initial model
Browse files- checkpoint-27580/config.json +108 -0
- checkpoint-27580/optimizer.pt +3 -0
- checkpoint-27580/preprocessor_config.json +10 -0
- checkpoint-27580/pytorch_model.bin +3 -0
- checkpoint-27580/rng_state.pth +3 -0
- checkpoint-27580/scheduler.pt +3 -0
- checkpoint-27580/trainer_state.json +1188 -0
- checkpoint-27580/training_args.bin +3 -0
- preprocessor_config.json +10 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- vocab.json +1 -0
checkpoint-27580/config.json
ADDED
|
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "wav2vec2-xls-r-parlaspeech-hr",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 768,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": true,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": true,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.0,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"gradient_checkpointing": false,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_size": 1024,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 4096,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.1,
|
| 63 |
+
"mask_feature_length": 10,
|
| 64 |
+
"mask_feature_min_masks": 0,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_masks": 2,
|
| 68 |
+
"mask_time_prob": 0.05,
|
| 69 |
+
"model_type": "wav2vec2",
|
| 70 |
+
"num_adapter_layers": 3,
|
| 71 |
+
"num_attention_heads": 16,
|
| 72 |
+
"num_codevector_groups": 2,
|
| 73 |
+
"num_codevectors_per_group": 320,
|
| 74 |
+
"num_conv_pos_embedding_groups": 16,
|
| 75 |
+
"num_conv_pos_embeddings": 128,
|
| 76 |
+
"num_feat_extract_layers": 7,
|
| 77 |
+
"num_hidden_layers": 24,
|
| 78 |
+
"num_negatives": 100,
|
| 79 |
+
"output_hidden_size": 1024,
|
| 80 |
+
"pad_token_id": 51,
|
| 81 |
+
"proj_codevector_dim": 768,
|
| 82 |
+
"tdnn_dilation": [
|
| 83 |
+
1,
|
| 84 |
+
2,
|
| 85 |
+
3,
|
| 86 |
+
1,
|
| 87 |
+
1
|
| 88 |
+
],
|
| 89 |
+
"tdnn_dim": [
|
| 90 |
+
512,
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
1500
|
| 95 |
+
],
|
| 96 |
+
"tdnn_kernel": [
|
| 97 |
+
5,
|
| 98 |
+
3,
|
| 99 |
+
3,
|
| 100 |
+
1,
|
| 101 |
+
1
|
| 102 |
+
],
|
| 103 |
+
"torch_dtype": "float32",
|
| 104 |
+
"transformers_version": "4.19.4",
|
| 105 |
+
"use_weighted_layer_sum": false,
|
| 106 |
+
"vocab_size": 52,
|
| 107 |
+
"xvector_output_dim": 512
|
| 108 |
+
}
|
checkpoint-27580/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5f8299516b9ad5674adea7e55b1c7a95d926f40cfd3442aefdc924a8f07d12e
|
| 3 |
+
size 2490585729
|
checkpoint-27580/preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
checkpoint-27580/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebcc0e6de114a1b32dd61d08501c2d612b86127839a6a5af5ad043f13e0c6c18
|
| 3 |
+
size 1262115053
|
checkpoint-27580/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf893b16f17408fec1ecf14a11c8fcb078ec42d8751a9a4bb4b8bd106a8940fa
|
| 3 |
+
size 14639
|
checkpoint-27580/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec32b769a697437dd8fe11d81712b4a1c7668000c56fb3a98d0dd5222985d4ae
|
| 3 |
+
size 627
|
checkpoint-27580/trainer_state.json
ADDED
|
@@ -0,0 +1,1188 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.09356310187531677,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-xls-r-slavic-pomak/checkpoint-27580",
|
| 4 |
+
"epoch": 34.99936588459099,
|
| 5 |
+
"global_step": 27580,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 0.00011999999999999999,
|
| 13 |
+
"loss": 5.3636,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.51,
|
| 18 |
+
"learning_rate": 0.00023999999999999998,
|
| 19 |
+
"loss": 1.2489,
|
| 20 |
+
"step": 400
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.76,
|
| 24 |
+
"learning_rate": 0.0002988921713441654,
|
| 25 |
+
"loss": 0.8015,
|
| 26 |
+
"step": 600
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"eval_cer": 0.07991897799971923,
|
| 31 |
+
"eval_loss": 0.34495434165000916,
|
| 32 |
+
"eval_runtime": 40.7653,
|
| 33 |
+
"eval_samples_per_second": 38.685,
|
| 34 |
+
"eval_steps_per_second": 4.857,
|
| 35 |
+
"eval_wer": 0.33421186011150533,
|
| 36 |
+
"step": 788
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 1.02,
|
| 40 |
+
"learning_rate": 0.00029667651403249626,
|
| 41 |
+
"loss": 0.6801,
|
| 42 |
+
"step": 800
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.27,
|
| 46 |
+
"learning_rate": 0.00029446085672082717,
|
| 47 |
+
"loss": 0.6282,
|
| 48 |
+
"step": 1000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.52,
|
| 52 |
+
"learning_rate": 0.00029224519940915803,
|
| 53 |
+
"loss": 0.5794,
|
| 54 |
+
"step": 1200
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.78,
|
| 58 |
+
"learning_rate": 0.0002900295420974889,
|
| 59 |
+
"loss": 0.5688,
|
| 60 |
+
"step": 1400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.0,
|
| 64 |
+
"eval_cer": 0.05737721356516856,
|
| 65 |
+
"eval_loss": 0.2721518874168396,
|
| 66 |
+
"eval_runtime": 42.4122,
|
| 67 |
+
"eval_samples_per_second": 37.183,
|
| 68 |
+
"eval_steps_per_second": 4.668,
|
| 69 |
+
"eval_wer": 0.22199695894576787,
|
| 70 |
+
"step": 1576
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 2.03,
|
| 74 |
+
"learning_rate": 0.0002878138847858198,
|
| 75 |
+
"loss": 0.5417,
|
| 76 |
+
"step": 1600
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.28,
|
| 80 |
+
"learning_rate": 0.00028559822747415065,
|
| 81 |
+
"loss": 0.4888,
|
| 82 |
+
"step": 1800
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 2.54,
|
| 86 |
+
"learning_rate": 0.0002833825701624815,
|
| 87 |
+
"loss": 0.5066,
|
| 88 |
+
"step": 2000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.79,
|
| 92 |
+
"learning_rate": 0.0002811669128508124,
|
| 93 |
+
"loss": 0.4869,
|
| 94 |
+
"step": 2200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.0,
|
| 98 |
+
"eval_cer": 0.053346168501694644,
|
| 99 |
+
"eval_loss": 0.2440285086631775,
|
| 100 |
+
"eval_runtime": 40.5216,
|
| 101 |
+
"eval_samples_per_second": 38.917,
|
| 102 |
+
"eval_steps_per_second": 4.886,
|
| 103 |
+
"eval_wer": 0.20293968575772936,
|
| 104 |
+
"step": 2364
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.05,
|
| 108 |
+
"learning_rate": 0.0002789512555391432,
|
| 109 |
+
"loss": 0.4609,
|
| 110 |
+
"step": 2400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.3,
|
| 114 |
+
"learning_rate": 0.00027673559822747414,
|
| 115 |
+
"loss": 0.4369,
|
| 116 |
+
"step": 2600
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.55,
|
| 120 |
+
"learning_rate": 0.000274519940915805,
|
| 121 |
+
"loss": 0.4398,
|
| 122 |
+
"step": 2800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.81,
|
| 126 |
+
"learning_rate": 0.00027230428360413585,
|
| 127 |
+
"loss": 0.4494,
|
| 128 |
+
"step": 3000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.0,
|
| 132 |
+
"eval_cer": 0.04845276056394521,
|
| 133 |
+
"eval_loss": 0.2370358109474182,
|
| 134 |
+
"eval_runtime": 40.4374,
|
| 135 |
+
"eval_samples_per_second": 38.999,
|
| 136 |
+
"eval_steps_per_second": 4.896,
|
| 137 |
+
"eval_wer": 0.18459199189052206,
|
| 138 |
+
"step": 3152
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 4.06,
|
| 142 |
+
"learning_rate": 0.00027008862629246676,
|
| 143 |
+
"loss": 0.433,
|
| 144 |
+
"step": 3200
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 4.31,
|
| 148 |
+
"learning_rate": 0.0002678729689807976,
|
| 149 |
+
"loss": 0.4144,
|
| 150 |
+
"step": 3400
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.57,
|
| 154 |
+
"learning_rate": 0.0002656573116691285,
|
| 155 |
+
"loss": 0.4018,
|
| 156 |
+
"step": 3600
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 4.82,
|
| 160 |
+
"learning_rate": 0.0002634416543574594,
|
| 161 |
+
"loss": 0.3969,
|
| 162 |
+
"step": 3800
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 5.0,
|
| 166 |
+
"eval_cer": 0.04522391352305317,
|
| 167 |
+
"eval_loss": 0.23047970235347748,
|
| 168 |
+
"eval_runtime": 41.6141,
|
| 169 |
+
"eval_samples_per_second": 37.896,
|
| 170 |
+
"eval_steps_per_second": 4.758,
|
| 171 |
+
"eval_wer": 0.1727318803852002,
|
| 172 |
+
"step": 3940
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 5.08,
|
| 176 |
+
"learning_rate": 0.0002612259970457902,
|
| 177 |
+
"loss": 0.3934,
|
| 178 |
+
"step": 4000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 5.33,
|
| 182 |
+
"learning_rate": 0.0002590103397341211,
|
| 183 |
+
"loss": 0.3676,
|
| 184 |
+
"step": 4200
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 5.58,
|
| 188 |
+
"learning_rate": 0.00025679468242245196,
|
| 189 |
+
"loss": 0.3725,
|
| 190 |
+
"step": 4400
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 5.84,
|
| 194 |
+
"learning_rate": 0.0002545790251107828,
|
| 195 |
+
"loss": 0.373,
|
| 196 |
+
"step": 4600
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 6.0,
|
| 200 |
+
"eval_cer": 0.0472494635300724,
|
| 201 |
+
"eval_loss": 0.21691733598709106,
|
| 202 |
+
"eval_runtime": 40.268,
|
| 203 |
+
"eval_samples_per_second": 39.163,
|
| 204 |
+
"eval_steps_per_second": 4.917,
|
| 205 |
+
"eval_wer": 0.17465788139888494,
|
| 206 |
+
"step": 4728
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.09,
|
| 210 |
+
"learning_rate": 0.0002523633677991137,
|
| 211 |
+
"loss": 0.3554,
|
| 212 |
+
"step": 4800
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.34,
|
| 216 |
+
"learning_rate": 0.0002501477104874446,
|
| 217 |
+
"loss": 0.332,
|
| 218 |
+
"step": 5000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.6,
|
| 222 |
+
"learning_rate": 0.00024793205317577544,
|
| 223 |
+
"loss": 0.3462,
|
| 224 |
+
"step": 5200
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.85,
|
| 228 |
+
"learning_rate": 0.00024571639586410635,
|
| 229 |
+
"loss": 0.3483,
|
| 230 |
+
"step": 5400
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 7.0,
|
| 234 |
+
"eval_cer": 0.04143352786635381,
|
| 235 |
+
"eval_loss": 0.22267436981201172,
|
| 236 |
+
"eval_runtime": 40.4667,
|
| 237 |
+
"eval_samples_per_second": 38.97,
|
| 238 |
+
"eval_steps_per_second": 4.893,
|
| 239 |
+
"eval_wer": 0.15701976685250887,
|
| 240 |
+
"step": 5516
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.11,
|
| 244 |
+
"learning_rate": 0.0002435007385524372,
|
| 245 |
+
"loss": 0.3532,
|
| 246 |
+
"step": 5600
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 7.36,
|
| 250 |
+
"learning_rate": 0.00024128508124076806,
|
| 251 |
+
"loss": 0.3249,
|
| 252 |
+
"step": 5800
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 7.61,
|
| 256 |
+
"learning_rate": 0.00023906942392909895,
|
| 257 |
+
"loss": 0.3436,
|
| 258 |
+
"step": 6000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 7.87,
|
| 262 |
+
"learning_rate": 0.00023685376661742983,
|
| 263 |
+
"loss": 0.31,
|
| 264 |
+
"step": 6200
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 8.0,
|
| 268 |
+
"eval_cer": 0.0421153961855484,
|
| 269 |
+
"eval_loss": 0.21147780120372772,
|
| 270 |
+
"eval_runtime": 40.4158,
|
| 271 |
+
"eval_samples_per_second": 39.019,
|
| 272 |
+
"eval_steps_per_second": 4.899,
|
| 273 |
+
"eval_wer": 0.15306639635073493,
|
| 274 |
+
"step": 6304
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 8.12,
|
| 278 |
+
"learning_rate": 0.0002346381093057607,
|
| 279 |
+
"loss": 0.3175,
|
| 280 |
+
"step": 6400
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 8.38,
|
| 284 |
+
"learning_rate": 0.00023242245199409157,
|
| 285 |
+
"loss": 0.2886,
|
| 286 |
+
"step": 6600
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 8.63,
|
| 290 |
+
"learning_rate": 0.00023020679468242246,
|
| 291 |
+
"loss": 0.2919,
|
| 292 |
+
"step": 6800
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 8.88,
|
| 296 |
+
"learning_rate": 0.0002279911373707533,
|
| 297 |
+
"loss": 0.3127,
|
| 298 |
+
"step": 7000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 9.0,
|
| 302 |
+
"eval_cer": 0.04123297836070834,
|
| 303 |
+
"eval_loss": 0.2211967408657074,
|
| 304 |
+
"eval_runtime": 40.3337,
|
| 305 |
+
"eval_samples_per_second": 39.099,
|
| 306 |
+
"eval_steps_per_second": 4.909,
|
| 307 |
+
"eval_wer": 0.1559047136340598,
|
| 308 |
+
"step": 7092
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 9.14,
|
| 312 |
+
"learning_rate": 0.00022577548005908417,
|
| 313 |
+
"loss": 0.3022,
|
| 314 |
+
"step": 7200
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 9.39,
|
| 318 |
+
"learning_rate": 0.00022355982274741506,
|
| 319 |
+
"loss": 0.2737,
|
| 320 |
+
"step": 7400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 9.64,
|
| 324 |
+
"learning_rate": 0.0002213441654357459,
|
| 325 |
+
"loss": 0.2681,
|
| 326 |
+
"step": 7600
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 9.9,
|
| 330 |
+
"learning_rate": 0.0002191285081240768,
|
| 331 |
+
"loss": 0.2795,
|
| 332 |
+
"step": 7800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 10.0,
|
| 336 |
+
"eval_cer": 0.038124461023203576,
|
| 337 |
+
"eval_loss": 0.21809443831443787,
|
| 338 |
+
"eval_runtime": 40.3044,
|
| 339 |
+
"eval_samples_per_second": 39.127,
|
| 340 |
+
"eval_steps_per_second": 4.913,
|
| 341 |
+
"eval_wer": 0.1395843892549417,
|
| 342 |
+
"step": 7880
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 10.15,
|
| 346 |
+
"learning_rate": 0.00021691285081240765,
|
| 347 |
+
"loss": 0.2714,
|
| 348 |
+
"step": 8000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 10.41,
|
| 352 |
+
"learning_rate": 0.00021469719350073854,
|
| 353 |
+
"loss": 0.2575,
|
| 354 |
+
"step": 8200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 10.66,
|
| 358 |
+
"learning_rate": 0.00021248153618906942,
|
| 359 |
+
"loss": 0.2758,
|
| 360 |
+
"step": 8400
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 10.91,
|
| 364 |
+
"learning_rate": 0.00021026587887740025,
|
| 365 |
+
"loss": 0.2676,
|
| 366 |
+
"step": 8600
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 11.0,
|
| 370 |
+
"eval_cer": 0.03782363676473537,
|
| 371 |
+
"eval_loss": 0.22525641322135925,
|
| 372 |
+
"eval_runtime": 40.3665,
|
| 373 |
+
"eval_samples_per_second": 39.067,
|
| 374 |
+
"eval_steps_per_second": 4.905,
|
| 375 |
+
"eval_wer": 0.1375570197668525,
|
| 376 |
+
"step": 8668
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 11.17,
|
| 380 |
+
"learning_rate": 0.00020805022156573114,
|
| 381 |
+
"loss": 0.2398,
|
| 382 |
+
"step": 8800
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 11.42,
|
| 386 |
+
"learning_rate": 0.00020583456425406202,
|
| 387 |
+
"loss": 0.2476,
|
| 388 |
+
"step": 9000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 11.67,
|
| 392 |
+
"learning_rate": 0.00020361890694239288,
|
| 393 |
+
"loss": 0.2491,
|
| 394 |
+
"step": 9200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 11.93,
|
| 398 |
+
"learning_rate": 0.00020140324963072376,
|
| 399 |
+
"loss": 0.2385,
|
| 400 |
+
"step": 9400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 12.0,
|
| 404 |
+
"eval_cer": 0.038024186270380844,
|
| 405 |
+
"eval_loss": 0.24045319855213165,
|
| 406 |
+
"eval_runtime": 40.4554,
|
| 407 |
+
"eval_samples_per_second": 38.981,
|
| 408 |
+
"eval_steps_per_second": 4.894,
|
| 409 |
+
"eval_wer": 0.1352255448555499,
|
| 410 |
+
"step": 9456
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 12.18,
|
| 414 |
+
"learning_rate": 0.00019918759231905465,
|
| 415 |
+
"loss": 0.2425,
|
| 416 |
+
"step": 9600
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 12.44,
|
| 420 |
+
"learning_rate": 0.0001969719350073855,
|
| 421 |
+
"loss": 0.2223,
|
| 422 |
+
"step": 9800
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 12.69,
|
| 426 |
+
"learning_rate": 0.0001947562776957164,
|
| 427 |
+
"loss": 0.2448,
|
| 428 |
+
"step": 10000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 12.94,
|
| 432 |
+
"learning_rate": 0.00019254062038404727,
|
| 433 |
+
"loss": 0.216,
|
| 434 |
+
"step": 10200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 13.0,
|
| 438 |
+
"eval_cer": 0.035657702103764316,
|
| 439 |
+
"eval_loss": 0.2175922691822052,
|
| 440 |
+
"eval_runtime": 40.4091,
|
| 441 |
+
"eval_samples_per_second": 39.026,
|
| 442 |
+
"eval_steps_per_second": 4.9,
|
| 443 |
+
"eval_wer": 0.12579827673593513,
|
| 444 |
+
"step": 10244
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 13.2,
|
| 448 |
+
"learning_rate": 0.0001903249630723781,
|
| 449 |
+
"loss": 0.2153,
|
| 450 |
+
"step": 10400
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 13.45,
|
| 454 |
+
"learning_rate": 0.00018810930576070899,
|
| 455 |
+
"loss": 0.2215,
|
| 456 |
+
"step": 10600
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 13.71,
|
| 460 |
+
"learning_rate": 0.00018589364844903987,
|
| 461 |
+
"loss": 0.2102,
|
| 462 |
+
"step": 10800
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 13.96,
|
| 466 |
+
"learning_rate": 0.00018367799113737073,
|
| 467 |
+
"loss": 0.2197,
|
| 468 |
+
"step": 11000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 14.0,
|
| 472 |
+
"eval_cer": 0.03569781200489341,
|
| 473 |
+
"eval_loss": 0.21506036818027496,
|
| 474 |
+
"eval_runtime": 40.487,
|
| 475 |
+
"eval_samples_per_second": 38.951,
|
| 476 |
+
"eval_steps_per_second": 4.89,
|
| 477 |
+
"eval_wer": 0.12630511910795741,
|
| 478 |
+
"step": 11032
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 14.21,
|
| 482 |
+
"learning_rate": 0.0001814623338257016,
|
| 483 |
+
"loss": 0.1977,
|
| 484 |
+
"step": 11200
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 14.47,
|
| 488 |
+
"learning_rate": 0.0001792466765140325,
|
| 489 |
+
"loss": 0.1966,
|
| 490 |
+
"step": 11400
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 14.72,
|
| 494 |
+
"learning_rate": 0.00017703101920236335,
|
| 495 |
+
"loss": 0.1966,
|
| 496 |
+
"step": 11600
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 14.97,
|
| 500 |
+
"learning_rate": 0.00017481536189069424,
|
| 501 |
+
"loss": 0.2036,
|
| 502 |
+
"step": 11800
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 15.0,
|
| 506 |
+
"eval_cer": 0.036981328841024404,
|
| 507 |
+
"eval_loss": 0.21347209811210632,
|
| 508 |
+
"eval_runtime": 40.2901,
|
| 509 |
+
"eval_samples_per_second": 39.141,
|
| 510 |
+
"eval_steps_per_second": 4.914,
|
| 511 |
+
"eval_wer": 0.12965027876330462,
|
| 512 |
+
"step": 11820
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 15.23,
|
| 516 |
+
"learning_rate": 0.00017259970457902512,
|
| 517 |
+
"loss": 0.178,
|
| 518 |
+
"step": 12000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 15.48,
|
| 522 |
+
"learning_rate": 0.00017038404726735595,
|
| 523 |
+
"loss": 0.1952,
|
| 524 |
+
"step": 12200
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 15.74,
|
| 528 |
+
"learning_rate": 0.00016816838995568683,
|
| 529 |
+
"loss": 0.1791,
|
| 530 |
+
"step": 12400
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 15.99,
|
| 534 |
+
"learning_rate": 0.0001659527326440177,
|
| 535 |
+
"loss": 0.1902,
|
| 536 |
+
"step": 12600
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 16.0,
|
| 540 |
+
"eval_cer": 0.037302208050057156,
|
| 541 |
+
"eval_loss": 0.22892530262470245,
|
| 542 |
+
"eval_runtime": 40.2075,
|
| 543 |
+
"eval_samples_per_second": 39.222,
|
| 544 |
+
"eval_steps_per_second": 4.924,
|
| 545 |
+
"eval_wer": 0.13127217435377597,
|
| 546 |
+
"step": 12608
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 16.24,
|
| 550 |
+
"learning_rate": 0.00016373707533234858,
|
| 551 |
+
"loss": 0.1775,
|
| 552 |
+
"step": 12800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 16.5,
|
| 556 |
+
"learning_rate": 0.00016152141802067946,
|
| 557 |
+
"loss": 0.1704,
|
| 558 |
+
"step": 13000
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 16.75,
|
| 562 |
+
"learning_rate": 0.00015930576070901032,
|
| 563 |
+
"loss": 0.1753,
|
| 564 |
+
"step": 13200
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 17.0,
|
| 568 |
+
"eval_cer": 0.03613902091731344,
|
| 569 |
+
"eval_loss": 0.23591692745685577,
|
| 570 |
+
"eval_runtime": 41.9181,
|
| 571 |
+
"eval_samples_per_second": 37.621,
|
| 572 |
+
"eval_steps_per_second": 4.723,
|
| 573 |
+
"eval_wer": 0.12083122149011657,
|
| 574 |
+
"step": 13396
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 17.01,
|
| 578 |
+
"learning_rate": 0.0001570901033973412,
|
| 579 |
+
"loss": 0.1875,
|
| 580 |
+
"step": 13400
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 17.26,
|
| 584 |
+
"learning_rate": 0.00015487444608567209,
|
| 585 |
+
"loss": 0.1639,
|
| 586 |
+
"step": 13600
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 17.51,
|
| 590 |
+
"learning_rate": 0.00015265878877400292,
|
| 591 |
+
"loss": 0.1637,
|
| 592 |
+
"step": 13800
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 17.77,
|
| 596 |
+
"learning_rate": 0.0001504431314623338,
|
| 597 |
+
"loss": 0.1679,
|
| 598 |
+
"step": 14000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 18.0,
|
| 602 |
+
"eval_cer": 0.03790385656699356,
|
| 603 |
+
"eval_loss": 0.22805896401405334,
|
| 604 |
+
"eval_runtime": 40.624,
|
| 605 |
+
"eval_samples_per_second": 38.819,
|
| 606 |
+
"eval_steps_per_second": 4.874,
|
| 607 |
+
"eval_wer": 0.12103395843892549,
|
| 608 |
+
"step": 14184
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 18.02,
|
| 612 |
+
"learning_rate": 0.00014822747415066468,
|
| 613 |
+
"loss": 0.1639,
|
| 614 |
+
"step": 14200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 18.27,
|
| 618 |
+
"learning_rate": 0.00014601181683899557,
|
| 619 |
+
"loss": 0.1567,
|
| 620 |
+
"step": 14400
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 18.53,
|
| 624 |
+
"learning_rate": 0.00014379615952732642,
|
| 625 |
+
"loss": 0.1548,
|
| 626 |
+
"step": 14600
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 18.78,
|
| 630 |
+
"learning_rate": 0.00014158050221565728,
|
| 631 |
+
"loss": 0.1558,
|
| 632 |
+
"step": 14800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 19.0,
|
| 636 |
+
"eval_cer": 0.035376932795860655,
|
| 637 |
+
"eval_loss": 0.23306047916412354,
|
| 638 |
+
"eval_runtime": 40.2559,
|
| 639 |
+
"eval_samples_per_second": 39.174,
|
| 640 |
+
"eval_steps_per_second": 4.919,
|
| 641 |
+
"eval_wer": 0.11910795742524075,
|
| 642 |
+
"step": 14972
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 19.04,
|
| 646 |
+
"learning_rate": 0.00013936484490398817,
|
| 647 |
+
"loss": 0.1471,
|
| 648 |
+
"step": 15000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 19.29,
|
| 652 |
+
"learning_rate": 0.00013714918759231905,
|
| 653 |
+
"loss": 0.1493,
|
| 654 |
+
"step": 15200
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 19.54,
|
| 658 |
+
"learning_rate": 0.0001349335302806499,
|
| 659 |
+
"loss": 0.1462,
|
| 660 |
+
"step": 15400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 19.8,
|
| 664 |
+
"learning_rate": 0.0001327178729689808,
|
| 665 |
+
"loss": 0.1411,
|
| 666 |
+
"step": 15600
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 20.0,
|
| 670 |
+
"eval_cer": 0.03383270160239055,
|
| 671 |
+
"eval_loss": 0.23826096951961517,
|
| 672 |
+
"eval_runtime": 40.4774,
|
| 673 |
+
"eval_samples_per_second": 38.96,
|
| 674 |
+
"eval_steps_per_second": 4.892,
|
| 675 |
+
"eval_wer": 0.11525595539787126,
|
| 676 |
+
"step": 15760
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 20.05,
|
| 680 |
+
"learning_rate": 0.00013050221565731165,
|
| 681 |
+
"loss": 0.1495,
|
| 682 |
+
"step": 15800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 20.3,
|
| 686 |
+
"learning_rate": 0.00012828655834564253,
|
| 687 |
+
"loss": 0.141,
|
| 688 |
+
"step": 16000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 20.56,
|
| 692 |
+
"learning_rate": 0.00012607090103397342,
|
| 693 |
+
"loss": 0.1279,
|
| 694 |
+
"step": 16200
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 20.81,
|
| 698 |
+
"learning_rate": 0.00012385524372230427,
|
| 699 |
+
"loss": 0.1438,
|
| 700 |
+
"step": 16400
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 21.0,
|
| 704 |
+
"eval_cer": 0.033652207047309625,
|
| 705 |
+
"eval_loss": 0.2469082772731781,
|
| 706 |
+
"eval_runtime": 40.7985,
|
| 707 |
+
"eval_samples_per_second": 38.653,
|
| 708 |
+
"eval_steps_per_second": 4.853,
|
| 709 |
+
"eval_wer": 0.11373542828180436,
|
| 710 |
+
"step": 16548
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 21.07,
|
| 714 |
+
"learning_rate": 0.00012163958641063514,
|
| 715 |
+
"loss": 0.1426,
|
| 716 |
+
"step": 16600
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 21.32,
|
| 720 |
+
"learning_rate": 0.00011942392909896601,
|
| 721 |
+
"loss": 0.1313,
|
| 722 |
+
"step": 16800
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 21.57,
|
| 726 |
+
"learning_rate": 0.0001172082717872969,
|
| 727 |
+
"loss": 0.1349,
|
| 728 |
+
"step": 17000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 21.83,
|
| 732 |
+
"learning_rate": 0.00011499261447562776,
|
| 733 |
+
"loss": 0.1262,
|
| 734 |
+
"step": 17200
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 22.0,
|
| 738 |
+
"eval_cer": 0.03353187734392235,
|
| 739 |
+
"eval_loss": 0.25218185782432556,
|
| 740 |
+
"eval_runtime": 40.3798,
|
| 741 |
+
"eval_samples_per_second": 39.054,
|
| 742 |
+
"eval_steps_per_second": 4.903,
|
| 743 |
+
"eval_wer": 0.11211353269133299,
|
| 744 |
+
"step": 17336
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 22.08,
|
| 748 |
+
"learning_rate": 0.00011277695716395863,
|
| 749 |
+
"loss": 0.1223,
|
| 750 |
+
"step": 17400
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 22.33,
|
| 754 |
+
"learning_rate": 0.00011056129985228951,
|
| 755 |
+
"loss": 0.1162,
|
| 756 |
+
"step": 17600
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 22.59,
|
| 760 |
+
"learning_rate": 0.00010834564254062038,
|
| 761 |
+
"loss": 0.1181,
|
| 762 |
+
"step": 17800
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 22.84,
|
| 766 |
+
"learning_rate": 0.00010612998522895124,
|
| 767 |
+
"loss": 0.1109,
|
| 768 |
+
"step": 18000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 23.0,
|
| 772 |
+
"eval_cer": 0.032007701101016785,
|
| 773 |
+
"eval_loss": 0.2507190406322479,
|
| 774 |
+
"eval_runtime": 40.4365,
|
| 775 |
+
"eval_samples_per_second": 38.999,
|
| 776 |
+
"eval_steps_per_second": 4.897,
|
| 777 |
+
"eval_wer": 0.10562595032944753,
|
| 778 |
+
"step": 18124
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 23.1,
|
| 782 |
+
"learning_rate": 0.00010391432791728212,
|
| 783 |
+
"loss": 0.1212,
|
| 784 |
+
"step": 18200
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 23.35,
|
| 788 |
+
"learning_rate": 0.00010169867060561299,
|
| 789 |
+
"loss": 0.1089,
|
| 790 |
+
"step": 18400
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 23.6,
|
| 794 |
+
"learning_rate": 9.948301329394386e-05,
|
| 795 |
+
"loss": 0.1055,
|
| 796 |
+
"step": 18600
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 23.86,
|
| 800 |
+
"learning_rate": 9.726735598227472e-05,
|
| 801 |
+
"loss": 0.122,
|
| 802 |
+
"step": 18800
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 24.0,
|
| 806 |
+
"eval_cer": 0.03224836050779135,
|
| 807 |
+
"eval_loss": 0.2498846799135208,
|
| 808 |
+
"eval_runtime": 40.3499,
|
| 809 |
+
"eval_samples_per_second": 39.083,
|
| 810 |
+
"eval_steps_per_second": 4.907,
|
| 811 |
+
"eval_wer": 0.10876837303598581,
|
| 812 |
+
"step": 18912
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 24.11,
|
| 816 |
+
"learning_rate": 9.50516986706056e-05,
|
| 817 |
+
"loss": 0.1089,
|
| 818 |
+
"step": 19000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 24.37,
|
| 822 |
+
"learning_rate": 9.283604135893648e-05,
|
| 823 |
+
"loss": 0.1053,
|
| 824 |
+
"step": 19200
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 24.62,
|
| 828 |
+
"learning_rate": 9.062038404726735e-05,
|
| 829 |
+
"loss": 0.1104,
|
| 830 |
+
"step": 19400
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 24.87,
|
| 834 |
+
"learning_rate": 8.840472673559823e-05,
|
| 835 |
+
"loss": 0.1012,
|
| 836 |
+
"step": 19600
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 25.0,
|
| 840 |
+
"eval_cer": 0.031045063473918538,
|
| 841 |
+
"eval_loss": 0.24705451726913452,
|
| 842 |
+
"eval_runtime": 40.2856,
|
| 843 |
+
"eval_samples_per_second": 39.146,
|
| 844 |
+
"eval_steps_per_second": 4.915,
|
| 845 |
+
"eval_wer": 0.10258489609731374,
|
| 846 |
+
"step": 19700
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 25.13,
|
| 850 |
+
"learning_rate": 8.618906942392909e-05,
|
| 851 |
+
"loss": 0.1187,
|
| 852 |
+
"step": 19800
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 25.38,
|
| 856 |
+
"learning_rate": 8.397341211225996e-05,
|
| 857 |
+
"loss": 0.1013,
|
| 858 |
+
"step": 20000
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 25.63,
|
| 862 |
+
"learning_rate": 8.175775480059084e-05,
|
| 863 |
+
"loss": 0.1013,
|
| 864 |
+
"step": 20200
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 25.89,
|
| 868 |
+
"learning_rate": 7.954209748892171e-05,
|
| 869 |
+
"loss": 0.1005,
|
| 870 |
+
"step": 20400
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 26.0,
|
| 874 |
+
"eval_cer": 0.03116539317730582,
|
| 875 |
+
"eval_loss": 0.25858306884765625,
|
| 876 |
+
"eval_runtime": 40.7124,
|
| 877 |
+
"eval_samples_per_second": 38.735,
|
| 878 |
+
"eval_steps_per_second": 4.863,
|
| 879 |
+
"eval_wer": 0.1042067916877851,
|
| 880 |
+
"step": 20488
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 26.14,
|
| 884 |
+
"learning_rate": 7.732644017725257e-05,
|
| 885 |
+
"loss": 0.0904,
|
| 886 |
+
"step": 20600
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 26.4,
|
| 890 |
+
"learning_rate": 7.511078286558345e-05,
|
| 891 |
+
"loss": 0.0923,
|
| 892 |
+
"step": 20800
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 26.65,
|
| 896 |
+
"learning_rate": 7.289512555391432e-05,
|
| 897 |
+
"loss": 0.0956,
|
| 898 |
+
"step": 21000
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 26.9,
|
| 902 |
+
"learning_rate": 7.06794682422452e-05,
|
| 903 |
+
"loss": 0.0902,
|
| 904 |
+
"step": 21200
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 27.0,
|
| 908 |
+
"eval_cer": 0.030563744660369412,
|
| 909 |
+
"eval_loss": 0.25710704922676086,
|
| 910 |
+
"eval_runtime": 40.3222,
|
| 911 |
+
"eval_samples_per_second": 39.11,
|
| 912 |
+
"eval_steps_per_second": 4.91,
|
| 913 |
+
"eval_wer": 0.10146984287886468,
|
| 914 |
+
"step": 21276
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 27.16,
|
| 918 |
+
"learning_rate": 6.846381093057607e-05,
|
| 919 |
+
"loss": 0.0812,
|
| 920 |
+
"step": 21400
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 27.41,
|
| 924 |
+
"learning_rate": 6.624815361890694e-05,
|
| 925 |
+
"loss": 0.0866,
|
| 926 |
+
"step": 21600
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 27.66,
|
| 930 |
+
"learning_rate": 6.40324963072378e-05,
|
| 931 |
+
"loss": 0.0907,
|
| 932 |
+
"step": 21800
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 27.92,
|
| 936 |
+
"learning_rate": 6.181683899556868e-05,
|
| 937 |
+
"loss": 0.0829,
|
| 938 |
+
"step": 22000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 28.0,
|
| 942 |
+
"eval_cer": 0.03272967932134047,
|
| 943 |
+
"eval_loss": 0.25568026304244995,
|
| 944 |
+
"eval_runtime": 40.5085,
|
| 945 |
+
"eval_samples_per_second": 38.93,
|
| 946 |
+
"eval_steps_per_second": 4.888,
|
| 947 |
+
"eval_wer": 0.1036999493157628,
|
| 948 |
+
"step": 22064
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 28.17,
|
| 952 |
+
"learning_rate": 5.960118168389955e-05,
|
| 953 |
+
"loss": 0.0857,
|
| 954 |
+
"step": 22200
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 28.43,
|
| 958 |
+
"learning_rate": 5.7385524372230425e-05,
|
| 959 |
+
"loss": 0.0826,
|
| 960 |
+
"step": 22400
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 28.68,
|
| 964 |
+
"learning_rate": 5.5169867060561295e-05,
|
| 965 |
+
"loss": 0.0817,
|
| 966 |
+
"step": 22600
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 28.93,
|
| 970 |
+
"learning_rate": 5.2954209748892166e-05,
|
| 971 |
+
"loss": 0.0773,
|
| 972 |
+
"step": 22800
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 29.0,
|
| 976 |
+
"eval_cer": 0.03084451396827307,
|
| 977 |
+
"eval_loss": 0.25824475288391113,
|
| 978 |
+
"eval_runtime": 40.434,
|
| 979 |
+
"eval_samples_per_second": 39.002,
|
| 980 |
+
"eval_steps_per_second": 4.897,
|
| 981 |
+
"eval_wer": 0.09761784085149519,
|
| 982 |
+
"step": 22852
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 29.19,
|
| 986 |
+
"learning_rate": 5.073855243722304e-05,
|
| 987 |
+
"loss": 0.0888,
|
| 988 |
+
"step": 23000
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 29.44,
|
| 992 |
+
"learning_rate": 4.852289512555391e-05,
|
| 993 |
+
"loss": 0.0782,
|
| 994 |
+
"step": 23200
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 29.69,
|
| 998 |
+
"learning_rate": 4.6307237813884784e-05,
|
| 999 |
+
"loss": 0.0757,
|
| 1000 |
+
"step": 23400
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 29.95,
|
| 1004 |
+
"learning_rate": 4.4091580502215655e-05,
|
| 1005 |
+
"loss": 0.0795,
|
| 1006 |
+
"step": 23600
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 30.0,
|
| 1010 |
+
"eval_cer": 0.02968132683552935,
|
| 1011 |
+
"eval_loss": 0.25763553380966187,
|
| 1012 |
+
"eval_runtime": 40.3436,
|
| 1013 |
+
"eval_samples_per_second": 39.089,
|
| 1014 |
+
"eval_steps_per_second": 4.908,
|
| 1015 |
+
"eval_wer": 0.09670552458185504,
|
| 1016 |
+
"step": 23640
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 30.2,
|
| 1020 |
+
"learning_rate": 4.1875923190546526e-05,
|
| 1021 |
+
"loss": 0.0678,
|
| 1022 |
+
"step": 23800
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 30.46,
|
| 1026 |
+
"learning_rate": 3.9660265878877396e-05,
|
| 1027 |
+
"loss": 0.0708,
|
| 1028 |
+
"step": 24000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 30.71,
|
| 1032 |
+
"learning_rate": 3.744460856720827e-05,
|
| 1033 |
+
"loss": 0.0764,
|
| 1034 |
+
"step": 24200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 30.96,
|
| 1038 |
+
"learning_rate": 3.522895125553914e-05,
|
| 1039 |
+
"loss": 0.0737,
|
| 1040 |
+
"step": 24400
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 31.0,
|
| 1044 |
+
"eval_cer": 0.02891923871407657,
|
| 1045 |
+
"eval_loss": 0.2654246985912323,
|
| 1046 |
+
"eval_runtime": 40.3564,
|
| 1047 |
+
"eval_samples_per_second": 39.077,
|
| 1048 |
+
"eval_steps_per_second": 4.906,
|
| 1049 |
+
"eval_wer": 0.09508362899138367,
|
| 1050 |
+
"step": 24428
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 31.22,
|
| 1054 |
+
"learning_rate": 3.301329394387001e-05,
|
| 1055 |
+
"loss": 0.0688,
|
| 1056 |
+
"step": 24600
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 31.47,
|
| 1060 |
+
"learning_rate": 3.0797636632200885e-05,
|
| 1061 |
+
"loss": 0.0669,
|
| 1062 |
+
"step": 24800
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 31.73,
|
| 1066 |
+
"learning_rate": 2.8581979320531756e-05,
|
| 1067 |
+
"loss": 0.0679,
|
| 1068 |
+
"step": 25000
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 31.98,
|
| 1072 |
+
"learning_rate": 2.6366322008862626e-05,
|
| 1073 |
+
"loss": 0.0744,
|
| 1074 |
+
"step": 25200
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 32.0,
|
| 1078 |
+
"eval_cer": 0.029741491687222992,
|
| 1079 |
+
"eval_loss": 0.26390764117240906,
|
| 1080 |
+
"eval_runtime": 40.4656,
|
| 1081 |
+
"eval_samples_per_second": 38.971,
|
| 1082 |
+
"eval_steps_per_second": 4.893,
|
| 1083 |
+
"eval_wer": 0.09640141915864166,
|
| 1084 |
+
"step": 25216
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 32.23,
|
| 1088 |
+
"learning_rate": 2.4150664697193497e-05,
|
| 1089 |
+
"loss": 0.0618,
|
| 1090 |
+
"step": 25400
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 32.49,
|
| 1094 |
+
"learning_rate": 2.1935007385524368e-05,
|
| 1095 |
+
"loss": 0.0675,
|
| 1096 |
+
"step": 25600
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 32.74,
|
| 1100 |
+
"learning_rate": 1.971935007385524e-05,
|
| 1101 |
+
"loss": 0.0637,
|
| 1102 |
+
"step": 25800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 32.99,
|
| 1106 |
+
"learning_rate": 1.7503692762186116e-05,
|
| 1107 |
+
"loss": 0.0656,
|
| 1108 |
+
"step": 26000
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 33.0,
|
| 1112 |
+
"eval_cer": 0.029300282774802962,
|
| 1113 |
+
"eval_loss": 0.26851221919059753,
|
| 1114 |
+
"eval_runtime": 40.4956,
|
| 1115 |
+
"eval_samples_per_second": 38.943,
|
| 1116 |
+
"eval_steps_per_second": 4.889,
|
| 1117 |
+
"eval_wer": 0.09488089204257476,
|
| 1118 |
+
"step": 26004
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 33.25,
|
| 1122 |
+
"learning_rate": 1.5288035450516986e-05,
|
| 1123 |
+
"loss": 0.0639,
|
| 1124 |
+
"step": 26200
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 33.5,
|
| 1128 |
+
"learning_rate": 1.3072378138847857e-05,
|
| 1129 |
+
"loss": 0.0615,
|
| 1130 |
+
"step": 26400
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 33.76,
|
| 1134 |
+
"learning_rate": 1.0856720827178729e-05,
|
| 1135 |
+
"loss": 0.0661,
|
| 1136 |
+
"step": 26600
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 34.0,
|
| 1140 |
+
"eval_cer": 0.029079678318592945,
|
| 1141 |
+
"eval_loss": 0.2691967487335205,
|
| 1142 |
+
"eval_runtime": 40.2387,
|
| 1143 |
+
"eval_samples_per_second": 39.191,
|
| 1144 |
+
"eval_steps_per_second": 4.921,
|
| 1145 |
+
"eval_wer": 0.0937658388241257,
|
| 1146 |
+
"step": 26792
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 34.01,
|
| 1150 |
+
"learning_rate": 8.6410635155096e-06,
|
| 1151 |
+
"loss": 0.0608,
|
| 1152 |
+
"step": 26800
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 34.26,
|
| 1156 |
+
"learning_rate": 6.425406203840472e-06,
|
| 1157 |
+
"loss": 0.0637,
|
| 1158 |
+
"step": 27000
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 34.52,
|
| 1162 |
+
"learning_rate": 4.209748892171343e-06,
|
| 1163 |
+
"loss": 0.057,
|
| 1164 |
+
"step": 27200
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 34.77,
|
| 1168 |
+
"learning_rate": 1.9940915805022156e-06,
|
| 1169 |
+
"loss": 0.0603,
|
| 1170 |
+
"step": 27400
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 35.0,
|
| 1174 |
+
"eval_cer": 0.028899183763512023,
|
| 1175 |
+
"eval_loss": 0.26925861835479736,
|
| 1176 |
+
"eval_runtime": 40.4005,
|
| 1177 |
+
"eval_samples_per_second": 39.034,
|
| 1178 |
+
"eval_steps_per_second": 4.901,
|
| 1179 |
+
"eval_wer": 0.09356310187531677,
|
| 1180 |
+
"step": 27580
|
| 1181 |
+
}
|
| 1182 |
+
],
|
| 1183 |
+
"max_steps": 27580,
|
| 1184 |
+
"num_train_epochs": 35,
|
| 1185 |
+
"total_flos": 3.481960163555136e+19,
|
| 1186 |
+
"trial_name": null,
|
| 1187 |
+
"trial_params": null
|
| 1188 |
+
}
|
checkpoint-27580/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c92f16cf911fa8458cf07acc2550bc466107aef785e00031d56f06c4c769ab7c
|
| 3 |
+
size 3259
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "replace_word_delimiter_char": " ", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2Processor"}
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"|": 0, "a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6, "g": 7, "h": 8, "i": 9, "j": 10, "k": 11, "l": 12, "m": 13, "n": 14, "o": 15, "p": 16, "r": 17, "s": 18, "t": 19, "u": 20, "v": 21, "x": 22, "y": 23, "z": 24, "à": 25, "á": 26, "æ": 27, "ç": 28, "é": 29, "ì": 30, "í": 31, "ò": 32, "ó": 33, "ø": 34, "ú": 35, "ü": 36, "ý": 37, "č": 38, "ļ": 39, "ņ": 40, "ş": 41, "š": 42, "ž": 43, "ǘ": 44, "ǧ": 45, "ǵ": 46, "ǽ": 47, "ǿ": 48, "ό": 49, "[UNK]": 50, "[PAD]": 51}
|