checkpoint 14400 of ksponspeech train 1/20
Browse files- config.json +76 -0
- optimizer.pt +3 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- trainer_state.json +520 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"apply_spec_augment": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Wav2Vec2ForCTC"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"conv_bias": true,
|
| 11 |
+
"conv_dim": [
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512
|
| 19 |
+
],
|
| 20 |
+
"conv_kernel": [
|
| 21 |
+
10,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
2,
|
| 27 |
+
2
|
| 28 |
+
],
|
| 29 |
+
"conv_stride": [
|
| 30 |
+
5,
|
| 31 |
+
2,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2
|
| 37 |
+
],
|
| 38 |
+
"ctc_loss_reduction": "mean",
|
| 39 |
+
"ctc_zero_infinity": false,
|
| 40 |
+
"do_stable_layer_norm": true,
|
| 41 |
+
"eos_token_id": 2,
|
| 42 |
+
"feat_extract_activation": "gelu",
|
| 43 |
+
"feat_extract_dropout": 0.0,
|
| 44 |
+
"feat_extract_norm": "layer",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"final_dropout": 0.0,
|
| 47 |
+
"gradient_checkpointing": true,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 1024,
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 4096,
|
| 53 |
+
"layer_norm_eps": 1e-05,
|
| 54 |
+
"layerdrop": 0.1,
|
| 55 |
+
"mask_channel_length": 10,
|
| 56 |
+
"mask_channel_min_space": 1,
|
| 57 |
+
"mask_channel_other": 0.0,
|
| 58 |
+
"mask_channel_prob": 0.0,
|
| 59 |
+
"mask_channel_selection": "static",
|
| 60 |
+
"mask_feature_length": 10,
|
| 61 |
+
"mask_feature_prob": 0.0,
|
| 62 |
+
"mask_time_length": 10,
|
| 63 |
+
"mask_time_min_space": 1,
|
| 64 |
+
"mask_time_other": 0.0,
|
| 65 |
+
"mask_time_prob": 0.05,
|
| 66 |
+
"mask_time_selection": "static",
|
| 67 |
+
"model_type": "wav2vec2",
|
| 68 |
+
"num_attention_heads": 16,
|
| 69 |
+
"num_conv_pos_embedding_groups": 16,
|
| 70 |
+
"num_conv_pos_embeddings": 128,
|
| 71 |
+
"num_feat_extract_layers": 7,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"pad_token_id": 2311,
|
| 74 |
+
"transformers_version": "4.4.0",
|
| 75 |
+
"vocab_size": 2312
|
| 76 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0beead7fc4aedab6b178be20d0d10d8146f60d78993cf9914392fe044fe3dea3
|
| 3 |
+
size 2509035527
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_size": 1,
|
| 4 |
+
"padding_side": "right",
|
| 5 |
+
"padding_value": 0.0,
|
| 6 |
+
"return_attention_mask": true,
|
| 7 |
+
"sampling_rate": 16000
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80f3e75e686889c1ce44ae0f43b56eed84907b048a98f19ddbd6d5c88d36e13d
|
| 3 |
+
size 1271413015
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25cc75a516875728337724d2b5b66f0d58cc615f64055082cceb126c9e8d11ef
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,520 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.75128998968008,
|
| 5 |
+
"global_step": 14400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.83,
|
| 12 |
+
"learning_rate": 0.00023999999999999998,
|
| 13 |
+
"loss": 24.132,
|
| 14 |
+
"step": 400
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.83,
|
| 18 |
+
"eval_cer": 1.0,
|
| 19 |
+
"eval_loss": 4.829305171966553,
|
| 20 |
+
"eval_runtime": 666.1596,
|
| 21 |
+
"eval_samples_per_second": 3.82,
|
| 22 |
+
"step": 400
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.65,
|
| 26 |
+
"learning_rate": 0.0002935805991440799,
|
| 27 |
+
"loss": 4.7165,
|
| 28 |
+
"step": 800
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.65,
|
| 32 |
+
"eval_cer": 1.0,
|
| 33 |
+
"eval_loss": 4.607272148132324,
|
| 34 |
+
"eval_runtime": 670.8713,
|
| 35 |
+
"eval_samples_per_second": 3.794,
|
| 36 |
+
"step": 800
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 2.48,
|
| 40 |
+
"learning_rate": 0.00028502139800285305,
|
| 41 |
+
"loss": 2.9637,
|
| 42 |
+
"step": 1200
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 2.48,
|
| 46 |
+
"eval_cer": 0.34719755926062756,
|
| 47 |
+
"eval_loss": 1.3734972476959229,
|
| 48 |
+
"eval_runtime": 671.6914,
|
| 49 |
+
"eval_samples_per_second": 3.789,
|
| 50 |
+
"step": 1200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.31,
|
| 54 |
+
"learning_rate": 0.0002764621968616262,
|
| 55 |
+
"loss": 1.3373,
|
| 56 |
+
"step": 1600
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.31,
|
| 60 |
+
"eval_cer": 0.26771859991180924,
|
| 61 |
+
"eval_loss": 0.9831737875938416,
|
| 62 |
+
"eval_runtime": 669.2245,
|
| 63 |
+
"eval_samples_per_second": 3.803,
|
| 64 |
+
"step": 1600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.13,
|
| 68 |
+
"learning_rate": 0.0002679029957203994,
|
| 69 |
+
"loss": 1.0378,
|
| 70 |
+
"step": 2000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 4.13,
|
| 74 |
+
"eval_cer": 0.24304901738788448,
|
| 75 |
+
"eval_loss": 0.8199232220649719,
|
| 76 |
+
"eval_runtime": 670.0042,
|
| 77 |
+
"eval_samples_per_second": 3.798,
|
| 78 |
+
"step": 2000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.96,
|
| 82 |
+
"learning_rate": 0.00025934379457917263,
|
| 83 |
+
"loss": 0.8747,
|
| 84 |
+
"step": 2400
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.96,
|
| 88 |
+
"eval_cer": 0.21944010773576136,
|
| 89 |
+
"eval_loss": 0.7384957075119019,
|
| 90 |
+
"eval_runtime": 673.0546,
|
| 91 |
+
"eval_samples_per_second": 3.781,
|
| 92 |
+
"step": 2400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.78,
|
| 96 |
+
"learning_rate": 0.0002507845934379458,
|
| 97 |
+
"loss": 0.7648,
|
| 98 |
+
"step": 2800
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 5.78,
|
| 102 |
+
"eval_cer": 0.21221799806933703,
|
| 103 |
+
"eval_loss": 0.7131306529045105,
|
| 104 |
+
"eval_runtime": 671.747,
|
| 105 |
+
"eval_samples_per_second": 3.789,
|
| 106 |
+
"step": 2800
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 6.61,
|
| 110 |
+
"learning_rate": 0.00024222539229671895,
|
| 111 |
+
"loss": 0.6863,
|
| 112 |
+
"step": 3200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 6.61,
|
| 116 |
+
"eval_cer": 0.19983553611650717,
|
| 117 |
+
"eval_loss": 0.6777553558349609,
|
| 118 |
+
"eval_runtime": 679.4426,
|
| 119 |
+
"eval_samples_per_second": 3.746,
|
| 120 |
+
"step": 3200
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.44,
|
| 124 |
+
"learning_rate": 0.00023366619115549212,
|
| 125 |
+
"loss": 0.6269,
|
| 126 |
+
"step": 3600
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 7.44,
|
| 130 |
+
"eval_cer": 0.18854950005362953,
|
| 131 |
+
"eval_loss": 0.6489213705062866,
|
| 132 |
+
"eval_runtime": 682.8113,
|
| 133 |
+
"eval_samples_per_second": 3.727,
|
| 134 |
+
"step": 3600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 8.26,
|
| 138 |
+
"learning_rate": 0.00022510699001426532,
|
| 139 |
+
"loss": 0.5893,
|
| 140 |
+
"step": 4000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 8.26,
|
| 144 |
+
"eval_cer": 0.19089728157885327,
|
| 145 |
+
"eval_loss": 0.6407034993171692,
|
| 146 |
+
"eval_runtime": 687.6153,
|
| 147 |
+
"eval_samples_per_second": 3.701,
|
| 148 |
+
"step": 4000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 9.09,
|
| 152 |
+
"learning_rate": 0.0002165477888730385,
|
| 153 |
+
"loss": 0.5301,
|
| 154 |
+
"step": 4400
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 9.09,
|
| 158 |
+
"eval_cer": 0.1817445089322957,
|
| 159 |
+
"eval_loss": 0.6261754035949707,
|
| 160 |
+
"eval_runtime": 681.4189,
|
| 161 |
+
"eval_samples_per_second": 3.735,
|
| 162 |
+
"step": 4400
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 9.92,
|
| 166 |
+
"learning_rate": 0.0002079885877318117,
|
| 167 |
+
"loss": 0.4909,
|
| 168 |
+
"step": 4800
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.92,
|
| 172 |
+
"eval_cer": 0.17955165715239127,
|
| 173 |
+
"eval_loss": 0.623394250869751,
|
| 174 |
+
"eval_runtime": 672.8551,
|
| 175 |
+
"eval_samples_per_second": 3.782,
|
| 176 |
+
"step": 4800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 10.74,
|
| 180 |
+
"learning_rate": 0.00019942938659058487,
|
| 181 |
+
"loss": 0.4602,
|
| 182 |
+
"step": 5200
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 10.74,
|
| 186 |
+
"eval_cer": 0.17870550238949337,
|
| 187 |
+
"eval_loss": 0.6308777928352356,
|
| 188 |
+
"eval_runtime": 673.8601,
|
| 189 |
+
"eval_samples_per_second": 3.777,
|
| 190 |
+
"step": 5200
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 11.57,
|
| 194 |
+
"learning_rate": 0.00019087018544935804,
|
| 195 |
+
"loss": 0.4244,
|
| 196 |
+
"step": 5600
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 11.57,
|
| 200 |
+
"eval_cer": 0.17706086355456507,
|
| 201 |
+
"eval_loss": 0.6224568486213684,
|
| 202 |
+
"eval_runtime": 677.416,
|
| 203 |
+
"eval_samples_per_second": 3.757,
|
| 204 |
+
"step": 5600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 12.4,
|
| 208 |
+
"learning_rate": 0.0001823109843081312,
|
| 209 |
+
"loss": 0.3936,
|
| 210 |
+
"step": 6000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 12.4,
|
| 214 |
+
"eval_cer": 0.17660799199132393,
|
| 215 |
+
"eval_loss": 0.6143100261688232,
|
| 216 |
+
"eval_runtime": 683.1583,
|
| 217 |
+
"eval_samples_per_second": 3.725,
|
| 218 |
+
"step": 6000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 13.22,
|
| 222 |
+
"learning_rate": 0.0001737517831669044,
|
| 223 |
+
"loss": 0.3792,
|
| 224 |
+
"step": 6400
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 13.22,
|
| 228 |
+
"eval_cer": 0.17075641468733985,
|
| 229 |
+
"eval_loss": 0.627468466758728,
|
| 230 |
+
"eval_runtime": 676.725,
|
| 231 |
+
"eval_samples_per_second": 3.761,
|
| 232 |
+
"step": 6400
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 14.05,
|
| 236 |
+
"learning_rate": 0.00016519258202567759,
|
| 237 |
+
"loss": 0.3555,
|
| 238 |
+
"step": 6800
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 14.05,
|
| 242 |
+
"eval_cer": 0.16945738836120083,
|
| 243 |
+
"eval_loss": 0.6296263933181763,
|
| 244 |
+
"eval_runtime": 678.0284,
|
| 245 |
+
"eval_samples_per_second": 3.754,
|
| 246 |
+
"step": 6800
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 14.88,
|
| 250 |
+
"learning_rate": 0.00015663338088445076,
|
| 251 |
+
"loss": 0.3282,
|
| 252 |
+
"step": 7200
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 14.88,
|
| 256 |
+
"eval_cer": 0.16861123359830293,
|
| 257 |
+
"eval_loss": 0.6353692412376404,
|
| 258 |
+
"eval_runtime": 677.9049,
|
| 259 |
+
"eval_samples_per_second": 3.754,
|
| 260 |
+
"step": 7200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 15.7,
|
| 264 |
+
"learning_rate": 0.00014807417974322396,
|
| 265 |
+
"loss": 0.3105,
|
| 266 |
+
"step": 7600
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 15.7,
|
| 270 |
+
"eval_cer": 0.16284308000333694,
|
| 271 |
+
"eval_loss": 0.6332650780677795,
|
| 272 |
+
"eval_runtime": 678.3344,
|
| 273 |
+
"eval_samples_per_second": 3.752,
|
| 274 |
+
"step": 7600
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 16.53,
|
| 278 |
+
"learning_rate": 0.00013951497860199713,
|
| 279 |
+
"loss": 0.2899,
|
| 280 |
+
"step": 8000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 16.53,
|
| 284 |
+
"eval_cer": 0.16626345207307916,
|
| 285 |
+
"eval_loss": 0.6411857604980469,
|
| 286 |
+
"eval_runtime": 680.0199,
|
| 287 |
+
"eval_samples_per_second": 3.743,
|
| 288 |
+
"step": 8000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 17.36,
|
| 292 |
+
"learning_rate": 0.0001309557774607703,
|
| 293 |
+
"loss": 0.2769,
|
| 294 |
+
"step": 8400
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 17.36,
|
| 298 |
+
"eval_cer": 0.1635343050209155,
|
| 299 |
+
"eval_loss": 0.629165768623352,
|
| 300 |
+
"eval_runtime": 679.3365,
|
| 301 |
+
"eval_samples_per_second": 3.746,
|
| 302 |
+
"step": 8400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 18.18,
|
| 306 |
+
"learning_rate": 0.0001223965763195435,
|
| 307 |
+
"loss": 0.2685,
|
| 308 |
+
"step": 8800
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 18.18,
|
| 312 |
+
"eval_cer": 0.16531003825572943,
|
| 313 |
+
"eval_loss": 0.6566583514213562,
|
| 314 |
+
"eval_runtime": 677.0871,
|
| 315 |
+
"eval_samples_per_second": 3.759,
|
| 316 |
+
"step": 8800
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 19.01,
|
| 320 |
+
"learning_rate": 0.00011383737517831669,
|
| 321 |
+
"loss": 0.2481,
|
| 322 |
+
"step": 9200
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 19.01,
|
| 326 |
+
"eval_cer": 0.16478566065618705,
|
| 327 |
+
"eval_loss": 0.652852475643158,
|
| 328 |
+
"eval_runtime": 678.2412,
|
| 329 |
+
"eval_samples_per_second": 3.752,
|
| 330 |
+
"step": 9200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 19.83,
|
| 334 |
+
"learning_rate": 0.00010527817403708987,
|
| 335 |
+
"loss": 0.2355,
|
| 336 |
+
"step": 9600
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 19.83,
|
| 340 |
+
"eval_cer": 0.1621041842948909,
|
| 341 |
+
"eval_loss": 0.6581872701644897,
|
| 342 |
+
"eval_runtime": 679.7908,
|
| 343 |
+
"eval_samples_per_second": 3.744,
|
| 344 |
+
"step": 9600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 20.66,
|
| 348 |
+
"learning_rate": 9.671897289586304e-05,
|
| 349 |
+
"loss": 0.2266,
|
| 350 |
+
"step": 10000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 20.66,
|
| 354 |
+
"eval_cer": 0.1623544554219452,
|
| 355 |
+
"eval_loss": 0.6539607644081116,
|
| 356 |
+
"eval_runtime": 679.8539,
|
| 357 |
+
"eval_samples_per_second": 3.743,
|
| 358 |
+
"step": 10000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 21.49,
|
| 362 |
+
"learning_rate": 8.815977175463624e-05,
|
| 363 |
+
"loss": 0.2183,
|
| 364 |
+
"step": 10400
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 21.49,
|
| 368 |
+
"eval_cer": 0.1604952984781132,
|
| 369 |
+
"eval_loss": 0.6687005758285522,
|
| 370 |
+
"eval_runtime": 678.9594,
|
| 371 |
+
"eval_samples_per_second": 3.748,
|
| 372 |
+
"step": 10400
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 22.31,
|
| 376 |
+
"learning_rate": 7.960057061340941e-05,
|
| 377 |
+
"loss": 0.2089,
|
| 378 |
+
"step": 10800
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 22.31,
|
| 382 |
+
"eval_cer": 0.16017352131475765,
|
| 383 |
+
"eval_loss": 0.6894858479499817,
|
| 384 |
+
"eval_runtime": 681.5459,
|
| 385 |
+
"eval_samples_per_second": 3.734,
|
| 386 |
+
"step": 10800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 23.14,
|
| 390 |
+
"learning_rate": 7.10413694721826e-05,
|
| 391 |
+
"loss": 0.2016,
|
| 392 |
+
"step": 11200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 23.14,
|
| 396 |
+
"eval_cer": 0.1619850075677222,
|
| 397 |
+
"eval_loss": 0.6835413575172424,
|
| 398 |
+
"eval_runtime": 682.6349,
|
| 399 |
+
"eval_samples_per_second": 3.728,
|
| 400 |
+
"step": 11200
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 23.97,
|
| 404 |
+
"learning_rate": 6.248216833095577e-05,
|
| 405 |
+
"loss": 0.1904,
|
| 406 |
+
"step": 11600
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 23.97,
|
| 410 |
+
"eval_cer": 0.16073365193245062,
|
| 411 |
+
"eval_loss": 0.6923198103904724,
|
| 412 |
+
"eval_runtime": 681.4566,
|
| 413 |
+
"eval_samples_per_second": 3.735,
|
| 414 |
+
"step": 11600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 24.79,
|
| 418 |
+
"learning_rate": 5.392296718972896e-05,
|
| 419 |
+
"loss": 0.1865,
|
| 420 |
+
"step": 12000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 24.79,
|
| 424 |
+
"eval_cer": 0.15916051913382354,
|
| 425 |
+
"eval_loss": 0.6866386532783508,
|
| 426 |
+
"eval_runtime": 680.6783,
|
| 427 |
+
"eval_samples_per_second": 3.739,
|
| 428 |
+
"step": 12000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 25.62,
|
| 432 |
+
"learning_rate": 4.536376604850214e-05,
|
| 433 |
+
"loss": 0.1754,
|
| 434 |
+
"step": 12400
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 25.62,
|
| 438 |
+
"eval_cer": 0.15712259709923845,
|
| 439 |
+
"eval_loss": 0.693188488483429,
|
| 440 |
+
"eval_runtime": 682.8906,
|
| 441 |
+
"eval_samples_per_second": 3.727,
|
| 442 |
+
"step": 12400
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 26.45,
|
| 446 |
+
"learning_rate": 3.680456490727532e-05,
|
| 447 |
+
"loss": 0.1739,
|
| 448 |
+
"step": 12800
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 26.45,
|
| 452 |
+
"eval_cer": 0.15602617120928625,
|
| 453 |
+
"eval_loss": 0.6990784406661987,
|
| 454 |
+
"eval_runtime": 682.7736,
|
| 455 |
+
"eval_samples_per_second": 3.727,
|
| 456 |
+
"step": 12800
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 27.27,
|
| 460 |
+
"learning_rate": 2.82453637660485e-05,
|
| 461 |
+
"loss": 0.1657,
|
| 462 |
+
"step": 13200
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 27.27,
|
| 466 |
+
"eval_cer": 0.1563002776817743,
|
| 467 |
+
"eval_loss": 0.7062773108482361,
|
| 468 |
+
"eval_runtime": 681.4615,
|
| 469 |
+
"eval_samples_per_second": 3.735,
|
| 470 |
+
"step": 13200
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 28.1,
|
| 474 |
+
"learning_rate": 1.968616262482168e-05,
|
| 475 |
+
"loss": 0.1584,
|
| 476 |
+
"step": 13600
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 28.1,
|
| 480 |
+
"eval_cer": 0.15688424364490103,
|
| 481 |
+
"eval_loss": 0.7029954791069031,
|
| 482 |
+
"eval_runtime": 682.9133,
|
| 483 |
+
"eval_samples_per_second": 3.727,
|
| 484 |
+
"step": 13600
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 28.92,
|
| 488 |
+
"learning_rate": 1.1126961483594864e-05,
|
| 489 |
+
"loss": 0.1625,
|
| 490 |
+
"step": 14000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 28.92,
|
| 494 |
+
"eval_cer": 0.1568604082994673,
|
| 495 |
+
"eval_loss": 0.6991938352584839,
|
| 496 |
+
"eval_runtime": 683.4936,
|
| 497 |
+
"eval_samples_per_second": 3.724,
|
| 498 |
+
"step": 14000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 29.75,
|
| 502 |
+
"learning_rate": 2.5677603423680457e-06,
|
| 503 |
+
"loss": 0.1582,
|
| 504 |
+
"step": 14400
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 29.75,
|
| 508 |
+
"eval_cer": 0.156514795790678,
|
| 509 |
+
"eval_loss": 0.7013677358627319,
|
| 510 |
+
"eval_runtime": 682.2827,
|
| 511 |
+
"eval_samples_per_second": 3.73,
|
| 512 |
+
"step": 14400
|
| 513 |
+
}
|
| 514 |
+
],
|
| 515 |
+
"max_steps": 14520,
|
| 516 |
+
"num_train_epochs": 30,
|
| 517 |
+
"total_flos": 1.668287739327801e+20,
|
| 518 |
+
"trial_name": null,
|
| 519 |
+
"trial_params": null
|
| 520 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c38279018176bffb3fc2cf5bca32a27e73c7872edc7ccf2dc3af476e282ac04d
|
| 3 |
+
size 2287
|