add best WER checkpoint at step 1200
Browse files- checkpoint-1200/config.json +41 -0
- checkpoint-1200/optimizer.pt +3 -0
- checkpoint-1200/preprocessor_config.json +0 -0
- checkpoint-1200/pytorch_model.bin +3 -0
- checkpoint-1200/rng_state.pth +3 -0
- checkpoint-1200/scaler.pt +3 -0
- checkpoint-1200/scheduler.pt +3 -0
- checkpoint-1200/trainer_state.json +412 -0
- checkpoint-1200/training_args.bin +3 -0
checkpoint-1200/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-medium",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"d_model": 1024,
|
| 15 |
+
"decoder_attention_heads": 16,
|
| 16 |
+
"decoder_ffn_dim": 4096,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 24,
|
| 19 |
+
"decoder_start_token_id": 50258,
|
| 20 |
+
"dropout": 0.0,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 24,
|
| 25 |
+
"eos_token_id": 50257,
|
| 26 |
+
"forced_decoder_ids": null,
|
| 27 |
+
"init_std": 0.02,
|
| 28 |
+
"is_encoder_decoder": true,
|
| 29 |
+
"max_length": 448,
|
| 30 |
+
"max_source_positions": 1500,
|
| 31 |
+
"max_target_positions": 448,
|
| 32 |
+
"model_type": "whisper",
|
| 33 |
+
"num_hidden_layers": 24,
|
| 34 |
+
"num_mel_bins": 80,
|
| 35 |
+
"pad_token_id": 50257,
|
| 36 |
+
"scale_embedding": false,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 51865
|
| 41 |
+
}
|
checkpoint-1200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d458445da730ae852fefdd1efbb7a3cbb6eb7e998ca01239c0b62a9ee24c2671
|
| 3 |
+
size 6111428695
|
checkpoint-1200/preprocessor_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1200/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2b160705165d9aa7c832adbda8e4aa2aba2bd16a3883e9801086c5a755b22b0
|
| 3 |
+
size 3055754841
|
checkpoint-1200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34ad700147ef2fad3ab0df1630fa483e649bc1c8bb65b171f22f772c0b8981ef
|
| 3 |
+
size 14575
|
checkpoint-1200/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17cc5d53f6ce64f7a24cc72e7c61187dc4c279eba99011d48c6c802500fc106d
|
| 3 |
+
size 557
|
checkpoint-1200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae020d01dc6e4c380fefd015708f5d69a276d355437f37b6781081a4054ee6d9
|
| 3 |
+
size 627
|
checkpoint-1200/trainer_state.json
ADDED
|
@@ -0,0 +1,412 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 50.544794188861985,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-1200",
|
| 4 |
+
"epoch": 171.42857142857142,
|
| 5 |
+
"global_step": 1200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.57,
|
| 12 |
+
"learning_rate": 4.4e-06,
|
| 13 |
+
"loss": 2.1832,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 7.14,
|
| 18 |
+
"learning_rate": 9.4e-06,
|
| 19 |
+
"loss": 0.7236,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 10.71,
|
| 24 |
+
"learning_rate": 9.76842105263158e-06,
|
| 25 |
+
"loss": 0.2181,
|
| 26 |
+
"step": 75
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 14.29,
|
| 30 |
+
"learning_rate": 9.505263157894738e-06,
|
| 31 |
+
"loss": 0.0334,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 14.29,
|
| 36 |
+
"eval_loss": 1.0348163843154907,
|
| 37 |
+
"eval_runtime": 1390.6808,
|
| 38 |
+
"eval_samples_per_second": 0.368,
|
| 39 |
+
"eval_steps_per_second": 0.023,
|
| 40 |
+
"eval_wer": 50.090799031477,
|
| 41 |
+
"step": 100
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 17.86,
|
| 45 |
+
"learning_rate": 9.242105263157896e-06,
|
| 46 |
+
"loss": 0.0077,
|
| 47 |
+
"step": 125
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 21.43,
|
| 51 |
+
"learning_rate": 8.978947368421055e-06,
|
| 52 |
+
"loss": 0.0041,
|
| 53 |
+
"step": 150
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 25.0,
|
| 57 |
+
"learning_rate": 8.715789473684211e-06,
|
| 58 |
+
"loss": 0.0026,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 28.57,
|
| 63 |
+
"learning_rate": 8.45263157894737e-06,
|
| 64 |
+
"loss": 0.0021,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 28.57,
|
| 69 |
+
"eval_loss": 1.1971338987350464,
|
| 70 |
+
"eval_runtime": 1356.432,
|
| 71 |
+
"eval_samples_per_second": 0.377,
|
| 72 |
+
"eval_steps_per_second": 0.024,
|
| 73 |
+
"eval_wer": 49.48547215496368,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 32.14,
|
| 78 |
+
"learning_rate": 8.189473684210527e-06,
|
| 79 |
+
"loss": 0.0014,
|
| 80 |
+
"step": 225
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 35.71,
|
| 84 |
+
"learning_rate": 7.926315789473686e-06,
|
| 85 |
+
"loss": 0.0009,
|
| 86 |
+
"step": 250
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 39.29,
|
| 90 |
+
"learning_rate": 7.663157894736842e-06,
|
| 91 |
+
"loss": 0.0008,
|
| 92 |
+
"step": 275
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 42.86,
|
| 96 |
+
"learning_rate": 7.4e-06,
|
| 97 |
+
"loss": 0.0007,
|
| 98 |
+
"step": 300
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 42.86,
|
| 102 |
+
"eval_loss": 1.265063762664795,
|
| 103 |
+
"eval_runtime": 1370.312,
|
| 104 |
+
"eval_samples_per_second": 0.374,
|
| 105 |
+
"eval_steps_per_second": 0.023,
|
| 106 |
+
"eval_wer": 49.73516949152542,
|
| 107 |
+
"step": 300
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 46.43,
|
| 111 |
+
"learning_rate": 7.1368421052631585e-06,
|
| 112 |
+
"loss": 0.0007,
|
| 113 |
+
"step": 325
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 50.0,
|
| 117 |
+
"learning_rate": 6.873684210526317e-06,
|
| 118 |
+
"loss": 0.0006,
|
| 119 |
+
"step": 350
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 53.57,
|
| 123 |
+
"learning_rate": 6.610526315789474e-06,
|
| 124 |
+
"loss": 0.0006,
|
| 125 |
+
"step": 375
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 57.14,
|
| 129 |
+
"learning_rate": 6.347368421052632e-06,
|
| 130 |
+
"loss": 0.0006,
|
| 131 |
+
"step": 400
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 57.14,
|
| 135 |
+
"eval_loss": 1.3083866834640503,
|
| 136 |
+
"eval_runtime": 1377.4534,
|
| 137 |
+
"eval_samples_per_second": 0.372,
|
| 138 |
+
"eval_steps_per_second": 0.023,
|
| 139 |
+
"eval_wer": 49.969733656174334,
|
| 140 |
+
"step": 400
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 60.71,
|
| 144 |
+
"learning_rate": 6.08421052631579e-06,
|
| 145 |
+
"loss": 0.0005,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 64.29,
|
| 150 |
+
"learning_rate": 5.8210526315789486e-06,
|
| 151 |
+
"loss": 0.0005,
|
| 152 |
+
"step": 450
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 67.86,
|
| 156 |
+
"learning_rate": 5.557894736842105e-06,
|
| 157 |
+
"loss": 0.0005,
|
| 158 |
+
"step": 475
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 71.43,
|
| 162 |
+
"learning_rate": 5.294736842105263e-06,
|
| 163 |
+
"loss": 0.0005,
|
| 164 |
+
"step": 500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 71.43,
|
| 168 |
+
"eval_loss": 1.3478939533233643,
|
| 169 |
+
"eval_runtime": 1368.2052,
|
| 170 |
+
"eval_samples_per_second": 0.374,
|
| 171 |
+
"eval_steps_per_second": 0.023,
|
| 172 |
+
"eval_wer": 50.060532687651325,
|
| 173 |
+
"step": 500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 75.0,
|
| 177 |
+
"learning_rate": 5.0315789473684214e-06,
|
| 178 |
+
"loss": 0.0005,
|
| 179 |
+
"step": 525
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 78.57,
|
| 183 |
+
"learning_rate": 4.76842105263158e-06,
|
| 184 |
+
"loss": 0.0005,
|
| 185 |
+
"step": 550
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 82.14,
|
| 189 |
+
"learning_rate": 4.505263157894737e-06,
|
| 190 |
+
"loss": 0.0004,
|
| 191 |
+
"step": 575
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 85.71,
|
| 195 |
+
"learning_rate": 4.242105263157895e-06,
|
| 196 |
+
"loss": 0.0004,
|
| 197 |
+
"step": 600
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 85.71,
|
| 201 |
+
"eval_loss": 1.3834712505340576,
|
| 202 |
+
"eval_runtime": 1377.4466,
|
| 203 |
+
"eval_samples_per_second": 0.372,
|
| 204 |
+
"eval_steps_per_second": 0.023,
|
| 205 |
+
"eval_wer": 50.30266343825666,
|
| 206 |
+
"step": 600
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 89.29,
|
| 210 |
+
"learning_rate": 3.9789473684210525e-06,
|
| 211 |
+
"loss": 0.0004,
|
| 212 |
+
"step": 625
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 92.86,
|
| 216 |
+
"learning_rate": 3.715789473684211e-06,
|
| 217 |
+
"loss": 0.0004,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 96.43,
|
| 222 |
+
"learning_rate": 3.4526315789473684e-06,
|
| 223 |
+
"loss": 0.0004,
|
| 224 |
+
"step": 675
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 100.0,
|
| 228 |
+
"learning_rate": 3.1894736842105266e-06,
|
| 229 |
+
"loss": 0.0004,
|
| 230 |
+
"step": 700
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 100.0,
|
| 234 |
+
"eval_loss": 1.4139070510864258,
|
| 235 |
+
"eval_runtime": 1374.8999,
|
| 236 |
+
"eval_samples_per_second": 0.372,
|
| 237 |
+
"eval_steps_per_second": 0.023,
|
| 238 |
+
"eval_wer": 50.45399515738499,
|
| 239 |
+
"step": 700
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 103.57,
|
| 243 |
+
"learning_rate": 2.9263157894736844e-06,
|
| 244 |
+
"loss": 0.0004,
|
| 245 |
+
"step": 725
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 107.14,
|
| 249 |
+
"learning_rate": 2.6631578947368426e-06,
|
| 250 |
+
"loss": 0.0004,
|
| 251 |
+
"step": 750
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 110.71,
|
| 255 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 256 |
+
"loss": 0.0004,
|
| 257 |
+
"step": 775
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 114.29,
|
| 261 |
+
"learning_rate": 2.136842105263158e-06,
|
| 262 |
+
"loss": 0.0004,
|
| 263 |
+
"step": 800
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 114.29,
|
| 267 |
+
"eval_loss": 1.4381563663482666,
|
| 268 |
+
"eval_runtime": 1378.8054,
|
| 269 |
+
"eval_samples_per_second": 0.371,
|
| 270 |
+
"eval_steps_per_second": 0.023,
|
| 271 |
+
"eval_wer": 50.46156174334141,
|
| 272 |
+
"step": 800
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 117.86,
|
| 276 |
+
"learning_rate": 1.8736842105263158e-06,
|
| 277 |
+
"loss": 0.0004,
|
| 278 |
+
"step": 825
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 121.43,
|
| 282 |
+
"learning_rate": 1.6105263157894738e-06,
|
| 283 |
+
"loss": 0.0004,
|
| 284 |
+
"step": 850
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 125.0,
|
| 288 |
+
"learning_rate": 1.3473684210526316e-06,
|
| 289 |
+
"loss": 0.0004,
|
| 290 |
+
"step": 875
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 128.57,
|
| 294 |
+
"learning_rate": 1.0842105263157895e-06,
|
| 295 |
+
"loss": 0.0004,
|
| 296 |
+
"step": 900
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 128.57,
|
| 300 |
+
"eval_loss": 1.454498291015625,
|
| 301 |
+
"eval_runtime": 1378.7878,
|
| 302 |
+
"eval_samples_per_second": 0.371,
|
| 303 |
+
"eval_steps_per_second": 0.023,
|
| 304 |
+
"eval_wer": 50.529661016949156,
|
| 305 |
+
"step": 900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 132.14,
|
| 309 |
+
"learning_rate": 8.210526315789474e-07,
|
| 310 |
+
"loss": 0.0003,
|
| 311 |
+
"step": 925
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 135.71,
|
| 315 |
+
"learning_rate": 5.578947368421053e-07,
|
| 316 |
+
"loss": 0.0003,
|
| 317 |
+
"step": 950
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 139.29,
|
| 321 |
+
"learning_rate": 2.9473684210526315e-07,
|
| 322 |
+
"loss": 0.0003,
|
| 323 |
+
"step": 975
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 142.86,
|
| 327 |
+
"learning_rate": 3.157894736842106e-08,
|
| 328 |
+
"loss": 0.0003,
|
| 329 |
+
"step": 1000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 142.86,
|
| 333 |
+
"eval_loss": 1.4602879285812378,
|
| 334 |
+
"eval_runtime": 1395.1761,
|
| 335 |
+
"eval_samples_per_second": 0.367,
|
| 336 |
+
"eval_steps_per_second": 0.023,
|
| 337 |
+
"eval_wer": 50.56749394673123,
|
| 338 |
+
"step": 1000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 146.43,
|
| 342 |
+
"learning_rate": 1.4957983193277311e-06,
|
| 343 |
+
"loss": 0.0003,
|
| 344 |
+
"step": 1025
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 150.0,
|
| 348 |
+
"learning_rate": 1.2857142857142856e-06,
|
| 349 |
+
"loss": 0.0003,
|
| 350 |
+
"step": 1050
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 153.57,
|
| 354 |
+
"learning_rate": 1.0756302521008405e-06,
|
| 355 |
+
"loss": 0.0003,
|
| 356 |
+
"step": 1075
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 157.14,
|
| 360 |
+
"learning_rate": 8.65546218487395e-07,
|
| 361 |
+
"loss": 0.0003,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 157.14,
|
| 366 |
+
"eval_loss": 1.4750263690948486,
|
| 367 |
+
"eval_runtime": 1385.2093,
|
| 368 |
+
"eval_samples_per_second": 0.37,
|
| 369 |
+
"eval_steps_per_second": 0.023,
|
| 370 |
+
"eval_wer": 50.559927360774815,
|
| 371 |
+
"step": 1100
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 160.71,
|
| 375 |
+
"learning_rate": 6.554621848739497e-07,
|
| 376 |
+
"loss": 0.0003,
|
| 377 |
+
"step": 1125
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 164.29,
|
| 381 |
+
"learning_rate": 4.4537815126050424e-07,
|
| 382 |
+
"loss": 0.0003,
|
| 383 |
+
"step": 1150
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 167.86,
|
| 387 |
+
"learning_rate": 2.3529411764705883e-07,
|
| 388 |
+
"loss": 0.0003,
|
| 389 |
+
"step": 1175
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 171.43,
|
| 393 |
+
"learning_rate": 2.521008403361345e-08,
|
| 394 |
+
"loss": 0.0003,
|
| 395 |
+
"step": 1200
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 171.43,
|
| 399 |
+
"eval_loss": 1.4806574583053589,
|
| 400 |
+
"eval_runtime": 1382.0563,
|
| 401 |
+
"eval_samples_per_second": 0.37,
|
| 402 |
+
"eval_steps_per_second": 0.023,
|
| 403 |
+
"eval_wer": 50.544794188861985,
|
| 404 |
+
"step": 1200
|
| 405 |
+
}
|
| 406 |
+
],
|
| 407 |
+
"max_steps": 1200,
|
| 408 |
+
"num_train_epochs": 172,
|
| 409 |
+
"total_flos": 7.541555843137536e+19,
|
| 410 |
+
"trial_name": null,
|
| 411 |
+
"trial_params": null
|
| 412 |
+
}
|
checkpoint-1200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9805d9bd109ed2a40594bfd2ab0a7e52952d96b67a7bca4857b01ac1f91b928
|
| 3 |
+
size 3579
|