Commit
•
daf256a
1
Parent(s):
170856c
up
Browse files- all_results.json +14 -0
- checkpoint-2000/config.json +291 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/preprocessor_config.json +9 -0
- checkpoint-2000/pytorch_model.bin +3 -0
- checkpoint-2000/rng_state_0.pth +3 -0
- checkpoint-2000/rng_state_1.pth +3 -0
- checkpoint-2000/rng_state_2.pth +3 -0
- checkpoint-2000/rng_state_3.pth +3 -0
- checkpoint-2000/rng_state_4.pth +3 -0
- checkpoint-2000/rng_state_5.pth +3 -0
- checkpoint-2000/rng_state_6.pth +3 -0
- checkpoint-2000/rng_state_7.pth +3 -0
- checkpoint-2000/scaler.pt +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/trainer_state.json +1261 -0
- checkpoint-2000/training_args.bin +3 -0
- config.json +291 -0
- create_model.py +28 -0
- emissions.csv +2 -0
- eval_results.json +9 -0
- merges.txt +0 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- run_librispeech.sh +31 -0
- run_speech_recognition_seq2seq.py +1 -0
- runs/Dec22_11-42-50_ip-172-31-1-149/1640173397.2128675/events.out.tfevents.1640173397.ip-172-31-1-149.489657.1 +3 -0
- runs/Dec22_11-42-50_ip-172-31-1-149/events.out.tfevents.1640173397.ip-172-31-1-149.489657.0 +3 -0
- runs/Dec22_11-42-50_ip-172-31-1-149/events.out.tfevents.1640176651.ip-172-31-1-149.489657.2 +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +1408 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 0.40512892603874207,
|
| 4 |
+
"eval_runtime": 60.9636,
|
| 5 |
+
"eval_samples": 2642,
|
| 6 |
+
"eval_samples_per_second": 43.337,
|
| 7 |
+
"eval_steps_per_second": 0.689,
|
| 8 |
+
"eval_wer": 0.11511305037683459,
|
| 9 |
+
"train_loss": 1.5071858802451146,
|
| 10 |
+
"train_runtime": 3191.5244,
|
| 11 |
+
"train_samples": 28538,
|
| 12 |
+
"train_samples_per_second": 44.709,
|
| 13 |
+
"train_steps_per_second": 0.699
|
| 14 |
+
}
|
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,291 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SpeechEncoderDecoderModel"
|
| 5 |
+
],
|
| 6 |
+
"decoder": {
|
| 7 |
+
"_name_or_path": "facebook/bart-base",
|
| 8 |
+
"activation_dropout": 0.1,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"add_bias_logits": false,
|
| 11 |
+
"add_cross_attention": true,
|
| 12 |
+
"add_final_layer_norm": false,
|
| 13 |
+
"architectures": [
|
| 14 |
+
"BartModel"
|
| 15 |
+
],
|
| 16 |
+
"attention_dropout": 0.1,
|
| 17 |
+
"bad_words_ids": null,
|
| 18 |
+
"bos_token_id": 0,
|
| 19 |
+
"chunk_size_feed_forward": 0,
|
| 20 |
+
"classif_dropout": 0.1,
|
| 21 |
+
"classifier_dropout": 0.0,
|
| 22 |
+
"cross_attention_hidden_size": null,
|
| 23 |
+
"d_model": 768,
|
| 24 |
+
"decoder_attention_heads": 12,
|
| 25 |
+
"decoder_ffn_dim": 3072,
|
| 26 |
+
"decoder_layerdrop": 0.0,
|
| 27 |
+
"decoder_layers": 6,
|
| 28 |
+
"decoder_start_token_id": 2,
|
| 29 |
+
"diversity_penalty": 0.0,
|
| 30 |
+
"do_sample": false,
|
| 31 |
+
"dropout": 0.1,
|
| 32 |
+
"early_stopping": true,
|
| 33 |
+
"encoder_attention_heads": 12,
|
| 34 |
+
"encoder_ffn_dim": 3072,
|
| 35 |
+
"encoder_layerdrop": 0.0,
|
| 36 |
+
"encoder_layers": 6,
|
| 37 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 38 |
+
"eos_token_id": 2,
|
| 39 |
+
"finetuning_task": null,
|
| 40 |
+
"forced_bos_token_id": null,
|
| 41 |
+
"forced_eos_token_id": 2,
|
| 42 |
+
"gradient_checkpointing": false,
|
| 43 |
+
"id2label": {
|
| 44 |
+
"0": "LABEL_0",
|
| 45 |
+
"1": "LABEL_1",
|
| 46 |
+
"2": "LABEL_2"
|
| 47 |
+
},
|
| 48 |
+
"init_std": 0.02,
|
| 49 |
+
"is_decoder": true,
|
| 50 |
+
"is_encoder_decoder": false,
|
| 51 |
+
"label2id": {
|
| 52 |
+
"LABEL_0": 0,
|
| 53 |
+
"LABEL_1": 1,
|
| 54 |
+
"LABEL_2": 2
|
| 55 |
+
},
|
| 56 |
+
"length_penalty": 1.0,
|
| 57 |
+
"max_length": 20,
|
| 58 |
+
"max_position_embeddings": 1024,
|
| 59 |
+
"min_length": 0,
|
| 60 |
+
"model_type": "bart",
|
| 61 |
+
"no_repeat_ngram_size": 3,
|
| 62 |
+
"normalize_before": false,
|
| 63 |
+
"normalize_embedding": true,
|
| 64 |
+
"num_beam_groups": 1,
|
| 65 |
+
"num_beams": 4,
|
| 66 |
+
"num_hidden_layers": 6,
|
| 67 |
+
"num_return_sequences": 1,
|
| 68 |
+
"output_attentions": false,
|
| 69 |
+
"output_hidden_states": false,
|
| 70 |
+
"output_scores": false,
|
| 71 |
+
"pad_token_id": 1,
|
| 72 |
+
"prefix": null,
|
| 73 |
+
"problem_type": null,
|
| 74 |
+
"pruned_heads": {},
|
| 75 |
+
"remove_invalid_values": false,
|
| 76 |
+
"repetition_penalty": 1.0,
|
| 77 |
+
"return_dict": true,
|
| 78 |
+
"return_dict_in_generate": false,
|
| 79 |
+
"scale_embedding": false,
|
| 80 |
+
"sep_token_id": null,
|
| 81 |
+
"task_specific_params": {
|
| 82 |
+
"summarization": {
|
| 83 |
+
"length_penalty": 1.0,
|
| 84 |
+
"max_length": 128,
|
| 85 |
+
"min_length": 12,
|
| 86 |
+
"num_beams": 4
|
| 87 |
+
},
|
| 88 |
+
"summarization_cnn": {
|
| 89 |
+
"length_penalty": 2.0,
|
| 90 |
+
"max_length": 142,
|
| 91 |
+
"min_length": 56,
|
| 92 |
+
"num_beams": 4
|
| 93 |
+
},
|
| 94 |
+
"summarization_xsum": {
|
| 95 |
+
"length_penalty": 1.0,
|
| 96 |
+
"max_length": 62,
|
| 97 |
+
"min_length": 11,
|
| 98 |
+
"num_beams": 6
|
| 99 |
+
}
|
| 100 |
+
},
|
| 101 |
+
"temperature": 1.0,
|
| 102 |
+
"tie_encoder_decoder": false,
|
| 103 |
+
"tie_word_embeddings": true,
|
| 104 |
+
"tokenizer_class": null,
|
| 105 |
+
"top_k": 50,
|
| 106 |
+
"top_p": 1.0,
|
| 107 |
+
"torch_dtype": "float32",
|
| 108 |
+
"torchscript": false,
|
| 109 |
+
"transformers_version": "4.15.0.dev0",
|
| 110 |
+
"use_bfloat16": false,
|
| 111 |
+
"use_cache": true,
|
| 112 |
+
"vocab_size": 50265
|
| 113 |
+
},
|
| 114 |
+
"decoder_start_token_id": 0,
|
| 115 |
+
"encoder": {
|
| 116 |
+
"_name_or_path": "facebook/wav2vec2-base",
|
| 117 |
+
"activation_dropout": 0.0,
|
| 118 |
+
"adapter_kernel_size": 3,
|
| 119 |
+
"adapter_stride": 2,
|
| 120 |
+
"add_adapter": false,
|
| 121 |
+
"add_cross_attention": false,
|
| 122 |
+
"apply_spec_augment": true,
|
| 123 |
+
"architectures": [
|
| 124 |
+
"Wav2Vec2ForPreTraining"
|
| 125 |
+
],
|
| 126 |
+
"attention_dropout": 0.1,
|
| 127 |
+
"bad_words_ids": null,
|
| 128 |
+
"bos_token_id": 1,
|
| 129 |
+
"chunk_size_feed_forward": 0,
|
| 130 |
+
"classifier_proj_size": 256,
|
| 131 |
+
"codevector_dim": 256,
|
| 132 |
+
"contrastive_logits_temperature": 0.1,
|
| 133 |
+
"conv_bias": false,
|
| 134 |
+
"conv_dim": [
|
| 135 |
+
512,
|
| 136 |
+
512,
|
| 137 |
+
512,
|
| 138 |
+
512,
|
| 139 |
+
512,
|
| 140 |
+
512,
|
| 141 |
+
512
|
| 142 |
+
],
|
| 143 |
+
"conv_kernel": [
|
| 144 |
+
10,
|
| 145 |
+
3,
|
| 146 |
+
3,
|
| 147 |
+
3,
|
| 148 |
+
3,
|
| 149 |
+
2,
|
| 150 |
+
2
|
| 151 |
+
],
|
| 152 |
+
"conv_stride": [
|
| 153 |
+
5,
|
| 154 |
+
2,
|
| 155 |
+
2,
|
| 156 |
+
2,
|
| 157 |
+
2,
|
| 158 |
+
2,
|
| 159 |
+
2
|
| 160 |
+
],
|
| 161 |
+
"cross_attention_hidden_size": null,
|
| 162 |
+
"ctc_loss_reduction": "sum",
|
| 163 |
+
"ctc_zero_infinity": false,
|
| 164 |
+
"decoder_start_token_id": null,
|
| 165 |
+
"diversity_loss_weight": 0.1,
|
| 166 |
+
"diversity_penalty": 0.0,
|
| 167 |
+
"do_sample": false,
|
| 168 |
+
"do_stable_layer_norm": false,
|
| 169 |
+
"early_stopping": false,
|
| 170 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 171 |
+
"eos_token_id": 2,
|
| 172 |
+
"feat_extract_activation": "gelu",
|
| 173 |
+
"feat_extract_norm": "group",
|
| 174 |
+
"feat_proj_dropout": 0.0,
|
| 175 |
+
"feat_quantizer_dropout": 0.0,
|
| 176 |
+
"final_dropout": 0.0,
|
| 177 |
+
"finetuning_task": null,
|
| 178 |
+
"forced_bos_token_id": null,
|
| 179 |
+
"forced_eos_token_id": null,
|
| 180 |
+
"freeze_feat_extract_train": true,
|
| 181 |
+
"hidden_act": "gelu",
|
| 182 |
+
"hidden_dropout": 0.1,
|
| 183 |
+
"hidden_size": 768,
|
| 184 |
+
"id2label": {
|
| 185 |
+
"0": "LABEL_0",
|
| 186 |
+
"1": "LABEL_1"
|
| 187 |
+
},
|
| 188 |
+
"initializer_range": 0.02,
|
| 189 |
+
"intermediate_size": 3072,
|
| 190 |
+
"is_decoder": false,
|
| 191 |
+
"is_encoder_decoder": false,
|
| 192 |
+
"label2id": {
|
| 193 |
+
"LABEL_0": 0,
|
| 194 |
+
"LABEL_1": 1
|
| 195 |
+
},
|
| 196 |
+
"layer_norm_eps": 1e-05,
|
| 197 |
+
"layerdrop": 0.0,
|
| 198 |
+
"length_penalty": 1.0,
|
| 199 |
+
"mask_channel_length": 10,
|
| 200 |
+
"mask_channel_min_space": 1,
|
| 201 |
+
"mask_channel_other": 0.0,
|
| 202 |
+
"mask_channel_prob": 0.0,
|
| 203 |
+
"mask_channel_selection": "static",
|
| 204 |
+
"mask_feature_length": 10,
|
| 205 |
+
"mask_feature_min_masks": 0,
|
| 206 |
+
"mask_feature_prob": 0.0,
|
| 207 |
+
"mask_time_length": 10,
|
| 208 |
+
"mask_time_min_masks": 2,
|
| 209 |
+
"mask_time_min_space": 1,
|
| 210 |
+
"mask_time_other": 0.0,
|
| 211 |
+
"mask_time_prob": 0.0,
|
| 212 |
+
"mask_time_selection": "static",
|
| 213 |
+
"max_length": 20,
|
| 214 |
+
"min_length": 0,
|
| 215 |
+
"model_type": "wav2vec2",
|
| 216 |
+
"no_mask_channel_overlap": false,
|
| 217 |
+
"no_mask_time_overlap": false,
|
| 218 |
+
"no_repeat_ngram_size": 0,
|
| 219 |
+
"num_adapter_layers": 3,
|
| 220 |
+
"num_attention_heads": 12,
|
| 221 |
+
"num_beam_groups": 1,
|
| 222 |
+
"num_beams": 1,
|
| 223 |
+
"num_codevector_groups": 2,
|
| 224 |
+
"num_codevectors_per_group": 320,
|
| 225 |
+
"num_conv_pos_embedding_groups": 16,
|
| 226 |
+
"num_conv_pos_embeddings": 128,
|
| 227 |
+
"num_feat_extract_layers": 7,
|
| 228 |
+
"num_hidden_layers": 12,
|
| 229 |
+
"num_negatives": 100,
|
| 230 |
+
"num_return_sequences": 1,
|
| 231 |
+
"output_attentions": false,
|
| 232 |
+
"output_hidden_size": 768,
|
| 233 |
+
"output_hidden_states": false,
|
| 234 |
+
"output_scores": false,
|
| 235 |
+
"pad_token_id": 0,
|
| 236 |
+
"prefix": null,
|
| 237 |
+
"problem_type": null,
|
| 238 |
+
"proj_codevector_dim": 256,
|
| 239 |
+
"pruned_heads": {},
|
| 240 |
+
"remove_invalid_values": false,
|
| 241 |
+
"repetition_penalty": 1.0,
|
| 242 |
+
"return_dict": true,
|
| 243 |
+
"return_dict_in_generate": false,
|
| 244 |
+
"sep_token_id": null,
|
| 245 |
+
"task_specific_params": null,
|
| 246 |
+
"tdnn_dilation": [
|
| 247 |
+
1,
|
| 248 |
+
2,
|
| 249 |
+
3,
|
| 250 |
+
1,
|
| 251 |
+
1
|
| 252 |
+
],
|
| 253 |
+
"tdnn_dim": [
|
| 254 |
+
512,
|
| 255 |
+
512,
|
| 256 |
+
512,
|
| 257 |
+
512,
|
| 258 |
+
1500
|
| 259 |
+
],
|
| 260 |
+
"tdnn_kernel": [
|
| 261 |
+
5,
|
| 262 |
+
3,
|
| 263 |
+
3,
|
| 264 |
+
1,
|
| 265 |
+
1
|
| 266 |
+
],
|
| 267 |
+
"temperature": 1.0,
|
| 268 |
+
"tie_encoder_decoder": false,
|
| 269 |
+
"tie_word_embeddings": true,
|
| 270 |
+
"tokenizer_class": null,
|
| 271 |
+
"top_k": 50,
|
| 272 |
+
"top_p": 1.0,
|
| 273 |
+
"torch_dtype": null,
|
| 274 |
+
"torchscript": false,
|
| 275 |
+
"transformers_version": "4.15.0.dev0",
|
| 276 |
+
"use_bfloat16": false,
|
| 277 |
+
"use_weighted_layer_sum": false,
|
| 278 |
+
"vocab_size": 32,
|
| 279 |
+
"xvector_output_dim": 512
|
| 280 |
+
},
|
| 281 |
+
"eos_token_id": 2,
|
| 282 |
+
"is_encoder_decoder": true,
|
| 283 |
+
"max_length": 40,
|
| 284 |
+
"model_type": "speech-encoder-decoder",
|
| 285 |
+
"pad_token_id": 1,
|
| 286 |
+
"processor_class": "Wav2Vec2Processor",
|
| 287 |
+
"tie_word_embeddings": false,
|
| 288 |
+
"torch_dtype": "float32",
|
| 289 |
+
"transformers_version": null,
|
| 290 |
+
"use_cache": false
|
| 291 |
+
}
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e76216221fafaca239b48abfe24695f6284b6d1ae9706fb0e96538b24ca33d24
|
| 3 |
+
size 1490403485
|
checkpoint-2000/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-2000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c168c067f3ef1a818e652b62239480599d84a6ea9a3cf62c467bb0d90eb0ccae
|
| 3 |
+
size 762044061
|
checkpoint-2000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ebc514b4c2ff1100143773b40b8ceddcd9390d17c0390cd40c92a0837864877
|
| 3 |
+
size 14503
|
checkpoint-2000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c11f5462e713e51522d9f31484953faff9f51d140a428f358a42f74b74cdb7aa
|
| 3 |
+
size 14567
|
checkpoint-2000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57efa632fbc5d2c57a77e926e9d1576ca04ceaf01cf846244f4ac4c85424d4b9
|
| 3 |
+
size 14567
|
checkpoint-2000/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:170ea09b95a6e1033c5f4f5df488f3b7f7f710a2d809295678c4b94a3146139f
|
| 3 |
+
size 14503
|
checkpoint-2000/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3db3e04745c8f7c97325c1168f43274ef056808b1fc2db2b93bfa692d6a6febd
|
| 3 |
+
size 14567
|
checkpoint-2000/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c48bbafee794ff028deda24ce43a6491dad82b4e7ab08d70c5608d22f2d9c3ee
|
| 3 |
+
size 14567
|
checkpoint-2000/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e2cd3af5cd3d234483896b951485610dd1f17938ccd32aa72a0c755812238d4
|
| 3 |
+
size 14567
|
checkpoint-2000/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebcb2521ccba33c6c7014e9feba616a2e19bd4cc85bbd7b24e7ed35b78b42a95
|
| 3 |
+
size 14567
|
checkpoint-2000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63d5772e08106162607021c01da6556e4f87840a3f024efbbb87d2ab17e06a8a
|
| 3 |
+
size 559
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:962d148339910b256662f48b08f7142018af7cb6572615fd9caff6809d635cbb
|
| 3 |
+
size 623
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,1261 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.484304932735426,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 2.2499999999999996e-06,
|
| 13 |
+
"loss": 11.7449,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 9.75e-06,
|
| 19 |
+
"loss": 9.9528,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 1.725e-05,
|
| 25 |
+
"loss": 7.8835,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"learning_rate": 2.475e-05,
|
| 31 |
+
"loss": 6.9463,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"learning_rate": 3.225e-05,
|
| 37 |
+
"loss": 6.3116,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.13,
|
| 42 |
+
"learning_rate": 3.975e-05,
|
| 43 |
+
"loss": 5.8418,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.16,
|
| 48 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 49 |
+
"loss": 5.6196,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.18,
|
| 54 |
+
"learning_rate": 5.4749999999999996e-05,
|
| 55 |
+
"loss": 5.4065,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.2,
|
| 60 |
+
"learning_rate": 6.225e-05,
|
| 61 |
+
"loss": 5.3217,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.22,
|
| 66 |
+
"learning_rate": 6.975e-05,
|
| 67 |
+
"loss": 5.2094,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.25,
|
| 72 |
+
"learning_rate": 7.725e-05,
|
| 73 |
+
"loss": 5.2388,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.27,
|
| 78 |
+
"learning_rate": 8.474999999999999e-05,
|
| 79 |
+
"loss": 5.1522,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.29,
|
| 84 |
+
"learning_rate": 9.224999999999999e-05,
|
| 85 |
+
"loss": 5.1679,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31,
|
| 90 |
+
"learning_rate": 9.975e-05,
|
| 91 |
+
"loss": 5.112,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.34,
|
| 96 |
+
"learning_rate": 0.00010724999999999999,
|
| 97 |
+
"loss": 5.1008,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.36,
|
| 102 |
+
"learning_rate": 0.00011475,
|
| 103 |
+
"loss": 5.1024,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.38,
|
| 108 |
+
"learning_rate": 0.00012225,
|
| 109 |
+
"loss": 5.0238,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.4,
|
| 114 |
+
"learning_rate": 0.00012974999999999998,
|
| 115 |
+
"loss": 5.061,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.43,
|
| 120 |
+
"learning_rate": 0.00013725,
|
| 121 |
+
"loss": 5.0293,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.45,
|
| 126 |
+
"learning_rate": 0.00014475,
|
| 127 |
+
"loss": 5.0315,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.47,
|
| 132 |
+
"learning_rate": 0.00015224999999999996,
|
| 133 |
+
"loss": 5.1264,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.49,
|
| 138 |
+
"learning_rate": 0.00015974999999999998,
|
| 139 |
+
"loss": 4.9663,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.52,
|
| 144 |
+
"learning_rate": 0.00016724999999999997,
|
| 145 |
+
"loss": 4.9525,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.54,
|
| 150 |
+
"learning_rate": 0.00017475,
|
| 151 |
+
"loss": 4.9757,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.56,
|
| 156 |
+
"learning_rate": 0.00018224999999999998,
|
| 157 |
+
"loss": 4.9203,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.58,
|
| 162 |
+
"learning_rate": 0.00018974999999999998,
|
| 163 |
+
"loss": 4.908,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.61,
|
| 168 |
+
"learning_rate": 0.00019724999999999997,
|
| 169 |
+
"loss": 4.845,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.63,
|
| 174 |
+
"learning_rate": 0.00020475,
|
| 175 |
+
"loss": 4.8367,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.65,
|
| 180 |
+
"learning_rate": 0.00021224999999999998,
|
| 181 |
+
"loss": 4.8034,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.67,
|
| 186 |
+
"learning_rate": 0.00021975,
|
| 187 |
+
"loss": 4.8215,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.7,
|
| 192 |
+
"learning_rate": 0.00022724999999999997,
|
| 193 |
+
"loss": 4.7776,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.72,
|
| 198 |
+
"learning_rate": 0.00023474999999999996,
|
| 199 |
+
"loss": 4.7166,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.74,
|
| 204 |
+
"learning_rate": 0.00024224999999999998,
|
| 205 |
+
"loss": 4.7696,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.76,
|
| 210 |
+
"learning_rate": 0.000249,
|
| 211 |
+
"loss": 4.6817,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.78,
|
| 216 |
+
"learning_rate": 0.00025649999999999995,
|
| 217 |
+
"loss": 4.7147,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.81,
|
| 222 |
+
"learning_rate": 0.00026399999999999997,
|
| 223 |
+
"loss": 4.6688,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.83,
|
| 228 |
+
"learning_rate": 0.0002715,
|
| 229 |
+
"loss": 4.5991,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.85,
|
| 234 |
+
"learning_rate": 0.000279,
|
| 235 |
+
"loss": 4.6401,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.87,
|
| 240 |
+
"learning_rate": 0.00028649999999999997,
|
| 241 |
+
"loss": 4.6029,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.9,
|
| 246 |
+
"learning_rate": 0.000294,
|
| 247 |
+
"loss": 4.5118,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.9,
|
| 252 |
+
"eval_loss": 4.247889041900635,
|
| 253 |
+
"eval_runtime": 51.3593,
|
| 254 |
+
"eval_samples_per_second": 51.442,
|
| 255 |
+
"eval_steps_per_second": 0.818,
|
| 256 |
+
"eval_wer": 0.9033518445061484,
|
| 257 |
+
"step": 400
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.92,
|
| 261 |
+
"learning_rate": 0.000299672131147541,
|
| 262 |
+
"loss": 4.5483,
|
| 263 |
+
"step": 410
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.94,
|
| 267 |
+
"learning_rate": 0.00029803278688524587,
|
| 268 |
+
"loss": 4.4005,
|
| 269 |
+
"step": 420
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.96,
|
| 273 |
+
"learning_rate": 0.00029639344262295076,
|
| 274 |
+
"loss": 4.327,
|
| 275 |
+
"step": 430
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.99,
|
| 279 |
+
"learning_rate": 0.0002947540983606557,
|
| 280 |
+
"loss": 4.2717,
|
| 281 |
+
"step": 440
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.01,
|
| 285 |
+
"learning_rate": 0.0002931147540983606,
|
| 286 |
+
"loss": 4.0139,
|
| 287 |
+
"step": 450
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.03,
|
| 291 |
+
"learning_rate": 0.00029147540983606555,
|
| 292 |
+
"loss": 3.8304,
|
| 293 |
+
"step": 460
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.05,
|
| 297 |
+
"learning_rate": 0.0002898360655737705,
|
| 298 |
+
"loss": 3.5825,
|
| 299 |
+
"step": 470
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.08,
|
| 303 |
+
"learning_rate": 0.0002881967213114754,
|
| 304 |
+
"loss": 3.4968,
|
| 305 |
+
"step": 480
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.1,
|
| 309 |
+
"learning_rate": 0.0002865573770491803,
|
| 310 |
+
"loss": 3.2832,
|
| 311 |
+
"step": 490
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.12,
|
| 315 |
+
"learning_rate": 0.00028491803278688524,
|
| 316 |
+
"loss": 3.1368,
|
| 317 |
+
"step": 500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.14,
|
| 321 |
+
"learning_rate": 0.00028327868852459013,
|
| 322 |
+
"loss": 2.899,
|
| 323 |
+
"step": 510
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.17,
|
| 327 |
+
"learning_rate": 0.0002816393442622951,
|
| 328 |
+
"loss": 2.6293,
|
| 329 |
+
"step": 520
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.19,
|
| 333 |
+
"learning_rate": 0.00028,
|
| 334 |
+
"loss": 2.7137,
|
| 335 |
+
"step": 530
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.21,
|
| 339 |
+
"learning_rate": 0.00027836065573770487,
|
| 340 |
+
"loss": 2.4069,
|
| 341 |
+
"step": 540
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.23,
|
| 345 |
+
"learning_rate": 0.0002767213114754098,
|
| 346 |
+
"loss": 2.3349,
|
| 347 |
+
"step": 550
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.26,
|
| 351 |
+
"learning_rate": 0.0002750819672131147,
|
| 352 |
+
"loss": 2.1064,
|
| 353 |
+
"step": 560
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.28,
|
| 357 |
+
"learning_rate": 0.00027344262295081966,
|
| 358 |
+
"loss": 1.9375,
|
| 359 |
+
"step": 570
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.3,
|
| 363 |
+
"learning_rate": 0.00027180327868852455,
|
| 364 |
+
"loss": 1.9395,
|
| 365 |
+
"step": 580
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.32,
|
| 369 |
+
"learning_rate": 0.0002701639344262295,
|
| 370 |
+
"loss": 1.7861,
|
| 371 |
+
"step": 590
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.35,
|
| 375 |
+
"learning_rate": 0.0002685245901639344,
|
| 376 |
+
"loss": 1.6795,
|
| 377 |
+
"step": 600
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.37,
|
| 381 |
+
"learning_rate": 0.0002668852459016393,
|
| 382 |
+
"loss": 1.5166,
|
| 383 |
+
"step": 610
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.39,
|
| 387 |
+
"learning_rate": 0.00026524590163934424,
|
| 388 |
+
"loss": 1.4776,
|
| 389 |
+
"step": 620
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.41,
|
| 393 |
+
"learning_rate": 0.00026360655737704913,
|
| 394 |
+
"loss": 1.4286,
|
| 395 |
+
"step": 630
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.43,
|
| 399 |
+
"learning_rate": 0.0002619672131147541,
|
| 400 |
+
"loss": 1.3071,
|
| 401 |
+
"step": 640
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.46,
|
| 405 |
+
"learning_rate": 0.00026032786885245903,
|
| 406 |
+
"loss": 1.277,
|
| 407 |
+
"step": 650
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.48,
|
| 411 |
+
"learning_rate": 0.0002586885245901639,
|
| 412 |
+
"loss": 1.2741,
|
| 413 |
+
"step": 660
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.5,
|
| 417 |
+
"learning_rate": 0.0002570491803278688,
|
| 418 |
+
"loss": 1.1661,
|
| 419 |
+
"step": 670
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.52,
|
| 423 |
+
"learning_rate": 0.00025540983606557377,
|
| 424 |
+
"loss": 1.1585,
|
| 425 |
+
"step": 680
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.55,
|
| 429 |
+
"learning_rate": 0.00025377049180327866,
|
| 430 |
+
"loss": 1.1015,
|
| 431 |
+
"step": 690
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.57,
|
| 435 |
+
"learning_rate": 0.0002521311475409836,
|
| 436 |
+
"loss": 1.0403,
|
| 437 |
+
"step": 700
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.59,
|
| 441 |
+
"learning_rate": 0.0002504918032786885,
|
| 442 |
+
"loss": 1.0001,
|
| 443 |
+
"step": 710
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.61,
|
| 447 |
+
"learning_rate": 0.0002488524590163934,
|
| 448 |
+
"loss": 0.9915,
|
| 449 |
+
"step": 720
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.64,
|
| 453 |
+
"learning_rate": 0.00024721311475409835,
|
| 454 |
+
"loss": 0.9481,
|
| 455 |
+
"step": 730
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.66,
|
| 459 |
+
"learning_rate": 0.00024557377049180324,
|
| 460 |
+
"loss": 0.9522,
|
| 461 |
+
"step": 740
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.68,
|
| 465 |
+
"learning_rate": 0.00024393442622950816,
|
| 466 |
+
"loss": 0.9063,
|
| 467 |
+
"step": 750
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.7,
|
| 471 |
+
"learning_rate": 0.00024229508196721308,
|
| 472 |
+
"loss": 0.8868,
|
| 473 |
+
"step": 760
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.73,
|
| 477 |
+
"learning_rate": 0.000240655737704918,
|
| 478 |
+
"loss": 0.818,
|
| 479 |
+
"step": 770
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.75,
|
| 483 |
+
"learning_rate": 0.00023901639344262293,
|
| 484 |
+
"loss": 0.8436,
|
| 485 |
+
"step": 780
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.77,
|
| 489 |
+
"learning_rate": 0.00023737704918032785,
|
| 490 |
+
"loss": 0.7755,
|
| 491 |
+
"step": 790
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.79,
|
| 495 |
+
"learning_rate": 0.00023573770491803274,
|
| 496 |
+
"loss": 0.8016,
|
| 497 |
+
"step": 800
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.79,
|
| 501 |
+
"eval_loss": 0.8116478323936462,
|
| 502 |
+
"eval_runtime": 61.0806,
|
| 503 |
+
"eval_samples_per_second": 43.254,
|
| 504 |
+
"eval_steps_per_second": 0.688,
|
| 505 |
+
"eval_wer": 0.2206862356207854,
|
| 506 |
+
"step": 800
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.82,
|
| 510 |
+
"learning_rate": 0.00023409836065573766,
|
| 511 |
+
"loss": 0.7623,
|
| 512 |
+
"step": 810
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.84,
|
| 516 |
+
"learning_rate": 0.00023245901639344258,
|
| 517 |
+
"loss": 0.7884,
|
| 518 |
+
"step": 820
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.86,
|
| 522 |
+
"learning_rate": 0.00023081967213114753,
|
| 523 |
+
"loss": 0.7791,
|
| 524 |
+
"step": 830
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.88,
|
| 528 |
+
"learning_rate": 0.00022918032786885245,
|
| 529 |
+
"loss": 0.7111,
|
| 530 |
+
"step": 840
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.91,
|
| 534 |
+
"learning_rate": 0.00022754098360655738,
|
| 535 |
+
"loss": 0.7063,
|
| 536 |
+
"step": 850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.93,
|
| 540 |
+
"learning_rate": 0.0002259016393442623,
|
| 541 |
+
"loss": 0.7307,
|
| 542 |
+
"step": 860
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.95,
|
| 546 |
+
"learning_rate": 0.0002242622950819672,
|
| 547 |
+
"loss": 0.6792,
|
| 548 |
+
"step": 870
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.97,
|
| 552 |
+
"learning_rate": 0.0002226229508196721,
|
| 553 |
+
"loss": 0.6496,
|
| 554 |
+
"step": 880
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.0,
|
| 558 |
+
"learning_rate": 0.00022098360655737703,
|
| 559 |
+
"loss": 0.6929,
|
| 560 |
+
"step": 890
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 2.02,
|
| 564 |
+
"learning_rate": 0.00021934426229508195,
|
| 565 |
+
"loss": 0.5503,
|
| 566 |
+
"step": 900
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 2.04,
|
| 570 |
+
"learning_rate": 0.00021770491803278688,
|
| 571 |
+
"loss": 0.4833,
|
| 572 |
+
"step": 910
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 2.06,
|
| 576 |
+
"learning_rate": 0.0002160655737704918,
|
| 577 |
+
"loss": 0.5107,
|
| 578 |
+
"step": 920
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.09,
|
| 582 |
+
"learning_rate": 0.0002144262295081967,
|
| 583 |
+
"loss": 0.4637,
|
| 584 |
+
"step": 930
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.11,
|
| 588 |
+
"learning_rate": 0.0002127868852459016,
|
| 589 |
+
"loss": 0.4764,
|
| 590 |
+
"step": 940
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.13,
|
| 594 |
+
"learning_rate": 0.00021114754098360653,
|
| 595 |
+
"loss": 0.4332,
|
| 596 |
+
"step": 950
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.15,
|
| 600 |
+
"learning_rate": 0.00020950819672131146,
|
| 601 |
+
"loss": 0.4599,
|
| 602 |
+
"step": 960
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 2.17,
|
| 606 |
+
"learning_rate": 0.00020786885245901638,
|
| 607 |
+
"loss": 0.4582,
|
| 608 |
+
"step": 970
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.2,
|
| 612 |
+
"learning_rate": 0.00020622950819672127,
|
| 613 |
+
"loss": 0.4112,
|
| 614 |
+
"step": 980
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.22,
|
| 618 |
+
"learning_rate": 0.0002045901639344262,
|
| 619 |
+
"loss": 0.4441,
|
| 620 |
+
"step": 990
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.24,
|
| 624 |
+
"learning_rate": 0.0002029508196721311,
|
| 625 |
+
"loss": 0.4575,
|
| 626 |
+
"step": 1000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.26,
|
| 630 |
+
"learning_rate": 0.00020131147540983603,
|
| 631 |
+
"loss": 0.4177,
|
| 632 |
+
"step": 1010
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.29,
|
| 636 |
+
"learning_rate": 0.00019967213114754098,
|
| 637 |
+
"loss": 0.4323,
|
| 638 |
+
"step": 1020
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.31,
|
| 642 |
+
"learning_rate": 0.0001980327868852459,
|
| 643 |
+
"loss": 0.4295,
|
| 644 |
+
"step": 1030
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.33,
|
| 648 |
+
"learning_rate": 0.00019639344262295083,
|
| 649 |
+
"loss": 0.4595,
|
| 650 |
+
"step": 1040
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.35,
|
| 654 |
+
"learning_rate": 0.00019475409836065572,
|
| 655 |
+
"loss": 0.4338,
|
| 656 |
+
"step": 1050
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.38,
|
| 660 |
+
"learning_rate": 0.00019311475409836064,
|
| 661 |
+
"loss": 0.4104,
|
| 662 |
+
"step": 1060
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.4,
|
| 666 |
+
"learning_rate": 0.00019147540983606556,
|
| 667 |
+
"loss": 0.4216,
|
| 668 |
+
"step": 1070
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.42,
|
| 672 |
+
"learning_rate": 0.00018983606557377048,
|
| 673 |
+
"loss": 0.3729,
|
| 674 |
+
"step": 1080
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.44,
|
| 678 |
+
"learning_rate": 0.0001881967213114754,
|
| 679 |
+
"loss": 0.3998,
|
| 680 |
+
"step": 1090
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.47,
|
| 684 |
+
"learning_rate": 0.0001865573770491803,
|
| 685 |
+
"loss": 0.4082,
|
| 686 |
+
"step": 1100
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.49,
|
| 690 |
+
"learning_rate": 0.00018491803278688522,
|
| 691 |
+
"loss": 0.38,
|
| 692 |
+
"step": 1110
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.51,
|
| 696 |
+
"learning_rate": 0.00018327868852459014,
|
| 697 |
+
"loss": 0.4157,
|
| 698 |
+
"step": 1120
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.53,
|
| 702 |
+
"learning_rate": 0.00018163934426229506,
|
| 703 |
+
"loss": 0.3937,
|
| 704 |
+
"step": 1130
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.56,
|
| 708 |
+
"learning_rate": 0.00017999999999999998,
|
| 709 |
+
"loss": 0.3772,
|
| 710 |
+
"step": 1140
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.58,
|
| 714 |
+
"learning_rate": 0.0001783606557377049,
|
| 715 |
+
"loss": 0.3767,
|
| 716 |
+
"step": 1150
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.6,
|
| 720 |
+
"learning_rate": 0.0001767213114754098,
|
| 721 |
+
"loss": 0.3602,
|
| 722 |
+
"step": 1160
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 2.62,
|
| 726 |
+
"learning_rate": 0.00017508196721311472,
|
| 727 |
+
"loss": 0.3876,
|
| 728 |
+
"step": 1170
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.65,
|
| 732 |
+
"learning_rate": 0.00017344262295081964,
|
| 733 |
+
"loss": 0.3776,
|
| 734 |
+
"step": 1180
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.67,
|
| 738 |
+
"learning_rate": 0.00017180327868852456,
|
| 739 |
+
"loss": 0.3617,
|
| 740 |
+
"step": 1190
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 2.69,
|
| 744 |
+
"learning_rate": 0.0001701639344262295,
|
| 745 |
+
"loss": 0.3627,
|
| 746 |
+
"step": 1200
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.69,
|
| 750 |
+
"eval_loss": 0.5665903091430664,
|
| 751 |
+
"eval_runtime": 61.1317,
|
| 752 |
+
"eval_samples_per_second": 43.218,
|
| 753 |
+
"eval_steps_per_second": 0.687,
|
| 754 |
+
"eval_wer": 0.1506743355811186,
|
| 755 |
+
"step": 1200
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.71,
|
| 759 |
+
"learning_rate": 0.00016852459016393443,
|
| 760 |
+
"loss": 0.3478,
|
| 761 |
+
"step": 1210
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.74,
|
| 765 |
+
"learning_rate": 0.00016688524590163935,
|
| 766 |
+
"loss": 0.3861,
|
| 767 |
+
"step": 1220
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.76,
|
| 771 |
+
"learning_rate": 0.00016524590163934425,
|
| 772 |
+
"loss": 0.3283,
|
| 773 |
+
"step": 1230
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.78,
|
| 777 |
+
"learning_rate": 0.00016360655737704917,
|
| 778 |
+
"loss": 0.3455,
|
| 779 |
+
"step": 1240
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.8,
|
| 783 |
+
"learning_rate": 0.0001619672131147541,
|
| 784 |
+
"loss": 0.3457,
|
| 785 |
+
"step": 1250
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.83,
|
| 789 |
+
"learning_rate": 0.000160327868852459,
|
| 790 |
+
"loss": 0.3178,
|
| 791 |
+
"step": 1260
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 2.85,
|
| 795 |
+
"learning_rate": 0.00015868852459016393,
|
| 796 |
+
"loss": 0.3429,
|
| 797 |
+
"step": 1270
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 2.87,
|
| 801 |
+
"learning_rate": 0.00015704918032786883,
|
| 802 |
+
"loss": 0.3253,
|
| 803 |
+
"step": 1280
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.89,
|
| 807 |
+
"learning_rate": 0.00015540983606557375,
|
| 808 |
+
"loss": 0.3469,
|
| 809 |
+
"step": 1290
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.91,
|
| 813 |
+
"learning_rate": 0.00015377049180327867,
|
| 814 |
+
"loss": 0.334,
|
| 815 |
+
"step": 1300
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.94,
|
| 819 |
+
"learning_rate": 0.0001521311475409836,
|
| 820 |
+
"loss": 0.3246,
|
| 821 |
+
"step": 1310
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.96,
|
| 825 |
+
"learning_rate": 0.0001504918032786885,
|
| 826 |
+
"loss": 0.3243,
|
| 827 |
+
"step": 1320
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.98,
|
| 831 |
+
"learning_rate": 0.00014885245901639343,
|
| 832 |
+
"loss": 0.3219,
|
| 833 |
+
"step": 1330
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.0,
|
| 837 |
+
"learning_rate": 0.00014721311475409836,
|
| 838 |
+
"loss": 0.3022,
|
| 839 |
+
"step": 1340
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 3.03,
|
| 843 |
+
"learning_rate": 0.00014557377049180328,
|
| 844 |
+
"loss": 0.1935,
|
| 845 |
+
"step": 1350
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.05,
|
| 849 |
+
"learning_rate": 0.0001439344262295082,
|
| 850 |
+
"loss": 0.2093,
|
| 851 |
+
"step": 1360
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.07,
|
| 855 |
+
"learning_rate": 0.0001422950819672131,
|
| 856 |
+
"loss": 0.1961,
|
| 857 |
+
"step": 1370
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.09,
|
| 861 |
+
"learning_rate": 0.00014065573770491801,
|
| 862 |
+
"loss": 0.1788,
|
| 863 |
+
"step": 1380
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.12,
|
| 867 |
+
"learning_rate": 0.00013901639344262294,
|
| 868 |
+
"loss": 0.2245,
|
| 869 |
+
"step": 1390
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 3.14,
|
| 873 |
+
"learning_rate": 0.00013737704918032786,
|
| 874 |
+
"loss": 0.1815,
|
| 875 |
+
"step": 1400
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 3.16,
|
| 879 |
+
"learning_rate": 0.00013573770491803278,
|
| 880 |
+
"loss": 0.2018,
|
| 881 |
+
"step": 1410
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 3.18,
|
| 885 |
+
"learning_rate": 0.0001340983606557377,
|
| 886 |
+
"loss": 0.1909,
|
| 887 |
+
"step": 1420
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.21,
|
| 891 |
+
"learning_rate": 0.00013245901639344262,
|
| 892 |
+
"loss": 0.2025,
|
| 893 |
+
"step": 1430
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 3.23,
|
| 897 |
+
"learning_rate": 0.00013081967213114754,
|
| 898 |
+
"loss": 0.1927,
|
| 899 |
+
"step": 1440
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 3.25,
|
| 903 |
+
"learning_rate": 0.00012918032786885246,
|
| 904 |
+
"loss": 0.1816,
|
| 905 |
+
"step": 1450
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 3.27,
|
| 909 |
+
"learning_rate": 0.00012754098360655736,
|
| 910 |
+
"loss": 0.2217,
|
| 911 |
+
"step": 1460
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 3.3,
|
| 915 |
+
"learning_rate": 0.00012590163934426228,
|
| 916 |
+
"loss": 0.2035,
|
| 917 |
+
"step": 1470
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.32,
|
| 921 |
+
"learning_rate": 0.0001242622950819672,
|
| 922 |
+
"loss": 0.1904,
|
| 923 |
+
"step": 1480
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 3.34,
|
| 927 |
+
"learning_rate": 0.00012262295081967212,
|
| 928 |
+
"loss": 0.1945,
|
| 929 |
+
"step": 1490
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.36,
|
| 933 |
+
"learning_rate": 0.00012098360655737703,
|
| 934 |
+
"loss": 0.1724,
|
| 935 |
+
"step": 1500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 3.39,
|
| 939 |
+
"learning_rate": 0.00011934426229508195,
|
| 940 |
+
"loss": 0.1704,
|
| 941 |
+
"step": 1510
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 3.41,
|
| 945 |
+
"learning_rate": 0.00011770491803278688,
|
| 946 |
+
"loss": 0.1745,
|
| 947 |
+
"step": 1520
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 3.43,
|
| 951 |
+
"learning_rate": 0.00011606557377049179,
|
| 952 |
+
"loss": 0.1756,
|
| 953 |
+
"step": 1530
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 3.45,
|
| 957 |
+
"learning_rate": 0.00011442622950819671,
|
| 958 |
+
"loss": 0.1864,
|
| 959 |
+
"step": 1540
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.48,
|
| 963 |
+
"learning_rate": 0.00011278688524590164,
|
| 964 |
+
"loss": 0.1799,
|
| 965 |
+
"step": 1550
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 3.5,
|
| 969 |
+
"learning_rate": 0.00011114754098360654,
|
| 970 |
+
"loss": 0.1974,
|
| 971 |
+
"step": 1560
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.52,
|
| 975 |
+
"learning_rate": 0.00010950819672131146,
|
| 976 |
+
"loss": 0.1751,
|
| 977 |
+
"step": 1570
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.54,
|
| 981 |
+
"learning_rate": 0.00010786885245901639,
|
| 982 |
+
"loss": 0.1792,
|
| 983 |
+
"step": 1580
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 3.57,
|
| 987 |
+
"learning_rate": 0.00010622950819672129,
|
| 988 |
+
"loss": 0.1887,
|
| 989 |
+
"step": 1590
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 3.59,
|
| 993 |
+
"learning_rate": 0.00010459016393442621,
|
| 994 |
+
"loss": 0.1763,
|
| 995 |
+
"step": 1600
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 3.59,
|
| 999 |
+
"eval_loss": 0.47305428981781006,
|
| 1000 |
+
"eval_runtime": 60.8349,
|
| 1001 |
+
"eval_samples_per_second": 43.429,
|
| 1002 |
+
"eval_steps_per_second": 0.69,
|
| 1003 |
+
"eval_wer": 0.12973026576755256,
|
| 1004 |
+
"step": 1600
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 3.61,
|
| 1008 |
+
"learning_rate": 0.00010295081967213115,
|
| 1009 |
+
"loss": 0.1767,
|
| 1010 |
+
"step": 1610
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 3.63,
|
| 1014 |
+
"learning_rate": 0.00010131147540983606,
|
| 1015 |
+
"loss": 0.1769,
|
| 1016 |
+
"step": 1620
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 3.65,
|
| 1020 |
+
"learning_rate": 9.967213114754098e-05,
|
| 1021 |
+
"loss": 0.1691,
|
| 1022 |
+
"step": 1630
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.68,
|
| 1026 |
+
"learning_rate": 9.80327868852459e-05,
|
| 1027 |
+
"loss": 0.1762,
|
| 1028 |
+
"step": 1640
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 3.7,
|
| 1032 |
+
"learning_rate": 9.639344262295081e-05,
|
| 1033 |
+
"loss": 0.1636,
|
| 1034 |
+
"step": 1650
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 3.72,
|
| 1038 |
+
"learning_rate": 9.475409836065573e-05,
|
| 1039 |
+
"loss": 0.1834,
|
| 1040 |
+
"step": 1660
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.74,
|
| 1044 |
+
"learning_rate": 9.311475409836065e-05,
|
| 1045 |
+
"loss": 0.1781,
|
| 1046 |
+
"step": 1670
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 3.77,
|
| 1050 |
+
"learning_rate": 9.147540983606556e-05,
|
| 1051 |
+
"loss": 0.1686,
|
| 1052 |
+
"step": 1680
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.79,
|
| 1056 |
+
"learning_rate": 8.983606557377048e-05,
|
| 1057 |
+
"loss": 0.1657,
|
| 1058 |
+
"step": 1690
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 3.81,
|
| 1062 |
+
"learning_rate": 8.819672131147541e-05,
|
| 1063 |
+
"loss": 0.1468,
|
| 1064 |
+
"step": 1700
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.83,
|
| 1068 |
+
"learning_rate": 8.655737704918032e-05,
|
| 1069 |
+
"loss": 0.1686,
|
| 1070 |
+
"step": 1710
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 3.86,
|
| 1074 |
+
"learning_rate": 8.491803278688524e-05,
|
| 1075 |
+
"loss": 0.1583,
|
| 1076 |
+
"step": 1720
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.88,
|
| 1080 |
+
"learning_rate": 8.327868852459016e-05,
|
| 1081 |
+
"loss": 0.1656,
|
| 1082 |
+
"step": 1730
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 3.9,
|
| 1086 |
+
"learning_rate": 8.163934426229507e-05,
|
| 1087 |
+
"loss": 0.1595,
|
| 1088 |
+
"step": 1740
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 3.92,
|
| 1092 |
+
"learning_rate": 7.999999999999999e-05,
|
| 1093 |
+
"loss": 0.1563,
|
| 1094 |
+
"step": 1750
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.95,
|
| 1098 |
+
"learning_rate": 7.836065573770491e-05,
|
| 1099 |
+
"loss": 0.1625,
|
| 1100 |
+
"step": 1760
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 3.97,
|
| 1104 |
+
"learning_rate": 7.672131147540982e-05,
|
| 1105 |
+
"loss": 0.1439,
|
| 1106 |
+
"step": 1770
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.99,
|
| 1110 |
+
"learning_rate": 7.508196721311474e-05,
|
| 1111 |
+
"loss": 0.1593,
|
| 1112 |
+
"step": 1780
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 4.01,
|
| 1116 |
+
"learning_rate": 7.344262295081966e-05,
|
| 1117 |
+
"loss": 0.1042,
|
| 1118 |
+
"step": 1790
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 4.04,
|
| 1122 |
+
"learning_rate": 7.180327868852459e-05,
|
| 1123 |
+
"loss": 0.0931,
|
| 1124 |
+
"step": 1800
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 4.06,
|
| 1128 |
+
"learning_rate": 7.01639344262295e-05,
|
| 1129 |
+
"loss": 0.0909,
|
| 1130 |
+
"step": 1810
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 4.08,
|
| 1134 |
+
"learning_rate": 6.852459016393443e-05,
|
| 1135 |
+
"loss": 0.0955,
|
| 1136 |
+
"step": 1820
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 4.1,
|
| 1140 |
+
"learning_rate": 6.688524590163934e-05,
|
| 1141 |
+
"loss": 0.102,
|
| 1142 |
+
"step": 1830
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 4.13,
|
| 1146 |
+
"learning_rate": 6.524590163934426e-05,
|
| 1147 |
+
"loss": 0.0889,
|
| 1148 |
+
"step": 1840
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 4.15,
|
| 1152 |
+
"learning_rate": 6.360655737704918e-05,
|
| 1153 |
+
"loss": 0.0968,
|
| 1154 |
+
"step": 1850
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 4.17,
|
| 1158 |
+
"learning_rate": 6.19672131147541e-05,
|
| 1159 |
+
"loss": 0.0809,
|
| 1160 |
+
"step": 1860
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 4.19,
|
| 1164 |
+
"learning_rate": 6.0327868852459015e-05,
|
| 1165 |
+
"loss": 0.0869,
|
| 1166 |
+
"step": 1870
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 4.22,
|
| 1170 |
+
"learning_rate": 5.868852459016393e-05,
|
| 1171 |
+
"loss": 0.0913,
|
| 1172 |
+
"step": 1880
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 4.24,
|
| 1176 |
+
"learning_rate": 5.7049180327868844e-05,
|
| 1177 |
+
"loss": 0.078,
|
| 1178 |
+
"step": 1890
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 4.26,
|
| 1182 |
+
"learning_rate": 5.5409836065573765e-05,
|
| 1183 |
+
"loss": 0.0924,
|
| 1184 |
+
"step": 1900
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 4.28,
|
| 1188 |
+
"learning_rate": 5.3770491803278686e-05,
|
| 1189 |
+
"loss": 0.0879,
|
| 1190 |
+
"step": 1910
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 4.3,
|
| 1194 |
+
"learning_rate": 5.21311475409836e-05,
|
| 1195 |
+
"loss": 0.0869,
|
| 1196 |
+
"step": 1920
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 4.33,
|
| 1200 |
+
"learning_rate": 5.049180327868852e-05,
|
| 1201 |
+
"loss": 0.0832,
|
| 1202 |
+
"step": 1930
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 4.35,
|
| 1206 |
+
"learning_rate": 4.8852459016393436e-05,
|
| 1207 |
+
"loss": 0.0777,
|
| 1208 |
+
"step": 1940
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 4.37,
|
| 1212 |
+
"learning_rate": 4.721311475409836e-05,
|
| 1213 |
+
"loss": 0.0905,
|
| 1214 |
+
"step": 1950
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 4.39,
|
| 1218 |
+
"learning_rate": 4.557377049180328e-05,
|
| 1219 |
+
"loss": 0.0818,
|
| 1220 |
+
"step": 1960
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 4.42,
|
| 1224 |
+
"learning_rate": 4.3934426229508194e-05,
|
| 1225 |
+
"loss": 0.0829,
|
| 1226 |
+
"step": 1970
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 4.44,
|
| 1230 |
+
"learning_rate": 4.229508196721311e-05,
|
| 1231 |
+
"loss": 0.0865,
|
| 1232 |
+
"step": 1980
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 4.46,
|
| 1236 |
+
"learning_rate": 4.065573770491803e-05,
|
| 1237 |
+
"loss": 0.0895,
|
| 1238 |
+
"step": 1990
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 4.48,
|
| 1242 |
+
"learning_rate": 3.901639344262295e-05,
|
| 1243 |
+
"loss": 0.089,
|
| 1244 |
+
"step": 2000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 4.48,
|
| 1248 |
+
"eval_loss": 0.41652488708496094,
|
| 1249 |
+
"eval_runtime": 61.3777,
|
| 1250 |
+
"eval_samples_per_second": 43.045,
|
| 1251 |
+
"eval_steps_per_second": 0.684,
|
| 1252 |
+
"eval_wer": 0.11800872669575566,
|
| 1253 |
+
"step": 2000
|
| 1254 |
+
}
|
| 1255 |
+
],
|
| 1256 |
+
"max_steps": 2230,
|
| 1257 |
+
"num_train_epochs": 5,
|
| 1258 |
+
"total_flos": 2.644259324881902e+19,
|
| 1259 |
+
"trial_name": null,
|
| 1260 |
+
"trial_params": null
|
| 1261 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c029da8afe9a48537ce0d99ab7e8b3fe26f7bf6fda6f49c143ff0da9917a35f4
|
| 3 |
+
size 3055
|
config.json
ADDED
|
@@ -0,0 +1,291 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SpeechEncoderDecoderModel"
|
| 5 |
+
],
|
| 6 |
+
"decoder": {
|
| 7 |
+
"_name_or_path": "facebook/bart-base",
|
| 8 |
+
"activation_dropout": 0.1,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"add_bias_logits": false,
|
| 11 |
+
"add_cross_attention": true,
|
| 12 |
+
"add_final_layer_norm": false,
|
| 13 |
+
"architectures": [
|
| 14 |
+
"BartModel"
|
| 15 |
+
],
|
| 16 |
+
"attention_dropout": 0.1,
|
| 17 |
+
"bad_words_ids": null,
|
| 18 |
+
"bos_token_id": 0,
|
| 19 |
+
"chunk_size_feed_forward": 0,
|
| 20 |
+
"classif_dropout": 0.1,
|
| 21 |
+
"classifier_dropout": 0.0,
|
| 22 |
+
"cross_attention_hidden_size": null,
|
| 23 |
+
"d_model": 768,
|
| 24 |
+
"decoder_attention_heads": 12,
|
| 25 |
+
"decoder_ffn_dim": 3072,
|
| 26 |
+
"decoder_layerdrop": 0.0,
|
| 27 |
+
"decoder_layers": 6,
|
| 28 |
+
"decoder_start_token_id": 2,
|
| 29 |
+
"diversity_penalty": 0.0,
|
| 30 |
+
"do_sample": false,
|
| 31 |
+
"dropout": 0.1,
|
| 32 |
+
"early_stopping": true,
|
| 33 |
+
"encoder_attention_heads": 12,
|
| 34 |
+
"encoder_ffn_dim": 3072,
|
| 35 |
+
"encoder_layerdrop": 0.0,
|
| 36 |
+
"encoder_layers": 6,
|
| 37 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 38 |
+
"eos_token_id": 2,
|
| 39 |
+
"finetuning_task": null,
|
| 40 |
+
"forced_bos_token_id": null,
|
| 41 |
+
"forced_eos_token_id": 2,
|
| 42 |
+
"gradient_checkpointing": false,
|
| 43 |
+
"id2label": {
|
| 44 |
+
"0": "LABEL_0",
|
| 45 |
+
"1": "LABEL_1",
|
| 46 |
+
"2": "LABEL_2"
|
| 47 |
+
},
|
| 48 |
+
"init_std": 0.02,
|
| 49 |
+
"is_decoder": true,
|
| 50 |
+
"is_encoder_decoder": false,
|
| 51 |
+
"label2id": {
|
| 52 |
+
"LABEL_0": 0,
|
| 53 |
+
"LABEL_1": 1,
|
| 54 |
+
"LABEL_2": 2
|
| 55 |
+
},
|
| 56 |
+
"length_penalty": 1.0,
|
| 57 |
+
"max_length": 20,
|
| 58 |
+
"max_position_embeddings": 1024,
|
| 59 |
+
"min_length": 0,
|
| 60 |
+
"model_type": "bart",
|
| 61 |
+
"no_repeat_ngram_size": 3,
|
| 62 |
+
"normalize_before": false,
|
| 63 |
+
"normalize_embedding": true,
|
| 64 |
+
"num_beam_groups": 1,
|
| 65 |
+
"num_beams": 4,
|
| 66 |
+
"num_hidden_layers": 6,
|
| 67 |
+
"num_return_sequences": 1,
|
| 68 |
+
"output_attentions": false,
|
| 69 |
+
"output_hidden_states": false,
|
| 70 |
+
"output_scores": false,
|
| 71 |
+
"pad_token_id": 1,
|
| 72 |
+
"prefix": null,
|
| 73 |
+
"problem_type": null,
|
| 74 |
+
"pruned_heads": {},
|
| 75 |
+
"remove_invalid_values": false,
|
| 76 |
+
"repetition_penalty": 1.0,
|
| 77 |
+
"return_dict": true,
|
| 78 |
+
"return_dict_in_generate": false,
|
| 79 |
+
"scale_embedding": false,
|
| 80 |
+
"sep_token_id": null,
|
| 81 |
+
"task_specific_params": {
|
| 82 |
+
"summarization": {
|
| 83 |
+
"length_penalty": 1.0,
|
| 84 |
+
"max_length": 128,
|
| 85 |
+
"min_length": 12,
|
| 86 |
+
"num_beams": 4
|
| 87 |
+
},
|
| 88 |
+
"summarization_cnn": {
|
| 89 |
+
"length_penalty": 2.0,
|
| 90 |
+
"max_length": 142,
|
| 91 |
+
"min_length": 56,
|
| 92 |
+
"num_beams": 4
|
| 93 |
+
},
|
| 94 |
+
"summarization_xsum": {
|
| 95 |
+
"length_penalty": 1.0,
|
| 96 |
+
"max_length": 62,
|
| 97 |
+
"min_length": 11,
|
| 98 |
+
"num_beams": 6
|
| 99 |
+
}
|
| 100 |
+
},
|
| 101 |
+
"temperature": 1.0,
|
| 102 |
+
"tie_encoder_decoder": false,
|
| 103 |
+
"tie_word_embeddings": true,
|
| 104 |
+
"tokenizer_class": null,
|
| 105 |
+
"top_k": 50,
|
| 106 |
+
"top_p": 1.0,
|
| 107 |
+
"torch_dtype": "float32",
|
| 108 |
+
"torchscript": false,
|
| 109 |
+
"transformers_version": "4.15.0.dev0",
|
| 110 |
+
"use_bfloat16": false,
|
| 111 |
+
"use_cache": true,
|
| 112 |
+
"vocab_size": 50265
|
| 113 |
+
},
|
| 114 |
+
"decoder_start_token_id": 0,
|
| 115 |
+
"encoder": {
|
| 116 |
+
"_name_or_path": "facebook/wav2vec2-base",
|
| 117 |
+
"activation_dropout": 0.0,
|
| 118 |
+
"adapter_kernel_size": 3,
|
| 119 |
+
"adapter_stride": 2,
|
| 120 |
+
"add_adapter": false,
|
| 121 |
+
"add_cross_attention": false,
|
| 122 |
+
"apply_spec_augment": true,
|
| 123 |
+
"architectures": [
|
| 124 |
+
"Wav2Vec2ForPreTraining"
|
| 125 |
+
],
|
| 126 |
+
"attention_dropout": 0.1,
|
| 127 |
+
"bad_words_ids": null,
|
| 128 |
+
"bos_token_id": 1,
|
| 129 |
+
"chunk_size_feed_forward": 0,
|
| 130 |
+
"classifier_proj_size": 256,
|
| 131 |
+
"codevector_dim": 256,
|
| 132 |
+
"contrastive_logits_temperature": 0.1,
|
| 133 |
+
"conv_bias": false,
|
| 134 |
+
"conv_dim": [
|
| 135 |
+
512,
|
| 136 |
+
512,
|
| 137 |
+
512,
|
| 138 |
+
512,
|
| 139 |
+
512,
|
| 140 |
+
512,
|
| 141 |
+
512
|
| 142 |
+
],
|
| 143 |
+
"conv_kernel": [
|
| 144 |
+
10,
|
| 145 |
+
3,
|
| 146 |
+
3,
|
| 147 |
+
3,
|
| 148 |
+
3,
|
| 149 |
+
2,
|
| 150 |
+
2
|
| 151 |
+
],
|
| 152 |
+
"conv_stride": [
|
| 153 |
+
5,
|
| 154 |
+
2,
|
| 155 |
+
2,
|
| 156 |
+
2,
|
| 157 |
+
2,
|
| 158 |
+
2,
|
| 159 |
+
2
|
| 160 |
+
],
|
| 161 |
+
"cross_attention_hidden_size": null,
|
| 162 |
+
"ctc_loss_reduction": "sum",
|
| 163 |
+
"ctc_zero_infinity": false,
|
| 164 |
+
"decoder_start_token_id": null,
|
| 165 |
+
"diversity_loss_weight": 0.1,
|
| 166 |
+
"diversity_penalty": 0.0,
|
| 167 |
+
"do_sample": false,
|
| 168 |
+
"do_stable_layer_norm": false,
|
| 169 |
+
"early_stopping": false,
|
| 170 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 171 |
+
"eos_token_id": 2,
|
| 172 |
+
"feat_extract_activation": "gelu",
|
| 173 |
+
"feat_extract_norm": "group",
|
| 174 |
+
"feat_proj_dropout": 0.0,
|
| 175 |
+
"feat_quantizer_dropout": 0.0,
|
| 176 |
+
"final_dropout": 0.0,
|
| 177 |
+
"finetuning_task": null,
|
| 178 |
+
"forced_bos_token_id": null,
|
| 179 |
+
"forced_eos_token_id": null,
|
| 180 |
+
"freeze_feat_extract_train": true,
|
| 181 |
+
"hidden_act": "gelu",
|
| 182 |
+
"hidden_dropout": 0.1,
|
| 183 |
+
"hidden_size": 768,
|
| 184 |
+
"id2label": {
|
| 185 |
+
"0": "LABEL_0",
|
| 186 |
+
"1": "LABEL_1"
|
| 187 |
+
},
|
| 188 |
+
"initializer_range": 0.02,
|
| 189 |
+
"intermediate_size": 3072,
|
| 190 |
+
"is_decoder": false,
|
| 191 |
+
"is_encoder_decoder": false,
|
| 192 |
+
"label2id": {
|
| 193 |
+
"LABEL_0": 0,
|
| 194 |
+
"LABEL_1": 1
|
| 195 |
+
},
|
| 196 |
+
"layer_norm_eps": 1e-05,
|
| 197 |
+
"layerdrop": 0.0,
|
| 198 |
+
"length_penalty": 1.0,
|
| 199 |
+
"mask_channel_length": 10,
|
| 200 |
+
"mask_channel_min_space": 1,
|
| 201 |
+
"mask_channel_other": 0.0,
|
| 202 |
+
"mask_channel_prob": 0.0,
|
| 203 |
+
"mask_channel_selection": "static",
|
| 204 |
+
"mask_feature_length": 10,
|
| 205 |
+
"mask_feature_min_masks": 0,
|
| 206 |
+
"mask_feature_prob": 0.0,
|
| 207 |
+
"mask_time_length": 10,
|
| 208 |
+
"mask_time_min_masks": 2,
|
| 209 |
+
"mask_time_min_space": 1,
|
| 210 |
+
"mask_time_other": 0.0,
|
| 211 |
+
"mask_time_prob": 0.0,
|
| 212 |
+
"mask_time_selection": "static",
|
| 213 |
+
"max_length": 20,
|
| 214 |
+
"min_length": 0,
|
| 215 |
+
"model_type": "wav2vec2",
|
| 216 |
+
"no_mask_channel_overlap": false,
|
| 217 |
+
"no_mask_time_overlap": false,
|
| 218 |
+
"no_repeat_ngram_size": 0,
|
| 219 |
+
"num_adapter_layers": 3,
|
| 220 |
+
"num_attention_heads": 12,
|
| 221 |
+
"num_beam_groups": 1,
|
| 222 |
+
"num_beams": 1,
|
| 223 |
+
"num_codevector_groups": 2,
|
| 224 |
+
"num_codevectors_per_group": 320,
|
| 225 |
+
"num_conv_pos_embedding_groups": 16,
|
| 226 |
+
"num_conv_pos_embeddings": 128,
|
| 227 |
+
"num_feat_extract_layers": 7,
|
| 228 |
+
"num_hidden_layers": 12,
|
| 229 |
+
"num_negatives": 100,
|
| 230 |
+
"num_return_sequences": 1,
|
| 231 |
+
"output_attentions": false,
|
| 232 |
+
"output_hidden_size": 768,
|
| 233 |
+
"output_hidden_states": false,
|
| 234 |
+
"output_scores": false,
|
| 235 |
+
"pad_token_id": 0,
|
| 236 |
+
"prefix": null,
|
| 237 |
+
"problem_type": null,
|
| 238 |
+
"proj_codevector_dim": 256,
|
| 239 |
+
"pruned_heads": {},
|
| 240 |
+
"remove_invalid_values": false,
|
| 241 |
+
"repetition_penalty": 1.0,
|
| 242 |
+
"return_dict": true,
|
| 243 |
+
"return_dict_in_generate": false,
|
| 244 |
+
"sep_token_id": null,
|
| 245 |
+
"task_specific_params": null,
|
| 246 |
+
"tdnn_dilation": [
|
| 247 |
+
1,
|
| 248 |
+
2,
|
| 249 |
+
3,
|
| 250 |
+
1,
|
| 251 |
+
1
|
| 252 |
+
],
|
| 253 |
+
"tdnn_dim": [
|
| 254 |
+
512,
|
| 255 |
+
512,
|
| 256 |
+
512,
|
| 257 |
+
512,
|
| 258 |
+
1500
|
| 259 |
+
],
|
| 260 |
+
"tdnn_kernel": [
|
| 261 |
+
5,
|
| 262 |
+
3,
|
| 263 |
+
3,
|
| 264 |
+
1,
|
| 265 |
+
1
|
| 266 |
+
],
|
| 267 |
+
"temperature": 1.0,
|
| 268 |
+
"tie_encoder_decoder": false,
|
| 269 |
+
"tie_word_embeddings": true,
|
| 270 |
+
"tokenizer_class": null,
|
| 271 |
+
"top_k": 50,
|
| 272 |
+
"top_p": 1.0,
|
| 273 |
+
"torch_dtype": null,
|
| 274 |
+
"torchscript": false,
|
| 275 |
+
"transformers_version": "4.15.0.dev0",
|
| 276 |
+
"use_bfloat16": false,
|
| 277 |
+
"use_weighted_layer_sum": false,
|
| 278 |
+
"vocab_size": 32,
|
| 279 |
+
"xvector_output_dim": 512
|
| 280 |
+
},
|
| 281 |
+
"eos_token_id": 2,
|
| 282 |
+
"is_encoder_decoder": true,
|
| 283 |
+
"max_length": 40,
|
| 284 |
+
"model_type": "speech-encoder-decoder",
|
| 285 |
+
"pad_token_id": 1,
|
| 286 |
+
"processor_class": "Wav2Vec2Processor",
|
| 287 |
+
"tie_word_embeddings": false,
|
| 288 |
+
"torch_dtype": "float32",
|
| 289 |
+
"transformers_version": null,
|
| 290 |
+
"use_cache": false
|
| 291 |
+
}
|
create_model.py
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python3
|
| 2 |
+
from transformers import SpeechEncoderDecoderModel, AutoFeatureExtractor, AutoTokenizer
|
| 3 |
+
import torch
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
encoder_id = "facebook/wav2vec2-base"
|
| 7 |
+
decoder_id = "facebook/bart-base"
|
| 8 |
+
|
| 9 |
+
model = SpeechEncoderDecoderModel.from_encoder_decoder_pretrained(encoder_id, decoder_id, encoder_add_adapter=True)
|
| 10 |
+
model.config.encoder.feat_proj_dropout = 0.0
|
| 11 |
+
model.config.encoder.mask_time_prob = 0.0
|
| 12 |
+
model.config.decoder_start_token_id = model.decoder.config.bos_token_id
|
| 13 |
+
model.config.pad_token_id = model.decoder.config.pad_token_id
|
| 14 |
+
model.config.eos_token_id = model.decoder.config.eos_token_id
|
| 15 |
+
model.config.max_length = 40
|
| 16 |
+
model.config.encoder.layerdrop = 0.0
|
| 17 |
+
model.config.use_cache = False
|
| 18 |
+
model.config.processor_class = "Wav2Vec2Processor"
|
| 19 |
+
|
| 20 |
+
# check if generation works
|
| 21 |
+
out = model.generate(torch.ones((1, 2000)))
|
| 22 |
+
|
| 23 |
+
model.save_pretrained("./")
|
| 24 |
+
|
| 25 |
+
feature_etxractor = AutoFeatureExtractor.from_pretrained(encoder_id)
|
| 26 |
+
feature_etxractor.save_pretrained("./")
|
| 27 |
+
tokenizer = AutoTokenizer.from_pretrained(decoder_id)
|
| 28 |
+
tokenizer.save_pretrained("./")
|
emissions.csv
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
| 1 |
+
timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
|
| 2 |
+
2021-12-22T12:36:27,6d18c71e-f1b0-4f2e-b8a5-07165da045c2,codecarbon,3187.4341790676117,0.42607577097711286,1.1584441842770876,USA,USA,nan,Y,aws,us-east-1
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 0.40512892603874207,
|
| 4 |
+
"eval_runtime": 60.9636,
|
| 5 |
+
"eval_samples": 2642,
|
| 6 |
+
"eval_samples_per_second": 43.337,
|
| 7 |
+
"eval_steps_per_second": 0.689,
|
| 8 |
+
"eval_wer": 0.11511305037683459
|
| 9 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4e7e4501561b66afbcb7f00bc6d95a9410c7dd4f7270d2d3ed5cc7b44d805e1
|
| 3 |
+
size 762044061
|
run_librispeech.sh
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env bash
|
| 2 |
+
#CUDA_VISIBLE_DEVICES="0" python run_speech_recognition_seq2seq.py \
|
| 3 |
+
python -m torch.distributed.launch \
|
| 4 |
+
--nproc_per_node 8 run_speech_recognition_seq2seq.py \
|
| 5 |
+
--dataset_name="librispeech_asr" \
|
| 6 |
+
--model_name_or_path="./" \
|
| 7 |
+
--dataset_config_name="clean" \
|
| 8 |
+
--train_split_name="train.100" \
|
| 9 |
+
--eval_split_name="validation" \
|
| 10 |
+
--output_dir="./" \
|
| 11 |
+
--preprocessing_num_workers="16" \
|
| 12 |
+
--length_column_name="input_length" \
|
| 13 |
+
--overwrite_output_dir \
|
| 14 |
+
--num_train_epochs="5" \
|
| 15 |
+
--per_device_train_batch_size="8" \
|
| 16 |
+
--per_device_eval_batch_size="8" \
|
| 17 |
+
--gradient_accumulation_steps="1" \
|
| 18 |
+
--learning_rate="3e-4" \
|
| 19 |
+
--warmup_steps="400" \
|
| 20 |
+
--evaluation_strategy="steps" \
|
| 21 |
+
--text_column_name="text" \
|
| 22 |
+
--save_steps="400" \
|
| 23 |
+
--eval_steps="400" \
|
| 24 |
+
--logging_steps="10" \
|
| 25 |
+
--save_total_limit="1" \
|
| 26 |
+
--freeze_feature_extractor \
|
| 27 |
+
--gradient_checkpointing \
|
| 28 |
+
--fp16 \
|
| 29 |
+
--group_by_length \
|
| 30 |
+
--predict_with_generate \
|
| 31 |
+
--do_train --do_eval
|
run_speech_recognition_seq2seq.py
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
/home/ubuntu/transformers/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py
|
runs/Dec22_11-42-50_ip-172-31-1-149/1640173397.2128675/events.out.tfevents.1640173397.ip-172-31-1-149.489657.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a3867d1f1945a775f70109031c11090a9837942f56d9aaf362e2f4d6c219151
|
| 3 |
+
size 4951
|
runs/Dec22_11-42-50_ip-172-31-1-149/events.out.tfevents.1640173397.ip-172-31-1-149.489657.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d5b9e5988021325cb2505e47ee82877c9669c63ec7492cde700493e59810cb0
|
| 3 |
+
size 46811
|
runs/Dec22_11-42-50_ip-172-31-1-149/events.out.tfevents.1640176651.ip-172-31-1-149.489657.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34cf8407b5897ebc16b68ccadd37b5c8e85970ade73f2d51944f395187b0f5ac
|
| 3 |
+
size 358
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "./", "tokenizer_class": "BartTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.5071858802451146,
|
| 4 |
+
"train_runtime": 3191.5244,
|
| 5 |
+
"train_samples": 28538,
|
| 6 |
+
"train_samples_per_second": 44.709,
|
| 7 |
+
"train_steps_per_second": 0.699
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1408 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 2230,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 2.2499999999999996e-06,
|
| 13 |
+
"loss": 11.7449,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 9.75e-06,
|
| 19 |
+
"loss": 9.9528,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 1.725e-05,
|
| 25 |
+
"loss": 7.8835,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"learning_rate": 2.475e-05,
|
| 31 |
+
"loss": 6.9463,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"learning_rate": 3.225e-05,
|
| 37 |
+
"loss": 6.3116,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.13,
|
| 42 |
+
"learning_rate": 3.975e-05,
|
| 43 |
+
"loss": 5.8418,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.16,
|
| 48 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 49 |
+
"loss": 5.6196,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.18,
|
| 54 |
+
"learning_rate": 5.4749999999999996e-05,
|
| 55 |
+
"loss": 5.4065,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.2,
|
| 60 |
+
"learning_rate": 6.225e-05,
|
| 61 |
+
"loss": 5.3217,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.22,
|
| 66 |
+
"learning_rate": 6.975e-05,
|
| 67 |
+
"loss": 5.2094,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.25,
|
| 72 |
+
"learning_rate": 7.725e-05,
|
| 73 |
+
"loss": 5.2388,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.27,
|
| 78 |
+
"learning_rate": 8.474999999999999e-05,
|
| 79 |
+
"loss": 5.1522,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.29,
|
| 84 |
+
"learning_rate": 9.224999999999999e-05,
|
| 85 |
+
"loss": 5.1679,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31,
|
| 90 |
+
"learning_rate": 9.975e-05,
|
| 91 |
+
"loss": 5.112,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.34,
|
| 96 |
+
"learning_rate": 0.00010724999999999999,
|
| 97 |
+
"loss": 5.1008,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.36,
|
| 102 |
+
"learning_rate": 0.00011475,
|
| 103 |
+
"loss": 5.1024,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.38,
|
| 108 |
+
"learning_rate": 0.00012225,
|
| 109 |
+
"loss": 5.0238,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.4,
|
| 114 |
+
"learning_rate": 0.00012974999999999998,
|
| 115 |
+
"loss": 5.061,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.43,
|
| 120 |
+
"learning_rate": 0.00013725,
|
| 121 |
+
"loss": 5.0293,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.45,
|
| 126 |
+
"learning_rate": 0.00014475,
|
| 127 |
+
"loss": 5.0315,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.47,
|
| 132 |
+
"learning_rate": 0.00015224999999999996,
|
| 133 |
+
"loss": 5.1264,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.49,
|
| 138 |
+
"learning_rate": 0.00015974999999999998,
|
| 139 |
+
"loss": 4.9663,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.52,
|
| 144 |
+
"learning_rate": 0.00016724999999999997,
|
| 145 |
+
"loss": 4.9525,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.54,
|
| 150 |
+
"learning_rate": 0.00017475,
|
| 151 |
+
"loss": 4.9757,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.56,
|
| 156 |
+
"learning_rate": 0.00018224999999999998,
|
| 157 |
+
"loss": 4.9203,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.58,
|
| 162 |
+
"learning_rate": 0.00018974999999999998,
|
| 163 |
+
"loss": 4.908,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.61,
|
| 168 |
+
"learning_rate": 0.00019724999999999997,
|
| 169 |
+
"loss": 4.845,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.63,
|
| 174 |
+
"learning_rate": 0.00020475,
|
| 175 |
+
"loss": 4.8367,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.65,
|
| 180 |
+
"learning_rate": 0.00021224999999999998,
|
| 181 |
+
"loss": 4.8034,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.67,
|
| 186 |
+
"learning_rate": 0.00021975,
|
| 187 |
+
"loss": 4.8215,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.7,
|
| 192 |
+
"learning_rate": 0.00022724999999999997,
|
| 193 |
+
"loss": 4.7776,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.72,
|
| 198 |
+
"learning_rate": 0.00023474999999999996,
|
| 199 |
+
"loss": 4.7166,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.74,
|
| 204 |
+
"learning_rate": 0.00024224999999999998,
|
| 205 |
+
"loss": 4.7696,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.76,
|
| 210 |
+
"learning_rate": 0.000249,
|
| 211 |
+
"loss": 4.6817,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.78,
|
| 216 |
+
"learning_rate": 0.00025649999999999995,
|
| 217 |
+
"loss": 4.7147,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.81,
|
| 222 |
+
"learning_rate": 0.00026399999999999997,
|
| 223 |
+
"loss": 4.6688,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.83,
|
| 228 |
+
"learning_rate": 0.0002715,
|
| 229 |
+
"loss": 4.5991,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.85,
|
| 234 |
+
"learning_rate": 0.000279,
|
| 235 |
+
"loss": 4.6401,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.87,
|
| 240 |
+
"learning_rate": 0.00028649999999999997,
|
| 241 |
+
"loss": 4.6029,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.9,
|
| 246 |
+
"learning_rate": 0.000294,
|
| 247 |
+
"loss": 4.5118,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.9,
|
| 252 |
+
"eval_loss": 4.247889041900635,
|
| 253 |
+
"eval_runtime": 51.3593,
|
| 254 |
+
"eval_samples_per_second": 51.442,
|
| 255 |
+
"eval_steps_per_second": 0.818,
|
| 256 |
+
"eval_wer": 0.9033518445061484,
|
| 257 |
+
"step": 400
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.92,
|
| 261 |
+
"learning_rate": 0.000299672131147541,
|
| 262 |
+
"loss": 4.5483,
|
| 263 |
+
"step": 410
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.94,
|
| 267 |
+
"learning_rate": 0.00029803278688524587,
|
| 268 |
+
"loss": 4.4005,
|
| 269 |
+
"step": 420
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.96,
|
| 273 |
+
"learning_rate": 0.00029639344262295076,
|
| 274 |
+
"loss": 4.327,
|
| 275 |
+
"step": 430
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.99,
|
| 279 |
+
"learning_rate": 0.0002947540983606557,
|
| 280 |
+
"loss": 4.2717,
|
| 281 |
+
"step": 440
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.01,
|
| 285 |
+
"learning_rate": 0.0002931147540983606,
|
| 286 |
+
"loss": 4.0139,
|
| 287 |
+
"step": 450
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.03,
|
| 291 |
+
"learning_rate": 0.00029147540983606555,
|
| 292 |
+
"loss": 3.8304,
|
| 293 |
+
"step": 460
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.05,
|
| 297 |
+
"learning_rate": 0.0002898360655737705,
|
| 298 |
+
"loss": 3.5825,
|
| 299 |
+
"step": 470
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.08,
|
| 303 |
+
"learning_rate": 0.0002881967213114754,
|
| 304 |
+
"loss": 3.4968,
|
| 305 |
+
"step": 480
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.1,
|
| 309 |
+
"learning_rate": 0.0002865573770491803,
|
| 310 |
+
"loss": 3.2832,
|
| 311 |
+
"step": 490
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.12,
|
| 315 |
+
"learning_rate": 0.00028491803278688524,
|
| 316 |
+
"loss": 3.1368,
|
| 317 |
+
"step": 500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.14,
|
| 321 |
+
"learning_rate": 0.00028327868852459013,
|
| 322 |
+
"loss": 2.899,
|
| 323 |
+
"step": 510
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.17,
|
| 327 |
+
"learning_rate": 0.0002816393442622951,
|
| 328 |
+
"loss": 2.6293,
|
| 329 |
+
"step": 520
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.19,
|
| 333 |
+
"learning_rate": 0.00028,
|
| 334 |
+
"loss": 2.7137,
|
| 335 |
+
"step": 530
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.21,
|
| 339 |
+
"learning_rate": 0.00027836065573770487,
|
| 340 |
+
"loss": 2.4069,
|
| 341 |
+
"step": 540
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.23,
|
| 345 |
+
"learning_rate": 0.0002767213114754098,
|
| 346 |
+
"loss": 2.3349,
|
| 347 |
+
"step": 550
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.26,
|
| 351 |
+
"learning_rate": 0.0002750819672131147,
|
| 352 |
+
"loss": 2.1064,
|
| 353 |
+
"step": 560
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.28,
|
| 357 |
+
"learning_rate": 0.00027344262295081966,
|
| 358 |
+
"loss": 1.9375,
|
| 359 |
+
"step": 570
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.3,
|
| 363 |
+
"learning_rate": 0.00027180327868852455,
|
| 364 |
+
"loss": 1.9395,
|
| 365 |
+
"step": 580
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.32,
|
| 369 |
+
"learning_rate": 0.0002701639344262295,
|
| 370 |
+
"loss": 1.7861,
|
| 371 |
+
"step": 590
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 1.35,
|
| 375 |
+
"learning_rate": 0.0002685245901639344,
|
| 376 |
+
"loss": 1.6795,
|
| 377 |
+
"step": 600
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.37,
|
| 381 |
+
"learning_rate": 0.0002668852459016393,
|
| 382 |
+
"loss": 1.5166,
|
| 383 |
+
"step": 610
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.39,
|
| 387 |
+
"learning_rate": 0.00026524590163934424,
|
| 388 |
+
"loss": 1.4776,
|
| 389 |
+
"step": 620
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.41,
|
| 393 |
+
"learning_rate": 0.00026360655737704913,
|
| 394 |
+
"loss": 1.4286,
|
| 395 |
+
"step": 630
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.43,
|
| 399 |
+
"learning_rate": 0.0002619672131147541,
|
| 400 |
+
"loss": 1.3071,
|
| 401 |
+
"step": 640
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.46,
|
| 405 |
+
"learning_rate": 0.00026032786885245903,
|
| 406 |
+
"loss": 1.277,
|
| 407 |
+
"step": 650
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.48,
|
| 411 |
+
"learning_rate": 0.0002586885245901639,
|
| 412 |
+
"loss": 1.2741,
|
| 413 |
+
"step": 660
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.5,
|
| 417 |
+
"learning_rate": 0.0002570491803278688,
|
| 418 |
+
"loss": 1.1661,
|
| 419 |
+
"step": 670
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.52,
|
| 423 |
+
"learning_rate": 0.00025540983606557377,
|
| 424 |
+
"loss": 1.1585,
|
| 425 |
+
"step": 680
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.55,
|
| 429 |
+
"learning_rate": 0.00025377049180327866,
|
| 430 |
+
"loss": 1.1015,
|
| 431 |
+
"step": 690
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.57,
|
| 435 |
+
"learning_rate": 0.0002521311475409836,
|
| 436 |
+
"loss": 1.0403,
|
| 437 |
+
"step": 700
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.59,
|
| 441 |
+
"learning_rate": 0.0002504918032786885,
|
| 442 |
+
"loss": 1.0001,
|
| 443 |
+
"step": 710
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.61,
|
| 447 |
+
"learning_rate": 0.0002488524590163934,
|
| 448 |
+
"loss": 0.9915,
|
| 449 |
+
"step": 720
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.64,
|
| 453 |
+
"learning_rate": 0.00024721311475409835,
|
| 454 |
+
"loss": 0.9481,
|
| 455 |
+
"step": 730
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.66,
|
| 459 |
+
"learning_rate": 0.00024557377049180324,
|
| 460 |
+
"loss": 0.9522,
|
| 461 |
+
"step": 740
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 1.68,
|
| 465 |
+
"learning_rate": 0.00024393442622950816,
|
| 466 |
+
"loss": 0.9063,
|
| 467 |
+
"step": 750
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 1.7,
|
| 471 |
+
"learning_rate": 0.00024229508196721308,
|
| 472 |
+
"loss": 0.8868,
|
| 473 |
+
"step": 760
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.73,
|
| 477 |
+
"learning_rate": 0.000240655737704918,
|
| 478 |
+
"loss": 0.818,
|
| 479 |
+
"step": 770
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.75,
|
| 483 |
+
"learning_rate": 0.00023901639344262293,
|
| 484 |
+
"loss": 0.8436,
|
| 485 |
+
"step": 780
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.77,
|
| 489 |
+
"learning_rate": 0.00023737704918032785,
|
| 490 |
+
"loss": 0.7755,
|
| 491 |
+
"step": 790
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.79,
|
| 495 |
+
"learning_rate": 0.00023573770491803274,
|
| 496 |
+
"loss": 0.8016,
|
| 497 |
+
"step": 800
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.79,
|
| 501 |
+
"eval_loss": 0.8116478323936462,
|
| 502 |
+
"eval_runtime": 61.0806,
|
| 503 |
+
"eval_samples_per_second": 43.254,
|
| 504 |
+
"eval_steps_per_second": 0.688,
|
| 505 |
+
"eval_wer": 0.2206862356207854,
|
| 506 |
+
"step": 800
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.82,
|
| 510 |
+
"learning_rate": 0.00023409836065573766,
|
| 511 |
+
"loss": 0.7623,
|
| 512 |
+
"step": 810
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.84,
|
| 516 |
+
"learning_rate": 0.00023245901639344258,
|
| 517 |
+
"loss": 0.7884,
|
| 518 |
+
"step": 820
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 1.86,
|
| 522 |
+
"learning_rate": 0.00023081967213114753,
|
| 523 |
+
"loss": 0.7791,
|
| 524 |
+
"step": 830
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 1.88,
|
| 528 |
+
"learning_rate": 0.00022918032786885245,
|
| 529 |
+
"loss": 0.7111,
|
| 530 |
+
"step": 840
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 1.91,
|
| 534 |
+
"learning_rate": 0.00022754098360655738,
|
| 535 |
+
"loss": 0.7063,
|
| 536 |
+
"step": 850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.93,
|
| 540 |
+
"learning_rate": 0.0002259016393442623,
|
| 541 |
+
"loss": 0.7307,
|
| 542 |
+
"step": 860
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.95,
|
| 546 |
+
"learning_rate": 0.0002242622950819672,
|
| 547 |
+
"loss": 0.6792,
|
| 548 |
+
"step": 870
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.97,
|
| 552 |
+
"learning_rate": 0.0002226229508196721,
|
| 553 |
+
"loss": 0.6496,
|
| 554 |
+
"step": 880
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.0,
|
| 558 |
+
"learning_rate": 0.00022098360655737703,
|
| 559 |
+
"loss": 0.6929,
|
| 560 |
+
"step": 890
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 2.02,
|
| 564 |
+
"learning_rate": 0.00021934426229508195,
|
| 565 |
+
"loss": 0.5503,
|
| 566 |
+
"step": 900
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 2.04,
|
| 570 |
+
"learning_rate": 0.00021770491803278688,
|
| 571 |
+
"loss": 0.4833,
|
| 572 |
+
"step": 910
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 2.06,
|
| 576 |
+
"learning_rate": 0.0002160655737704918,
|
| 577 |
+
"loss": 0.5107,
|
| 578 |
+
"step": 920
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.09,
|
| 582 |
+
"learning_rate": 0.0002144262295081967,
|
| 583 |
+
"loss": 0.4637,
|
| 584 |
+
"step": 930
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.11,
|
| 588 |
+
"learning_rate": 0.0002127868852459016,
|
| 589 |
+
"loss": 0.4764,
|
| 590 |
+
"step": 940
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.13,
|
| 594 |
+
"learning_rate": 0.00021114754098360653,
|
| 595 |
+
"loss": 0.4332,
|
| 596 |
+
"step": 950
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.15,
|
| 600 |
+
"learning_rate": 0.00020950819672131146,
|
| 601 |
+
"loss": 0.4599,
|
| 602 |
+
"step": 960
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 2.17,
|
| 606 |
+
"learning_rate": 0.00020786885245901638,
|
| 607 |
+
"loss": 0.4582,
|
| 608 |
+
"step": 970
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.2,
|
| 612 |
+
"learning_rate": 0.00020622950819672127,
|
| 613 |
+
"loss": 0.4112,
|
| 614 |
+
"step": 980
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.22,
|
| 618 |
+
"learning_rate": 0.0002045901639344262,
|
| 619 |
+
"loss": 0.4441,
|
| 620 |
+
"step": 990
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.24,
|
| 624 |
+
"learning_rate": 0.0002029508196721311,
|
| 625 |
+
"loss": 0.4575,
|
| 626 |
+
"step": 1000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.26,
|
| 630 |
+
"learning_rate": 0.00020131147540983603,
|
| 631 |
+
"loss": 0.4177,
|
| 632 |
+
"step": 1010
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.29,
|
| 636 |
+
"learning_rate": 0.00019967213114754098,
|
| 637 |
+
"loss": 0.4323,
|
| 638 |
+
"step": 1020
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.31,
|
| 642 |
+
"learning_rate": 0.0001980327868852459,
|
| 643 |
+
"loss": 0.4295,
|
| 644 |
+
"step": 1030
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.33,
|
| 648 |
+
"learning_rate": 0.00019639344262295083,
|
| 649 |
+
"loss": 0.4595,
|
| 650 |
+
"step": 1040
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.35,
|
| 654 |
+
"learning_rate": 0.00019475409836065572,
|
| 655 |
+
"loss": 0.4338,
|
| 656 |
+
"step": 1050
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.38,
|
| 660 |
+
"learning_rate": 0.00019311475409836064,
|
| 661 |
+
"loss": 0.4104,
|
| 662 |
+
"step": 1060
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.4,
|
| 666 |
+
"learning_rate": 0.00019147540983606556,
|
| 667 |
+
"loss": 0.4216,
|
| 668 |
+
"step": 1070
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.42,
|
| 672 |
+
"learning_rate": 0.00018983606557377048,
|
| 673 |
+
"loss": 0.3729,
|
| 674 |
+
"step": 1080
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.44,
|
| 678 |
+
"learning_rate": 0.0001881967213114754,
|
| 679 |
+
"loss": 0.3998,
|
| 680 |
+
"step": 1090
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.47,
|
| 684 |
+
"learning_rate": 0.0001865573770491803,
|
| 685 |
+
"loss": 0.4082,
|
| 686 |
+
"step": 1100
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 2.49,
|
| 690 |
+
"learning_rate": 0.00018491803278688522,
|
| 691 |
+
"loss": 0.38,
|
| 692 |
+
"step": 1110
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.51,
|
| 696 |
+
"learning_rate": 0.00018327868852459014,
|
| 697 |
+
"loss": 0.4157,
|
| 698 |
+
"step": 1120
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 2.53,
|
| 702 |
+
"learning_rate": 0.00018163934426229506,
|
| 703 |
+
"loss": 0.3937,
|
| 704 |
+
"step": 1130
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.56,
|
| 708 |
+
"learning_rate": 0.00017999999999999998,
|
| 709 |
+
"loss": 0.3772,
|
| 710 |
+
"step": 1140
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.58,
|
| 714 |
+
"learning_rate": 0.0001783606557377049,
|
| 715 |
+
"loss": 0.3767,
|
| 716 |
+
"step": 1150
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.6,
|
| 720 |
+
"learning_rate": 0.0001767213114754098,
|
| 721 |
+
"loss": 0.3602,
|
| 722 |
+
"step": 1160
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 2.62,
|
| 726 |
+
"learning_rate": 0.00017508196721311472,
|
| 727 |
+
"loss": 0.3876,
|
| 728 |
+
"step": 1170
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 2.65,
|
| 732 |
+
"learning_rate": 0.00017344262295081964,
|
| 733 |
+
"loss": 0.3776,
|
| 734 |
+
"step": 1180
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.67,
|
| 738 |
+
"learning_rate": 0.00017180327868852456,
|
| 739 |
+
"loss": 0.3617,
|
| 740 |
+
"step": 1190
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 2.69,
|
| 744 |
+
"learning_rate": 0.0001701639344262295,
|
| 745 |
+
"loss": 0.3627,
|
| 746 |
+
"step": 1200
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.69,
|
| 750 |
+
"eval_loss": 0.5665903091430664,
|
| 751 |
+
"eval_runtime": 61.1317,
|
| 752 |
+
"eval_samples_per_second": 43.218,
|
| 753 |
+
"eval_steps_per_second": 0.687,
|
| 754 |
+
"eval_wer": 0.1506743355811186,
|
| 755 |
+
"step": 1200
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.71,
|
| 759 |
+
"learning_rate": 0.00016852459016393443,
|
| 760 |
+
"loss": 0.3478,
|
| 761 |
+
"step": 1210
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 2.74,
|
| 765 |
+
"learning_rate": 0.00016688524590163935,
|
| 766 |
+
"loss": 0.3861,
|
| 767 |
+
"step": 1220
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.76,
|
| 771 |
+
"learning_rate": 0.00016524590163934425,
|
| 772 |
+
"loss": 0.3283,
|
| 773 |
+
"step": 1230
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.78,
|
| 777 |
+
"learning_rate": 0.00016360655737704917,
|
| 778 |
+
"loss": 0.3455,
|
| 779 |
+
"step": 1240
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.8,
|
| 783 |
+
"learning_rate": 0.0001619672131147541,
|
| 784 |
+
"loss": 0.3457,
|
| 785 |
+
"step": 1250
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.83,
|
| 789 |
+
"learning_rate": 0.000160327868852459,
|
| 790 |
+
"loss": 0.3178,
|
| 791 |
+
"step": 1260
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 2.85,
|
| 795 |
+
"learning_rate": 0.00015868852459016393,
|
| 796 |
+
"loss": 0.3429,
|
| 797 |
+
"step": 1270
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 2.87,
|
| 801 |
+
"learning_rate": 0.00015704918032786883,
|
| 802 |
+
"loss": 0.3253,
|
| 803 |
+
"step": 1280
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.89,
|
| 807 |
+
"learning_rate": 0.00015540983606557375,
|
| 808 |
+
"loss": 0.3469,
|
| 809 |
+
"step": 1290
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.91,
|
| 813 |
+
"learning_rate": 0.00015377049180327867,
|
| 814 |
+
"loss": 0.334,
|
| 815 |
+
"step": 1300
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 2.94,
|
| 819 |
+
"learning_rate": 0.0001521311475409836,
|
| 820 |
+
"loss": 0.3246,
|
| 821 |
+
"step": 1310
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.96,
|
| 825 |
+
"learning_rate": 0.0001504918032786885,
|
| 826 |
+
"loss": 0.3243,
|
| 827 |
+
"step": 1320
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.98,
|
| 831 |
+
"learning_rate": 0.00014885245901639343,
|
| 832 |
+
"loss": 0.3219,
|
| 833 |
+
"step": 1330
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.0,
|
| 837 |
+
"learning_rate": 0.00014721311475409836,
|
| 838 |
+
"loss": 0.3022,
|
| 839 |
+
"step": 1340
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 3.03,
|
| 843 |
+
"learning_rate": 0.00014557377049180328,
|
| 844 |
+
"loss": 0.1935,
|
| 845 |
+
"step": 1350
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.05,
|
| 849 |
+
"learning_rate": 0.0001439344262295082,
|
| 850 |
+
"loss": 0.2093,
|
| 851 |
+
"step": 1360
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.07,
|
| 855 |
+
"learning_rate": 0.0001422950819672131,
|
| 856 |
+
"loss": 0.1961,
|
| 857 |
+
"step": 1370
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.09,
|
| 861 |
+
"learning_rate": 0.00014065573770491801,
|
| 862 |
+
"loss": 0.1788,
|
| 863 |
+
"step": 1380
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.12,
|
| 867 |
+
"learning_rate": 0.00013901639344262294,
|
| 868 |
+
"loss": 0.2245,
|
| 869 |
+
"step": 1390
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 3.14,
|
| 873 |
+
"learning_rate": 0.00013737704918032786,
|
| 874 |
+
"loss": 0.1815,
|
| 875 |
+
"step": 1400
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 3.16,
|
| 879 |
+
"learning_rate": 0.00013573770491803278,
|
| 880 |
+
"loss": 0.2018,
|
| 881 |
+
"step": 1410
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 3.18,
|
| 885 |
+
"learning_rate": 0.0001340983606557377,
|
| 886 |
+
"loss": 0.1909,
|
| 887 |
+
"step": 1420
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 3.21,
|
| 891 |
+
"learning_rate": 0.00013245901639344262,
|
| 892 |
+
"loss": 0.2025,
|
| 893 |
+
"step": 1430
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 3.23,
|
| 897 |
+
"learning_rate": 0.00013081967213114754,
|
| 898 |
+
"loss": 0.1927,
|
| 899 |
+
"step": 1440
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 3.25,
|
| 903 |
+
"learning_rate": 0.00012918032786885246,
|
| 904 |
+
"loss": 0.1816,
|
| 905 |
+
"step": 1450
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 3.27,
|
| 909 |
+
"learning_rate": 0.00012754098360655736,
|
| 910 |
+
"loss": 0.2217,
|
| 911 |
+
"step": 1460
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 3.3,
|
| 915 |
+
"learning_rate": 0.00012590163934426228,
|
| 916 |
+
"loss": 0.2035,
|
| 917 |
+
"step": 1470
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 3.32,
|
| 921 |
+
"learning_rate": 0.0001242622950819672,
|
| 922 |
+
"loss": 0.1904,
|
| 923 |
+
"step": 1480
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 3.34,
|
| 927 |
+
"learning_rate": 0.00012262295081967212,
|
| 928 |
+
"loss": 0.1945,
|
| 929 |
+
"step": 1490
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 3.36,
|
| 933 |
+
"learning_rate": 0.00012098360655737703,
|
| 934 |
+
"loss": 0.1724,
|
| 935 |
+
"step": 1500
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 3.39,
|
| 939 |
+
"learning_rate": 0.00011934426229508195,
|
| 940 |
+
"loss": 0.1704,
|
| 941 |
+
"step": 1510
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 3.41,
|
| 945 |
+
"learning_rate": 0.00011770491803278688,
|
| 946 |
+
"loss": 0.1745,
|
| 947 |
+
"step": 1520
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 3.43,
|
| 951 |
+
"learning_rate": 0.00011606557377049179,
|
| 952 |
+
"loss": 0.1756,
|
| 953 |
+
"step": 1530
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 3.45,
|
| 957 |
+
"learning_rate": 0.00011442622950819671,
|
| 958 |
+
"loss": 0.1864,
|
| 959 |
+
"step": 1540
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 3.48,
|
| 963 |
+
"learning_rate": 0.00011278688524590164,
|
| 964 |
+
"loss": 0.1799,
|
| 965 |
+
"step": 1550
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 3.5,
|
| 969 |
+
"learning_rate": 0.00011114754098360654,
|
| 970 |
+
"loss": 0.1974,
|
| 971 |
+
"step": 1560
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 3.52,
|
| 975 |
+
"learning_rate": 0.00010950819672131146,
|
| 976 |
+
"loss": 0.1751,
|
| 977 |
+
"step": 1570
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.54,
|
| 981 |
+
"learning_rate": 0.00010786885245901639,
|
| 982 |
+
"loss": 0.1792,
|
| 983 |
+
"step": 1580
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 3.57,
|
| 987 |
+
"learning_rate": 0.00010622950819672129,
|
| 988 |
+
"loss": 0.1887,
|
| 989 |
+
"step": 1590
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 3.59,
|
| 993 |
+
"learning_rate": 0.00010459016393442621,
|
| 994 |
+
"loss": 0.1763,
|
| 995 |
+
"step": 1600
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 3.59,
|
| 999 |
+
"eval_loss": 0.47305428981781006,
|
| 1000 |
+
"eval_runtime": 60.8349,
|
| 1001 |
+
"eval_samples_per_second": 43.429,
|
| 1002 |
+
"eval_steps_per_second": 0.69,
|
| 1003 |
+
"eval_wer": 0.12973026576755256,
|
| 1004 |
+
"step": 1600
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 3.61,
|
| 1008 |
+
"learning_rate": 0.00010295081967213115,
|
| 1009 |
+
"loss": 0.1767,
|
| 1010 |
+
"step": 1610
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 3.63,
|
| 1014 |
+
"learning_rate": 0.00010131147540983606,
|
| 1015 |
+
"loss": 0.1769,
|
| 1016 |
+
"step": 1620
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 3.65,
|
| 1020 |
+
"learning_rate": 9.967213114754098e-05,
|
| 1021 |
+
"loss": 0.1691,
|
| 1022 |
+
"step": 1630
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.68,
|
| 1026 |
+
"learning_rate": 9.80327868852459e-05,
|
| 1027 |
+
"loss": 0.1762,
|
| 1028 |
+
"step": 1640
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 3.7,
|
| 1032 |
+
"learning_rate": 9.639344262295081e-05,
|
| 1033 |
+
"loss": 0.1636,
|
| 1034 |
+
"step": 1650
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 3.72,
|
| 1038 |
+
"learning_rate": 9.475409836065573e-05,
|
| 1039 |
+
"loss": 0.1834,
|
| 1040 |
+
"step": 1660
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.74,
|
| 1044 |
+
"learning_rate": 9.311475409836065e-05,
|
| 1045 |
+
"loss": 0.1781,
|
| 1046 |
+
"step": 1670
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 3.77,
|
| 1050 |
+
"learning_rate": 9.147540983606556e-05,
|
| 1051 |
+
"loss": 0.1686,
|
| 1052 |
+
"step": 1680
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.79,
|
| 1056 |
+
"learning_rate": 8.983606557377048e-05,
|
| 1057 |
+
"loss": 0.1657,
|
| 1058 |
+
"step": 1690
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 3.81,
|
| 1062 |
+
"learning_rate": 8.819672131147541e-05,
|
| 1063 |
+
"loss": 0.1468,
|
| 1064 |
+
"step": 1700
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.83,
|
| 1068 |
+
"learning_rate": 8.655737704918032e-05,
|
| 1069 |
+
"loss": 0.1686,
|
| 1070 |
+
"step": 1710
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 3.86,
|
| 1074 |
+
"learning_rate": 8.491803278688524e-05,
|
| 1075 |
+
"loss": 0.1583,
|
| 1076 |
+
"step": 1720
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.88,
|
| 1080 |
+
"learning_rate": 8.327868852459016e-05,
|
| 1081 |
+
"loss": 0.1656,
|
| 1082 |
+
"step": 1730
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 3.9,
|
| 1086 |
+
"learning_rate": 8.163934426229507e-05,
|
| 1087 |
+
"loss": 0.1595,
|
| 1088 |
+
"step": 1740
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 3.92,
|
| 1092 |
+
"learning_rate": 7.999999999999999e-05,
|
| 1093 |
+
"loss": 0.1563,
|
| 1094 |
+
"step": 1750
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.95,
|
| 1098 |
+
"learning_rate": 7.836065573770491e-05,
|
| 1099 |
+
"loss": 0.1625,
|
| 1100 |
+
"step": 1760
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 3.97,
|
| 1104 |
+
"learning_rate": 7.672131147540982e-05,
|
| 1105 |
+
"loss": 0.1439,
|
| 1106 |
+
"step": 1770
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.99,
|
| 1110 |
+
"learning_rate": 7.508196721311474e-05,
|
| 1111 |
+
"loss": 0.1593,
|
| 1112 |
+
"step": 1780
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 4.01,
|
| 1116 |
+
"learning_rate": 7.344262295081966e-05,
|
| 1117 |
+
"loss": 0.1042,
|
| 1118 |
+
"step": 1790
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 4.04,
|
| 1122 |
+
"learning_rate": 7.180327868852459e-05,
|
| 1123 |
+
"loss": 0.0931,
|
| 1124 |
+
"step": 1800
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 4.06,
|
| 1128 |
+
"learning_rate": 7.01639344262295e-05,
|
| 1129 |
+
"loss": 0.0909,
|
| 1130 |
+
"step": 1810
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 4.08,
|
| 1134 |
+
"learning_rate": 6.852459016393443e-05,
|
| 1135 |
+
"loss": 0.0955,
|
| 1136 |
+
"step": 1820
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 4.1,
|
| 1140 |
+
"learning_rate": 6.688524590163934e-05,
|
| 1141 |
+
"loss": 0.102,
|
| 1142 |
+
"step": 1830
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 4.13,
|
| 1146 |
+
"learning_rate": 6.524590163934426e-05,
|
| 1147 |
+
"loss": 0.0889,
|
| 1148 |
+
"step": 1840
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 4.15,
|
| 1152 |
+
"learning_rate": 6.360655737704918e-05,
|
| 1153 |
+
"loss": 0.0968,
|
| 1154 |
+
"step": 1850
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 4.17,
|
| 1158 |
+
"learning_rate": 6.19672131147541e-05,
|
| 1159 |
+
"loss": 0.0809,
|
| 1160 |
+
"step": 1860
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 4.19,
|
| 1164 |
+
"learning_rate": 6.0327868852459015e-05,
|
| 1165 |
+
"loss": 0.0869,
|
| 1166 |
+
"step": 1870
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 4.22,
|
| 1170 |
+
"learning_rate": 5.868852459016393e-05,
|
| 1171 |
+
"loss": 0.0913,
|
| 1172 |
+
"step": 1880
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 4.24,
|
| 1176 |
+
"learning_rate": 5.7049180327868844e-05,
|
| 1177 |
+
"loss": 0.078,
|
| 1178 |
+
"step": 1890
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 4.26,
|
| 1182 |
+
"learning_rate": 5.5409836065573765e-05,
|
| 1183 |
+
"loss": 0.0924,
|
| 1184 |
+
"step": 1900
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 4.28,
|
| 1188 |
+
"learning_rate": 5.3770491803278686e-05,
|
| 1189 |
+
"loss": 0.0879,
|
| 1190 |
+
"step": 1910
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 4.3,
|
| 1194 |
+
"learning_rate": 5.21311475409836e-05,
|
| 1195 |
+
"loss": 0.0869,
|
| 1196 |
+
"step": 1920
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 4.33,
|
| 1200 |
+
"learning_rate": 5.049180327868852e-05,
|
| 1201 |
+
"loss": 0.0832,
|
| 1202 |
+
"step": 1930
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 4.35,
|
| 1206 |
+
"learning_rate": 4.8852459016393436e-05,
|
| 1207 |
+
"loss": 0.0777,
|
| 1208 |
+
"step": 1940
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 4.37,
|
| 1212 |
+
"learning_rate": 4.721311475409836e-05,
|
| 1213 |
+
"loss": 0.0905,
|
| 1214 |
+
"step": 1950
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 4.39,
|
| 1218 |
+
"learning_rate": 4.557377049180328e-05,
|
| 1219 |
+
"loss": 0.0818,
|
| 1220 |
+
"step": 1960
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 4.42,
|
| 1224 |
+
"learning_rate": 4.3934426229508194e-05,
|
| 1225 |
+
"loss": 0.0829,
|
| 1226 |
+
"step": 1970
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 4.44,
|
| 1230 |
+
"learning_rate": 4.229508196721311e-05,
|
| 1231 |
+
"loss": 0.0865,
|
| 1232 |
+
"step": 1980
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 4.46,
|
| 1236 |
+
"learning_rate": 4.065573770491803e-05,
|
| 1237 |
+
"loss": 0.0895,
|
| 1238 |
+
"step": 1990
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 4.48,
|
| 1242 |
+
"learning_rate": 3.901639344262295e-05,
|
| 1243 |
+
"loss": 0.089,
|
| 1244 |
+
"step": 2000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 4.48,
|
| 1248 |
+
"eval_loss": 0.41652488708496094,
|
| 1249 |
+
"eval_runtime": 61.3777,
|
| 1250 |
+
"eval_samples_per_second": 43.045,
|
| 1251 |
+
"eval_steps_per_second": 0.684,
|
| 1252 |
+
"eval_wer": 0.11800872669575566,
|
| 1253 |
+
"step": 2000
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 4.51,
|
| 1257 |
+
"learning_rate": 3.7377049180327865e-05,
|
| 1258 |
+
"loss": 0.071,
|
| 1259 |
+
"step": 2010
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 4.53,
|
| 1263 |
+
"learning_rate": 3.5737704918032786e-05,
|
| 1264 |
+
"loss": 0.0847,
|
| 1265 |
+
"step": 2020
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 4.55,
|
| 1269 |
+
"learning_rate": 3.40983606557377e-05,
|
| 1270 |
+
"loss": 0.0898,
|
| 1271 |
+
"step": 2030
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 4.57,
|
| 1275 |
+
"learning_rate": 3.245901639344262e-05,
|
| 1276 |
+
"loss": 0.0798,
|
| 1277 |
+
"step": 2040
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 4.6,
|
| 1281 |
+
"learning_rate": 3.081967213114754e-05,
|
| 1282 |
+
"loss": 0.091,
|
| 1283 |
+
"step": 2050
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 4.62,
|
| 1287 |
+
"learning_rate": 2.9180327868852458e-05,
|
| 1288 |
+
"loss": 0.0867,
|
| 1289 |
+
"step": 2060
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 4.64,
|
| 1293 |
+
"learning_rate": 2.7540983606557373e-05,
|
| 1294 |
+
"loss": 0.0847,
|
| 1295 |
+
"step": 2070
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 4.66,
|
| 1299 |
+
"learning_rate": 2.5901639344262294e-05,
|
| 1300 |
+
"loss": 0.0861,
|
| 1301 |
+
"step": 2080
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 4.69,
|
| 1305 |
+
"learning_rate": 2.4262295081967212e-05,
|
| 1306 |
+
"loss": 0.0858,
|
| 1307 |
+
"step": 2090
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 4.71,
|
| 1311 |
+
"learning_rate": 2.262295081967213e-05,
|
| 1312 |
+
"loss": 0.0862,
|
| 1313 |
+
"step": 2100
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 4.73,
|
| 1317 |
+
"learning_rate": 2.0983606557377048e-05,
|
| 1318 |
+
"loss": 0.0781,
|
| 1319 |
+
"step": 2110
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 4.75,
|
| 1323 |
+
"learning_rate": 1.9344262295081962e-05,
|
| 1324 |
+
"loss": 0.0733,
|
| 1325 |
+
"step": 2120
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 4.78,
|
| 1329 |
+
"learning_rate": 1.7704918032786883e-05,
|
| 1330 |
+
"loss": 0.0701,
|
| 1331 |
+
"step": 2130
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 4.8,
|
| 1335 |
+
"learning_rate": 1.60655737704918e-05,
|
| 1336 |
+
"loss": 0.0688,
|
| 1337 |
+
"step": 2140
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 4.82,
|
| 1341 |
+
"learning_rate": 1.442622950819672e-05,
|
| 1342 |
+
"loss": 0.0793,
|
| 1343 |
+
"step": 2150
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 4.84,
|
| 1347 |
+
"learning_rate": 1.2786885245901637e-05,
|
| 1348 |
+
"loss": 0.074,
|
| 1349 |
+
"step": 2160
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 4.87,
|
| 1353 |
+
"learning_rate": 1.1147540983606557e-05,
|
| 1354 |
+
"loss": 0.0728,
|
| 1355 |
+
"step": 2170
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 4.89,
|
| 1359 |
+
"learning_rate": 9.508196721311474e-06,
|
| 1360 |
+
"loss": 0.0702,
|
| 1361 |
+
"step": 2180
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 4.91,
|
| 1365 |
+
"learning_rate": 7.868852459016392e-06,
|
| 1366 |
+
"loss": 0.0672,
|
| 1367 |
+
"step": 2190
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 4.93,
|
| 1371 |
+
"learning_rate": 6.229508196721311e-06,
|
| 1372 |
+
"loss": 0.0799,
|
| 1373 |
+
"step": 2200
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 4.96,
|
| 1377 |
+
"learning_rate": 4.590163934426229e-06,
|
| 1378 |
+
"loss": 0.0782,
|
| 1379 |
+
"step": 2210
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 4.98,
|
| 1383 |
+
"learning_rate": 2.9508196721311474e-06,
|
| 1384 |
+
"loss": 0.0856,
|
| 1385 |
+
"step": 2220
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 5.0,
|
| 1389 |
+
"learning_rate": 1.3114754098360655e-06,
|
| 1390 |
+
"loss": 0.0727,
|
| 1391 |
+
"step": 2230
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 5.0,
|
| 1395 |
+
"step": 2230,
|
| 1396 |
+
"total_flos": 2.9474537983075942e+19,
|
| 1397 |
+
"train_loss": 1.5071858802451146,
|
| 1398 |
+
"train_runtime": 3191.5244,
|
| 1399 |
+
"train_samples_per_second": 44.709,
|
| 1400 |
+
"train_steps_per_second": 0.699
|
| 1401 |
+
}
|
| 1402 |
+
],
|
| 1403 |
+
"max_steps": 2230,
|
| 1404 |
+
"num_train_epochs": 5,
|
| 1405 |
+
"total_flos": 2.9474537983075942e+19,
|
| 1406 |
+
"trial_name": null,
|
| 1407 |
+
"trial_params": null
|
| 1408 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c029da8afe9a48537ce0d99ab7e8b3fe26f7bf6fda6f49c143ff0da9917a35f4
|
| 3 |
+
size 3055
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|