad_norm": 20.954299926757812,
|
| 14 |
+
"learning_rate": 3.8e-07,
|
| 15 |
+
"loss": 3.1782,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.28735632183908044,
|
| 20 |
+
"grad_norm": 15.136266708374023,
|
| 21 |
+
"learning_rate": 8.8e-07,
|
| 22 |
+
"loss": 2.6873,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.43103448275862066,
|
| 27 |
+
"grad_norm": 11.3860502243042,
|
| 28 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 29 |
+
"loss": 1.9105,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.5747126436781609,
|
| 34 |
+
"grad_norm": 10.273317337036133,
|
| 35 |
+
"learning_rate": 1.8800000000000002e-06,
|
| 36 |
+
"loss": 1.3364,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.7183908045977011,
|
| 41 |
+
"grad_norm": 6.59112024307251,
|
| 42 |
+
"learning_rate": 2.38e-06,
|
| 43 |
+
"loss": 0.6904,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.8620689655172413,
|
| 48 |
+
"grad_norm": 7.895627021789551,
|
| 49 |
+
"learning_rate": 2.88e-06,
|
| 50 |
+
"loss": 0.2129,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.0057471264367817,
|
| 55 |
+
"grad_norm": 5.726414680480957,
|
| 56 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 57 |
+
"loss": 0.199,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.1494252873563218,
|
| 62 |
+
"grad_norm": 4.383477687835693,
|
| 63 |
+
"learning_rate": 3.88e-06,
|
| 64 |
+
"loss": 0.1546,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.293103448275862,
|
| 69 |
+
"grad_norm": 6.100410461425781,
|
| 70 |
+
"learning_rate": 4.38e-06,
|
| 71 |
+
"loss": 0.1425,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.4367816091954024,
|
| 76 |
+
"grad_norm": 6.457783222198486,
|
| 77 |
+
"learning_rate": 4.880000000000001e-06,
|
| 78 |
+
"loss": 0.149,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.5804597701149425,
|
| 83 |
+
"grad_norm": 6.307376384735107,
|
| 84 |
+
"learning_rate": 5.380000000000001e-06,
|
| 85 |
+
"loss": 0.162,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.7241379310344827,
|
| 90 |
+
"grad_norm": 3.9684667587280273,
|
| 91 |
+
"learning_rate": 5.8800000000000005e-06,
|
| 92 |
+
"loss": 0.1566,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.867816091954023,
|
| 97 |
+
"grad_norm": 5.163708686828613,
|
| 98 |
+
"learning_rate": 6.380000000000001e-06,
|
| 99 |
+
"loss": 0.1594,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.0114942528735633,
|
| 104 |
+
"grad_norm": 4.452727794647217,
|
| 105 |
+
"learning_rate": 6.88e-06,
|
| 106 |
+
"loss": 0.1557,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.1551724137931036,
|
| 111 |
+
"grad_norm": 3.3390920162200928,
|
| 112 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 113 |
+
"loss": 0.0866,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.2988505747126435,
|
| 118 |
+
"grad_norm": 4.967168807983398,
|
| 119 |
+
"learning_rate": 7.88e-06,
|
| 120 |
+
"loss": 0.0854,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.442528735632184,
|
| 125 |
+
"grad_norm": 4.6858439445495605,
|
| 126 |
+
"learning_rate": 8.380000000000001e-06,
|
| 127 |
+
"loss": 0.1005,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.586206896551724,
|
| 132 |
+
"grad_norm": 4.143185138702393,
|
| 133 |
+
"learning_rate": 8.880000000000001e-06,
|
| 134 |
+
"loss": 0.0865,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.7298850574712645,
|
| 139 |
+
"grad_norm": 4.719691753387451,
|
| 140 |
+
"learning_rate": 9.38e-06,
|
| 141 |
+
"loss": 0.1025,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.873563218390805,
|
| 146 |
+
"grad_norm": 8.654423713684082,
|
| 147 |
+
"learning_rate": 9.88e-06,
|
| 148 |
+
"loss": 0.1032,
|
| 149 |
+
"step": 500
|
| 150 |
+
}
|
| 151 |
+
],
|
| 152 |
+
"logging_steps": 25,
|
| 153 |
+
"max_steps": 522,
|
| 154 |
+
"num_input_tokens_seen": 0,
|
| 155 |
+
"num_train_epochs": 3,
|
| 156 |
+
"save_steps": 1000,
|
| 157 |
+
"stateful_callbacks": {
|
| 158 |
+
"TrainerControl": {
|
| 159 |
+
"args": {
|
| 160 |
+
"should_epoch_stop": false,
|
| 161 |
+
"should_evaluate": false,
|
| 162 |
+
"should_log": false,
|
| 163 |
+
"should_save": true,
|
| 164 |
+
"should_training_stop": true
|
| 165 |
+
},
|
| 166 |
+
"attributes": {}
|
| 167 |
+
}
|
| 168 |
+
},
|
| 169 |
+
"total_flos": 8.49041222565888e+18,
|
| 170 |
+
"train_batch_size": 16,
|
| 171 |
+
"trial_name": null,
|
| 172 |
+
"trial_params": null
|
| 173 |
+
}
|
checkpoint-522/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10c582f8f67285309fe103d1e5ed042f60eca0383067610987b83ac2cac6c292
|
| 3 |
+
size 5496
|
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"apply_spec_augment": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"WhisperForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"begin_suppress_tokens": null,
|
| 11 |
+
"bos_token_id": 50257,
|
| 12 |
+
"classifier_proj_size": 256,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 24,
|
| 18 |
+
"decoder_start_token_id": 50258,
|
| 19 |
+
"dropout": 0.0,
|
| 20 |
+
"encoder_attention_heads": 16,
|
| 21 |
+
"encoder_ffn_dim": 4096,
|
| 22 |
+
"encoder_layerdrop": 0.0,
|
| 23 |
+
"encoder_layers": 24,
|
| 24 |
+
"eos_token_id": 50257,
|
| 25 |
+
"forced_decoder_ids": [
|
| 26 |
+
[
|
| 27 |
+
1,
|
| 28 |
+
50259
|
| 29 |
+
],
|
| 30 |
+
[
|
| 31 |
+
2,
|
| 32 |
+
50359
|
| 33 |
+
],
|
| 34 |
+
[
|
| 35 |
+
3,
|
| 36 |
+
50363
|
| 37 |
+
]
|
| 38 |
+
],
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"mask_feature_length": 10,
|
| 42 |
+
"mask_feature_min_masks": 0,
|
| 43 |
+
"mask_feature_prob": 0.0,
|
| 44 |
+
"mask_time_length": 10,
|
| 45 |
+
"mask_time_min_masks": 2,
|
| 46 |
+
"mask_time_prob": 0.05,
|
| 47 |
+
"max_length": null,
|
| 48 |
+
"max_source_positions": 1500,
|
| 49 |
+
"max_target_positions": 448,
|
| 50 |
+
"median_filter_width": 7,
|
| 51 |
+
"model_type": "whisper",
|
| 52 |
+
"num_hidden_layers": 24,
|
| 53 |
+
"num_mel_bins": 80,
|
| 54 |
+
"pad_token_id": 50257,
|
| 55 |
+
"scale_embedding": false,
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.48.0.dev0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"use_weighted_layer_sum": false,
|
| 60 |
+
"vocab_size": 51865
|
| 61 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:454814e215d69c26398b5553e1d6d418787af98fedfaf313c3d7f4d485d2b514
|
| 3 |
+
size 3055544304
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 4 |
+
"feature_size": 80,
|
| 5 |
+
"hop_length": 160,
|
| 6 |
+
"n_fft": 400,
|
| 7 |
+
"n_samples": 480000,
|
| 8 |
+
"nb_max_frames": 3000,
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"padding_value": 0.0,
|
| 11 |
+
"processor_class": "WhisperProcessor",
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:10c582f8f67285309fe103d1e5ed042f60eca0383067610987b83ac2cac6c292
|
| 3 |
+
size 5496
|