Xinbo Yang commited on
Commit ·
1f9110e
1
Parent(s): 6261052
add model
Browse files- config.json +116 -0
- model.safetensors +3 -0
- preprocessor_config.json +10 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +49 -0
- trainer_state.json +474 -0
- training_args.bin +3 -0
- vocab.json +35 -0
config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-english",
|
| 3 |
+
"activation_dropout": 0.05,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 256,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": true,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.05,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.05,
|
| 58 |
+
"hidden_size": 1024,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 4096,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.05,
|
| 63 |
+
"mask_channel_length": 10,
|
| 64 |
+
"mask_channel_min_space": 1,
|
| 65 |
+
"mask_channel_other": 0.0,
|
| 66 |
+
"mask_channel_prob": 0.0,
|
| 67 |
+
"mask_channel_selection": "static",
|
| 68 |
+
"mask_feature_length": 10,
|
| 69 |
+
"mask_feature_min_masks": 0,
|
| 70 |
+
"mask_feature_prob": 0.0,
|
| 71 |
+
"mask_time_length": 10,
|
| 72 |
+
"mask_time_min_masks": 2,
|
| 73 |
+
"mask_time_min_space": 1,
|
| 74 |
+
"mask_time_other": 0.0,
|
| 75 |
+
"mask_time_prob": 0.05,
|
| 76 |
+
"mask_time_selection": "static",
|
| 77 |
+
"model_type": "wav2vec2",
|
| 78 |
+
"num_adapter_layers": 3,
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_codevector_groups": 2,
|
| 81 |
+
"num_codevectors_per_group": 320,
|
| 82 |
+
"num_conv_pos_embedding_groups": 16,
|
| 83 |
+
"num_conv_pos_embeddings": 128,
|
| 84 |
+
"num_feat_extract_layers": 7,
|
| 85 |
+
"num_hidden_layers": 24,
|
| 86 |
+
"num_negatives": 100,
|
| 87 |
+
"output_hidden_size": 1024,
|
| 88 |
+
"pad_token_id": 0,
|
| 89 |
+
"proj_codevector_dim": 256,
|
| 90 |
+
"tdnn_dilation": [
|
| 91 |
+
1,
|
| 92 |
+
2,
|
| 93 |
+
3,
|
| 94 |
+
1,
|
| 95 |
+
1
|
| 96 |
+
],
|
| 97 |
+
"tdnn_dim": [
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
1500
|
| 103 |
+
],
|
| 104 |
+
"tdnn_kernel": [
|
| 105 |
+
5,
|
| 106 |
+
3,
|
| 107 |
+
3,
|
| 108 |
+
1,
|
| 109 |
+
1
|
| 110 |
+
],
|
| 111 |
+
"torch_dtype": "float32",
|
| 112 |
+
"transformers_version": "4.48.3",
|
| 113 |
+
"use_weighted_layer_sum": false,
|
| 114 |
+
"vocab_size": 33,
|
| 115 |
+
"xvector_output_dim": 512
|
| 116 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a839a2ffaca65ecda84bcdcbf7a5691ae26b97a700e3302f1d903fd511eca30a
|
| 3 |
+
size 1261942780
|
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": "Wav2Vec2ProcessorWithLM",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:122306bb9b69b9f84306a0b76c3e47a10645c12d797561040e464ffae9a81093
|
| 3 |
+
size 14308
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a12bdff223423ea477b57dc354830e99828b709c7d95b535068964f89600921b
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "<pad>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<pad>",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": true,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": false
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": true,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": false
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": true,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": true,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": false
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": true,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": true,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": false
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"bos_token": "<s>",
|
| 37 |
+
"clean_up_tokenization_spaces": false,
|
| 38 |
+
"do_lower_case": false,
|
| 39 |
+
"eos_token": "</s>",
|
| 40 |
+
"extra_special_tokens": {},
|
| 41 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 42 |
+
"pad_token": "<pad>",
|
| 43 |
+
"processor_class": "Wav2Vec2Processor",
|
| 44 |
+
"replace_word_delimiter_char": " ",
|
| 45 |
+
"target_lang": null,
|
| 46 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 47 |
+
"unk_token": "<unk>",
|
| 48 |
+
"word_delimiter_token": "|"
|
| 49 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,474 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.18050541516245489,
|
| 3 |
+
"best_model_checkpoint": "/content/drive/MyDrive/Fine tune/wav2vec2_R/TestModel6/checkpoint-198",
|
| 4 |
+
"epoch": 27.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 243,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 2.6207005977630615,
|
| 14 |
+
"learning_rate": 2.0454545454545457e-05,
|
| 15 |
+
"loss": 1.4468,
|
| 16 |
+
"step": 9
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_loss": 0.9447698593139648,
|
| 21 |
+
"eval_runtime": 7.468,
|
| 22 |
+
"eval_samples_per_second": 1.339,
|
| 23 |
+
"eval_steps_per_second": 0.268,
|
| 24 |
+
"eval_wer": 0.3458483754512635,
|
| 25 |
+
"step": 9
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"grad_norm": 0.9180173873901367,
|
| 30 |
+
"learning_rate": 4.0909090909090915e-05,
|
| 31 |
+
"loss": 1.0187,
|
| 32 |
+
"step": 18
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"eval_loss": 0.7252597808837891,
|
| 37 |
+
"eval_runtime": 7.0648,
|
| 38 |
+
"eval_samples_per_second": 1.415,
|
| 39 |
+
"eval_steps_per_second": 0.283,
|
| 40 |
+
"eval_wer": 0.2772563176895307,
|
| 41 |
+
"step": 18
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"grad_norm": 1.0308139324188232,
|
| 46 |
+
"learning_rate": 4.9978417404589105e-05,
|
| 47 |
+
"loss": 0.7962,
|
| 48 |
+
"step": 27
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"eval_loss": 0.5624670386314392,
|
| 53 |
+
"eval_runtime": 7.0549,
|
| 54 |
+
"eval_samples_per_second": 1.417,
|
| 55 |
+
"eval_steps_per_second": 0.283,
|
| 56 |
+
"eval_wer": 0.30541516245487366,
|
| 57 |
+
"step": 27
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 1.6524184942245483,
|
| 62 |
+
"learning_rate": 4.983095894354858e-05,
|
| 63 |
+
"loss": 0.6591,
|
| 64 |
+
"step": 36
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_loss": 0.49896639585494995,
|
| 69 |
+
"eval_runtime": 6.9528,
|
| 70 |
+
"eval_samples_per_second": 1.438,
|
| 71 |
+
"eval_steps_per_second": 0.288,
|
| 72 |
+
"eval_wer": 0.24332129963898916,
|
| 73 |
+
"step": 36
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"grad_norm": 1.9104937314987183,
|
| 78 |
+
"learning_rate": 4.954463568772415e-05,
|
| 79 |
+
"loss": 0.5742,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.0,
|
| 84 |
+
"eval_loss": 0.4727010130882263,
|
| 85 |
+
"eval_runtime": 7.6581,
|
| 86 |
+
"eval_samples_per_second": 1.306,
|
| 87 |
+
"eval_steps_per_second": 0.261,
|
| 88 |
+
"eval_wer": 0.23249097472924188,
|
| 89 |
+
"step": 45
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"grad_norm": 0.7306864261627197,
|
| 94 |
+
"learning_rate": 4.912104887290587e-05,
|
| 95 |
+
"loss": 0.5231,
|
| 96 |
+
"step": 54
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 6.0,
|
| 100 |
+
"eval_loss": 0.4486842155456543,
|
| 101 |
+
"eval_runtime": 7.182,
|
| 102 |
+
"eval_samples_per_second": 1.392,
|
| 103 |
+
"eval_steps_per_second": 0.278,
|
| 104 |
+
"eval_wer": 0.21444043321299638,
|
| 105 |
+
"step": 54
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"grad_norm": 0.8729464411735535,
|
| 110 |
+
"learning_rate": 4.856256736838498e-05,
|
| 111 |
+
"loss": 0.4567,
|
| 112 |
+
"step": 63
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 7.0,
|
| 116 |
+
"eval_loss": 0.45621854066848755,
|
| 117 |
+
"eval_runtime": 7.0189,
|
| 118 |
+
"eval_samples_per_second": 1.425,
|
| 119 |
+
"eval_steps_per_second": 0.285,
|
| 120 |
+
"eval_wer": 0.21516245487364621,
|
| 121 |
+
"step": 63
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 0.9639649987220764,
|
| 126 |
+
"learning_rate": 4.787231442927587e-05,
|
| 127 |
+
"loss": 0.4934,
|
| 128 |
+
"step": 72
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_loss": 0.4438972473144531,
|
| 133 |
+
"eval_runtime": 7.0861,
|
| 134 |
+
"eval_samples_per_second": 1.411,
|
| 135 |
+
"eval_steps_per_second": 0.282,
|
| 136 |
+
"eval_wer": 0.2108303249097473,
|
| 137 |
+
"step": 72
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"grad_norm": 1.3746402263641357,
|
| 142 |
+
"learning_rate": 4.7054150230006604e-05,
|
| 143 |
+
"loss": 0.4516,
|
| 144 |
+
"step": 81
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.0,
|
| 148 |
+
"eval_loss": 0.43273186683654785,
|
| 149 |
+
"eval_runtime": 7.1786,
|
| 150 |
+
"eval_samples_per_second": 1.393,
|
| 151 |
+
"eval_steps_per_second": 0.279,
|
| 152 |
+
"eval_wer": 0.2043321299638989,
|
| 153 |
+
"step": 81
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 10.0,
|
| 157 |
+
"grad_norm": 0.5043421387672424,
|
| 158 |
+
"learning_rate": 4.61126502766577e-05,
|
| 159 |
+
"loss": 0.4069,
|
| 160 |
+
"step": 90
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 10.0,
|
| 164 |
+
"eval_loss": 0.43151745200157166,
|
| 165 |
+
"eval_runtime": 6.9477,
|
| 166 |
+
"eval_samples_per_second": 1.439,
|
| 167 |
+
"eval_steps_per_second": 0.288,
|
| 168 |
+
"eval_wer": 0.1956678700361011,
|
| 169 |
+
"step": 90
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 11.0,
|
| 173 |
+
"grad_norm": 1.8196316957473755,
|
| 174 |
+
"learning_rate": 4.50530798188761e-05,
|
| 175 |
+
"loss": 0.4235,
|
| 176 |
+
"step": 99
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 11.0,
|
| 180 |
+
"eval_loss": 0.4364844858646393,
|
| 181 |
+
"eval_runtime": 7.1022,
|
| 182 |
+
"eval_samples_per_second": 1.408,
|
| 183 |
+
"eval_steps_per_second": 0.282,
|
| 184 |
+
"eval_wer": 0.20144404332129964,
|
| 185 |
+
"step": 99
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 0.22248859703540802,
|
| 190 |
+
"learning_rate": 4.388136440446337e-05,
|
| 191 |
+
"loss": 0.4004,
|
| 192 |
+
"step": 108
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_loss": 0.43696457147598267,
|
| 197 |
+
"eval_runtime": 7.1886,
|
| 198 |
+
"eval_samples_per_second": 1.391,
|
| 199 |
+
"eval_steps_per_second": 0.278,
|
| 200 |
+
"eval_wer": 0.19855595667870035,
|
| 201 |
+
"step": 108
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 13.0,
|
| 205 |
+
"grad_norm": 1.0341521501541138,
|
| 206 |
+
"learning_rate": 4.2604056741308796e-05,
|
| 207 |
+
"loss": 0.3652,
|
| 208 |
+
"step": 117
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 13.0,
|
| 212 |
+
"eval_loss": 0.4320377707481384,
|
| 213 |
+
"eval_runtime": 7.073,
|
| 214 |
+
"eval_samples_per_second": 1.414,
|
| 215 |
+
"eval_steps_per_second": 0.283,
|
| 216 |
+
"eval_wer": 0.19494584837545126,
|
| 217 |
+
"step": 117
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 14.0,
|
| 221 |
+
"grad_norm": 0.6511925458908081,
|
| 222 |
+
"learning_rate": 4.12283000519888e-05,
|
| 223 |
+
"loss": 0.3597,
|
| 224 |
+
"step": 126
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.0,
|
| 228 |
+
"eval_loss": 0.4305523931980133,
|
| 229 |
+
"eval_runtime": 7.0901,
|
| 230 |
+
"eval_samples_per_second": 1.41,
|
| 231 |
+
"eval_steps_per_second": 0.282,
|
| 232 |
+
"eval_wer": 0.19133574007220217,
|
| 233 |
+
"step": 126
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.0,
|
| 237 |
+
"grad_norm": 0.897441565990448,
|
| 238 |
+
"learning_rate": 3.9761788125968744e-05,
|
| 239 |
+
"loss": 0.3701,
|
| 240 |
+
"step": 135
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.0,
|
| 244 |
+
"eval_loss": 0.4407591223716736,
|
| 245 |
+
"eval_runtime": 6.9823,
|
| 246 |
+
"eval_samples_per_second": 1.432,
|
| 247 |
+
"eval_steps_per_second": 0.286,
|
| 248 |
+
"eval_wer": 0.1891696750902527,
|
| 249 |
+
"step": 135
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 1.0737636089324951,
|
| 254 |
+
"learning_rate": 3.821272229281139e-05,
|
| 255 |
+
"loss": 0.3348,
|
| 256 |
+
"step": 144
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_loss": 0.4343787729740143,
|
| 261 |
+
"eval_runtime": 7.07,
|
| 262 |
+
"eval_samples_per_second": 1.414,
|
| 263 |
+
"eval_steps_per_second": 0.283,
|
| 264 |
+
"eval_wer": 0.18483754512635378,
|
| 265 |
+
"step": 144
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 17.0,
|
| 269 |
+
"grad_norm": 0.5325424671173096,
|
| 270 |
+
"learning_rate": 3.6589765557015145e-05,
|
| 271 |
+
"loss": 0.3248,
|
| 272 |
+
"step": 153
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 17.0,
|
| 276 |
+
"eval_loss": 0.42944344878196716,
|
| 277 |
+
"eval_runtime": 7.0553,
|
| 278 |
+
"eval_samples_per_second": 1.417,
|
| 279 |
+
"eval_steps_per_second": 0.283,
|
| 280 |
+
"eval_wer": 0.1884476534296029,
|
| 281 |
+
"step": 153
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 18.0,
|
| 285 |
+
"grad_norm": 0.6885235905647278,
|
| 286 |
+
"learning_rate": 3.490199415097892e-05,
|
| 287 |
+
"loss": 0.3212,
|
| 288 |
+
"step": 162
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 18.0,
|
| 292 |
+
"eval_loss": 0.4301183819770813,
|
| 293 |
+
"eval_runtime": 6.9613,
|
| 294 |
+
"eval_samples_per_second": 1.437,
|
| 295 |
+
"eval_steps_per_second": 0.287,
|
| 296 |
+
"eval_wer": 0.18483754512635378,
|
| 297 |
+
"step": 162
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 19.0,
|
| 301 |
+
"grad_norm": 0.5315946340560913,
|
| 302 |
+
"learning_rate": 3.31588467770289e-05,
|
| 303 |
+
"loss": 0.3484,
|
| 304 |
+
"step": 171
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 19.0,
|
| 308 |
+
"eval_loss": 0.4417332112789154,
|
| 309 |
+
"eval_runtime": 7.0935,
|
| 310 |
+
"eval_samples_per_second": 1.41,
|
| 311 |
+
"eval_steps_per_second": 0.282,
|
| 312 |
+
"eval_wer": 0.1855595667870036,
|
| 313 |
+
"step": 171
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 0.6952418684959412,
|
| 318 |
+
"learning_rate": 3.137007182236637e-05,
|
| 319 |
+
"loss": 0.3193,
|
| 320 |
+
"step": 180
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_loss": 0.4366823732852936,
|
| 325 |
+
"eval_runtime": 6.9514,
|
| 326 |
+
"eval_samples_per_second": 1.439,
|
| 327 |
+
"eval_steps_per_second": 0.288,
|
| 328 |
+
"eval_wer": 0.19061371841155234,
|
| 329 |
+
"step": 180
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.0,
|
| 333 |
+
"grad_norm": 0.3309876024723053,
|
| 334 |
+
"learning_rate": 2.9545672842132273e-05,
|
| 335 |
+
"loss": 0.3157,
|
| 336 |
+
"step": 189
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.0,
|
| 340 |
+
"eval_loss": 0.45008668303489685,
|
| 341 |
+
"eval_runtime": 6.9279,
|
| 342 |
+
"eval_samples_per_second": 1.443,
|
| 343 |
+
"eval_steps_per_second": 0.289,
|
| 344 |
+
"eval_wer": 0.18411552346570398,
|
| 345 |
+
"step": 189
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.0,
|
| 349 |
+
"grad_norm": 0.49725911021232605,
|
| 350 |
+
"learning_rate": 2.769585261546897e-05,
|
| 351 |
+
"loss": 0.2985,
|
| 352 |
+
"step": 198
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 22.0,
|
| 356 |
+
"eval_loss": 0.4500831961631775,
|
| 357 |
+
"eval_runtime": 6.8409,
|
| 358 |
+
"eval_samples_per_second": 1.462,
|
| 359 |
+
"eval_steps_per_second": 0.292,
|
| 360 |
+
"eval_wer": 0.18050541516245489,
|
| 361 |
+
"step": 198
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 23.0,
|
| 365 |
+
"grad_norm": 0.7141502499580383,
|
| 366 |
+
"learning_rate": 2.5830956087440665e-05,
|
| 367 |
+
"loss": 0.3284,
|
| 368 |
+
"step": 207
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 23.0,
|
| 372 |
+
"eval_loss": 0.4402785897254944,
|
| 373 |
+
"eval_runtime": 6.9521,
|
| 374 |
+
"eval_samples_per_second": 1.438,
|
| 375 |
+
"eval_steps_per_second": 0.288,
|
| 376 |
+
"eval_wer": 0.18483754512635378,
|
| 377 |
+
"step": 207
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 0.8570722341537476,
|
| 382 |
+
"learning_rate": 2.3961412515904336e-05,
|
| 383 |
+
"loss": 0.3057,
|
| 384 |
+
"step": 216
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_loss": 0.4428979754447937,
|
| 389 |
+
"eval_runtime": 6.9014,
|
| 390 |
+
"eval_samples_per_second": 1.449,
|
| 391 |
+
"eval_steps_per_second": 0.29,
|
| 392 |
+
"eval_wer": 0.1884476534296029,
|
| 393 |
+
"step": 216
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 25.0,
|
| 397 |
+
"grad_norm": 0.9306482076644897,
|
| 398 |
+
"learning_rate": 2.2097677146869242e-05,
|
| 399 |
+
"loss": 0.3062,
|
| 400 |
+
"step": 225
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 25.0,
|
| 404 |
+
"eval_loss": 0.4460487961769104,
|
| 405 |
+
"eval_runtime": 6.8059,
|
| 406 |
+
"eval_samples_per_second": 1.469,
|
| 407 |
+
"eval_steps_per_second": 0.294,
|
| 408 |
+
"eval_wer": 0.1855595667870036,
|
| 409 |
+
"step": 225
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 26.0,
|
| 413 |
+
"grad_norm": 0.28782331943511963,
|
| 414 |
+
"learning_rate": 2.0250172744520258e-05,
|
| 415 |
+
"loss": 0.2985,
|
| 416 |
+
"step": 234
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 26.0,
|
| 420 |
+
"eval_loss": 0.4484630227088928,
|
| 421 |
+
"eval_runtime": 6.8287,
|
| 422 |
+
"eval_samples_per_second": 1.464,
|
| 423 |
+
"eval_steps_per_second": 0.293,
|
| 424 |
+
"eval_wer": 0.18411552346570398,
|
| 425 |
+
"step": 234
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 27.0,
|
| 429 |
+
"grad_norm": 0.636001706123352,
|
| 430 |
+
"learning_rate": 1.842923130289262e-05,
|
| 431 |
+
"loss": 0.2732,
|
| 432 |
+
"step": 243
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.0,
|
| 436 |
+
"eval_loss": 0.449407160282135,
|
| 437 |
+
"eval_runtime": 6.9231,
|
| 438 |
+
"eval_samples_per_second": 1.444,
|
| 439 |
+
"eval_steps_per_second": 0.289,
|
| 440 |
+
"eval_wer": 0.18050541516245489,
|
| 441 |
+
"step": 243
|
| 442 |
+
}
|
| 443 |
+
],
|
| 444 |
+
"logging_steps": 500,
|
| 445 |
+
"max_steps": 400,
|
| 446 |
+
"num_input_tokens_seen": 0,
|
| 447 |
+
"num_train_epochs": 50,
|
| 448 |
+
"save_steps": 500,
|
| 449 |
+
"stateful_callbacks": {
|
| 450 |
+
"EarlyStoppingCallback": {
|
| 451 |
+
"args": {
|
| 452 |
+
"early_stopping_patience": 6,
|
| 453 |
+
"early_stopping_threshold": 0.0
|
| 454 |
+
},
|
| 455 |
+
"attributes": {
|
| 456 |
+
"early_stopping_patience_counter": 5
|
| 457 |
+
}
|
| 458 |
+
},
|
| 459 |
+
"TrainerControl": {
|
| 460 |
+
"args": {
|
| 461 |
+
"should_epoch_stop": false,
|
| 462 |
+
"should_evaluate": false,
|
| 463 |
+
"should_log": false,
|
| 464 |
+
"should_save": true,
|
| 465 |
+
"should_training_stop": false
|
| 466 |
+
},
|
| 467 |
+
"attributes": {}
|
| 468 |
+
}
|
| 469 |
+
},
|
| 470 |
+
"total_flos": 3.7013666477539267e+18,
|
| 471 |
+
"train_batch_size": 4,
|
| 472 |
+
"trial_name": null,
|
| 473 |
+
"trial_params": null
|
| 474 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68d59ad509bc3f83b9464870789ab0edc9cb84b77d08df5cbe12b1e7eebbdf21
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"'": 5,
|
| 3 |
+
"-": 6,
|
| 4 |
+
"</s>": 2,
|
| 5 |
+
"<pad>": 0,
|
| 6 |
+
"<s>": 1,
|
| 7 |
+
"<unk>": 3,
|
| 8 |
+
"a": 7,
|
| 9 |
+
"b": 8,
|
| 10 |
+
"c": 9,
|
| 11 |
+
"d": 10,
|
| 12 |
+
"e": 11,
|
| 13 |
+
"f": 12,
|
| 14 |
+
"g": 13,
|
| 15 |
+
"h": 14,
|
| 16 |
+
"i": 15,
|
| 17 |
+
"j": 16,
|
| 18 |
+
"k": 17,
|
| 19 |
+
"l": 18,
|
| 20 |
+
"m": 19,
|
| 21 |
+
"n": 20,
|
| 22 |
+
"o": 21,
|
| 23 |
+
"p": 22,
|
| 24 |
+
"q": 23,
|
| 25 |
+
"r": 24,
|
| 26 |
+
"s": 25,
|
| 27 |
+
"t": 26,
|
| 28 |
+
"u": 27,
|
| 29 |
+
"v": 28,
|
| 30 |
+
"w": 29,
|
| 31 |
+
"x": 30,
|
| 32 |
+
"y": 31,
|
| 33 |
+
"z": 32,
|
| 34 |
+
"|": 4
|
| 35 |
+
}
|