Initial upload of fine-tuned model
Browse files- added_tokens.json +4 -0
- config.json +116 -0
- model.safetensors +3 -0
- optimizer.pt +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 +40 -0
- trainer_state.json +369 -0
- training_args.bin +3 -0
- vocab.json +42 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 41,
|
| 3 |
+
"[PAD]": 40
|
| 4 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 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": 768,
|
| 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": false,
|
| 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": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 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_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": 40,
|
| 89 |
+
"proj_codevector_dim": 768,
|
| 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": "4.46.2",
|
| 113 |
+
"use_weighted_layer_sum": false,
|
| 114 |
+
"vocab_size": 41,
|
| 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:ee55bfec7db41e7ac3285da5f500b66565407b1819658c87c928bf92d0733b9b
|
| 3 |
+
size 1261975580
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6f1a15abc9fb647c6dbccb3f57e21ef3afff406edeed77c4d5ff19bbcaa4788
|
| 3 |
+
size 2490495926
|
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": false,
|
| 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:c7bfff24010ce5addc6819311c0cd527c781debae728b3dab4834c2c00f9e1e4
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac12145411f3297fd1c8f86688bb9872f0cff0af2b85faa89a991e959ebe046c
|
| 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 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"39": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": true,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": false
|
| 10 |
+
},
|
| 11 |
+
"40": {
|
| 12 |
+
"content": "[PAD]",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": true,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": false
|
| 18 |
+
},
|
| 19 |
+
"41": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"bos_token": "<s>",
|
| 29 |
+
"clean_up_tokenization_spaces": false,
|
| 30 |
+
"do_lower_case": false,
|
| 31 |
+
"eos_token": "</s>",
|
| 32 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 33 |
+
"pad_token": "[PAD]",
|
| 34 |
+
"processor_class": "Wav2Vec2Processor",
|
| 35 |
+
"replace_word_delimiter_char": " ",
|
| 36 |
+
"target_lang": null,
|
| 37 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 38 |
+
"unk_token": "[UNK]",
|
| 39 |
+
"word_delimiter_token": "|"
|
| 40 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,369 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 10584,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.6613756613756614,
|
| 13 |
+
"grad_norm": 4.001740455627441,
|
| 14 |
+
"learning_rate": 0.00014879999999999998,
|
| 15 |
+
"loss": 5.2541,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.6613756613756614,
|
| 20 |
+
"eval_cer": 0.7718660765772963,
|
| 21 |
+
"eval_loss": 2.2404534816741943,
|
| 22 |
+
"eval_runtime": 101.1138,
|
| 23 |
+
"eval_samples_per_second": 29.897,
|
| 24 |
+
"eval_steps_per_second": 3.738,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.3227513227513228,
|
| 29 |
+
"grad_norm": 2.757023334503174,
|
| 30 |
+
"learning_rate": 0.0002988,
|
| 31 |
+
"loss": 0.9761,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.3227513227513228,
|
| 36 |
+
"eval_cer": 0.15263351698359653,
|
| 37 |
+
"eval_loss": 0.45504969358444214,
|
| 38 |
+
"eval_runtime": 101.0814,
|
| 39 |
+
"eval_samples_per_second": 29.907,
|
| 40 |
+
"eval_steps_per_second": 3.74,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.9841269841269842,
|
| 45 |
+
"grad_norm": 3.515296697616577,
|
| 46 |
+
"learning_rate": 0.000284474123539232,
|
| 47 |
+
"loss": 0.6318,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.9841269841269842,
|
| 52 |
+
"eval_cer": 0.14255243759249095,
|
| 53 |
+
"eval_loss": 0.37101900577545166,
|
| 54 |
+
"eval_runtime": 100.1575,
|
| 55 |
+
"eval_samples_per_second": 30.182,
|
| 56 |
+
"eval_steps_per_second": 3.774,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.6455026455026456,
|
| 61 |
+
"grad_norm": 3.0831806659698486,
|
| 62 |
+
"learning_rate": 0.00026882303839732884,
|
| 63 |
+
"loss": 0.4988,
|
| 64 |
+
"step": 2000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.6455026455026456,
|
| 68 |
+
"eval_cer": 0.13127635774978635,
|
| 69 |
+
"eval_loss": 0.33260029554367065,
|
| 70 |
+
"eval_runtime": 101.862,
|
| 71 |
+
"eval_samples_per_second": 29.677,
|
| 72 |
+
"eval_steps_per_second": 3.711,
|
| 73 |
+
"step": 2000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 3.306878306878307,
|
| 77 |
+
"grad_norm": 2.1343257427215576,
|
| 78 |
+
"learning_rate": 0.00025317195325542566,
|
| 79 |
+
"loss": 0.4457,
|
| 80 |
+
"step": 2500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.306878306878307,
|
| 84 |
+
"eval_cer": 0.12462048314146165,
|
| 85 |
+
"eval_loss": 0.3164725601673126,
|
| 86 |
+
"eval_runtime": 100.3857,
|
| 87 |
+
"eval_samples_per_second": 30.114,
|
| 88 |
+
"eval_steps_per_second": 3.765,
|
| 89 |
+
"step": 2500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.9682539682539684,
|
| 93 |
+
"grad_norm": 2.458252191543579,
|
| 94 |
+
"learning_rate": 0.00023752086811352254,
|
| 95 |
+
"loss": 0.4001,
|
| 96 |
+
"step": 3000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.9682539682539684,
|
| 100 |
+
"eval_cer": 0.11981963830393308,
|
| 101 |
+
"eval_loss": 0.29567044973373413,
|
| 102 |
+
"eval_runtime": 100.366,
|
| 103 |
+
"eval_samples_per_second": 30.12,
|
| 104 |
+
"eval_steps_per_second": 3.766,
|
| 105 |
+
"step": 3000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 4.62962962962963,
|
| 109 |
+
"grad_norm": 2.29879093170166,
|
| 110 |
+
"learning_rate": 0.00022186978297161937,
|
| 111 |
+
"loss": 0.3481,
|
| 112 |
+
"step": 3500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.62962962962963,
|
| 116 |
+
"eval_cer": 0.11847873663440629,
|
| 117 |
+
"eval_loss": 0.2806848883628845,
|
| 118 |
+
"eval_runtime": 100.2612,
|
| 119 |
+
"eval_samples_per_second": 30.151,
|
| 120 |
+
"eval_steps_per_second": 3.77,
|
| 121 |
+
"step": 3500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 5.291005291005291,
|
| 125 |
+
"grad_norm": 1.4758025407791138,
|
| 126 |
+
"learning_rate": 0.0002062186978297162,
|
| 127 |
+
"loss": 0.3299,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.291005291005291,
|
| 132 |
+
"eval_cer": 0.1159011484510154,
|
| 133 |
+
"eval_loss": 0.27692675590515137,
|
| 134 |
+
"eval_runtime": 100.2812,
|
| 135 |
+
"eval_samples_per_second": 30.145,
|
| 136 |
+
"eval_steps_per_second": 3.769,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.9523809523809526,
|
| 141 |
+
"grad_norm": 1.3598448038101196,
|
| 142 |
+
"learning_rate": 0.00019056761268781302,
|
| 143 |
+
"loss": 0.3024,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 5.9523809523809526,
|
| 148 |
+
"eval_cer": 0.11369178715096608,
|
| 149 |
+
"eval_loss": 0.26659879088401794,
|
| 150 |
+
"eval_runtime": 100.6471,
|
| 151 |
+
"eval_samples_per_second": 30.036,
|
| 152 |
+
"eval_steps_per_second": 3.756,
|
| 153 |
+
"step": 4500
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 6.613756613756614,
|
| 157 |
+
"grad_norm": 1.0627385377883911,
|
| 158 |
+
"learning_rate": 0.00017491652754590985,
|
| 159 |
+
"loss": 0.273,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 6.613756613756614,
|
| 164 |
+
"eval_cer": 0.11135041998707731,
|
| 165 |
+
"eval_loss": 0.2516258955001831,
|
| 166 |
+
"eval_runtime": 100.6836,
|
| 167 |
+
"eval_samples_per_second": 30.025,
|
| 168 |
+
"eval_steps_per_second": 3.754,
|
| 169 |
+
"step": 5000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 7.275132275132275,
|
| 173 |
+
"grad_norm": 2.335207223892212,
|
| 174 |
+
"learning_rate": 0.00015926544240400667,
|
| 175 |
+
"loss": 0.2573,
|
| 176 |
+
"step": 5500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 7.275132275132275,
|
| 180 |
+
"eval_cer": 0.11219108890942313,
|
| 181 |
+
"eval_loss": 0.26705658435821533,
|
| 182 |
+
"eval_runtime": 100.0417,
|
| 183 |
+
"eval_samples_per_second": 30.217,
|
| 184 |
+
"eval_steps_per_second": 3.778,
|
| 185 |
+
"step": 5500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.936507936507937,
|
| 189 |
+
"grad_norm": 2.2953412532806396,
|
| 190 |
+
"learning_rate": 0.0001436143572621035,
|
| 191 |
+
"loss": 0.2373,
|
| 192 |
+
"step": 6000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 7.936507936507937,
|
| 196 |
+
"eval_cer": 0.11012068115025742,
|
| 197 |
+
"eval_loss": 0.26839447021484375,
|
| 198 |
+
"eval_runtime": 101.1747,
|
| 199 |
+
"eval_samples_per_second": 29.879,
|
| 200 |
+
"eval_steps_per_second": 3.736,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 8.597883597883598,
|
| 205 |
+
"grad_norm": 1.1853502988815308,
|
| 206 |
+
"learning_rate": 0.00012796327212020032,
|
| 207 |
+
"loss": 0.2196,
|
| 208 |
+
"step": 6500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 8.597883597883598,
|
| 212 |
+
"eval_cer": 0.10729992427032022,
|
| 213 |
+
"eval_loss": 0.2642601728439331,
|
| 214 |
+
"eval_runtime": 100.6315,
|
| 215 |
+
"eval_samples_per_second": 30.04,
|
| 216 |
+
"eval_steps_per_second": 3.756,
|
| 217 |
+
"step": 6500
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 9.25925925925926,
|
| 221 |
+
"grad_norm": 4.087908744812012,
|
| 222 |
+
"learning_rate": 0.00011234348914858095,
|
| 223 |
+
"loss": 0.2033,
|
| 224 |
+
"step": 7000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 9.25925925925926,
|
| 228 |
+
"eval_cer": 0.10725129053101096,
|
| 229 |
+
"eval_loss": 0.27656930685043335,
|
| 230 |
+
"eval_runtime": 100.8711,
|
| 231 |
+
"eval_samples_per_second": 29.969,
|
| 232 |
+
"eval_steps_per_second": 3.747,
|
| 233 |
+
"step": 7000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 9.920634920634921,
|
| 237 |
+
"grad_norm": 2.103665351867676,
|
| 238 |
+
"learning_rate": 9.669240400667779e-05,
|
| 239 |
+
"loss": 0.1958,
|
| 240 |
+
"step": 7500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 9.920634920634921,
|
| 244 |
+
"eval_cer": 0.10560469107154023,
|
| 245 |
+
"eval_loss": 0.2580866515636444,
|
| 246 |
+
"eval_runtime": 100.9128,
|
| 247 |
+
"eval_samples_per_second": 29.957,
|
| 248 |
+
"eval_steps_per_second": 3.746,
|
| 249 |
+
"step": 7500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 10.582010582010582,
|
| 253 |
+
"grad_norm": 4.133511066436768,
|
| 254 |
+
"learning_rate": 8.104131886477462e-05,
|
| 255 |
+
"loss": 0.1758,
|
| 256 |
+
"step": 8000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 10.582010582010582,
|
| 260 |
+
"eval_cer": 0.10472233608692934,
|
| 261 |
+
"eval_loss": 0.2675287127494812,
|
| 262 |
+
"eval_runtime": 100.8159,
|
| 263 |
+
"eval_samples_per_second": 29.985,
|
| 264 |
+
"eval_steps_per_second": 3.749,
|
| 265 |
+
"step": 8000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 11.243386243386244,
|
| 269 |
+
"grad_norm": 1.4247872829437256,
|
| 270 |
+
"learning_rate": 6.539023372287144e-05,
|
| 271 |
+
"loss": 0.1671,
|
| 272 |
+
"step": 8500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 11.243386243386244,
|
| 276 |
+
"eval_cer": 0.1042429463708809,
|
| 277 |
+
"eval_loss": 0.2661576569080353,
|
| 278 |
+
"eval_runtime": 100.6921,
|
| 279 |
+
"eval_samples_per_second": 30.022,
|
| 280 |
+
"eval_steps_per_second": 3.754,
|
| 281 |
+
"step": 8500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 11.904761904761905,
|
| 285 |
+
"grad_norm": 1.5974005460739136,
|
| 286 |
+
"learning_rate": 4.9739148580968275e-05,
|
| 287 |
+
"loss": 0.1589,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 11.904761904761905,
|
| 292 |
+
"eval_cer": 0.10398588232024622,
|
| 293 |
+
"eval_loss": 0.26505181193351746,
|
| 294 |
+
"eval_runtime": 100.6474,
|
| 295 |
+
"eval_samples_per_second": 30.036,
|
| 296 |
+
"eval_steps_per_second": 3.756,
|
| 297 |
+
"step": 9000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 12.566137566137566,
|
| 301 |
+
"grad_norm": 0.8991349935531616,
|
| 302 |
+
"learning_rate": 3.408806343906511e-05,
|
| 303 |
+
"loss": 0.1433,
|
| 304 |
+
"step": 9500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 12.566137566137566,
|
| 308 |
+
"eval_cer": 0.10245044569348238,
|
| 309 |
+
"eval_loss": 0.2700433135032654,
|
| 310 |
+
"eval_runtime": 101.5429,
|
| 311 |
+
"eval_samples_per_second": 29.771,
|
| 312 |
+
"eval_steps_per_second": 3.723,
|
| 313 |
+
"step": 9500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 13.227513227513228,
|
| 317 |
+
"grad_norm": 1.4518108367919922,
|
| 318 |
+
"learning_rate": 1.8436978297161933e-05,
|
| 319 |
+
"loss": 0.1394,
|
| 320 |
+
"step": 10000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 13.227513227513228,
|
| 324 |
+
"eval_cer": 0.10264498065071943,
|
| 325 |
+
"eval_loss": 0.27115562558174133,
|
| 326 |
+
"eval_runtime": 108.9106,
|
| 327 |
+
"eval_samples_per_second": 27.757,
|
| 328 |
+
"eval_steps_per_second": 3.471,
|
| 329 |
+
"step": 10000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.88888888888889,
|
| 333 |
+
"grad_norm": 1.4964288473129272,
|
| 334 |
+
"learning_rate": 2.817195325542571e-06,
|
| 335 |
+
"loss": 0.1318,
|
| 336 |
+
"step": 10500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 13.88888888888889,
|
| 340 |
+
"eval_cer": 0.10274919580638213,
|
| 341 |
+
"eval_loss": 0.2701418697834015,
|
| 342 |
+
"eval_runtime": 100.2825,
|
| 343 |
+
"eval_samples_per_second": 30.145,
|
| 344 |
+
"eval_steps_per_second": 3.769,
|
| 345 |
+
"step": 10500
|
| 346 |
+
}
|
| 347 |
+
],
|
| 348 |
+
"logging_steps": 500,
|
| 349 |
+
"max_steps": 10584,
|
| 350 |
+
"num_input_tokens_seen": 0,
|
| 351 |
+
"num_train_epochs": 14,
|
| 352 |
+
"save_steps": 500,
|
| 353 |
+
"stateful_callbacks": {
|
| 354 |
+
"TrainerControl": {
|
| 355 |
+
"args": {
|
| 356 |
+
"should_epoch_stop": false,
|
| 357 |
+
"should_evaluate": false,
|
| 358 |
+
"should_log": false,
|
| 359 |
+
"should_save": true,
|
| 360 |
+
"should_training_stop": true
|
| 361 |
+
},
|
| 362 |
+
"attributes": {}
|
| 363 |
+
}
|
| 364 |
+
},
|
| 365 |
+
"total_flos": 2.0915180948213604e+19,
|
| 366 |
+
"train_batch_size": 8,
|
| 367 |
+
"trial_name": null,
|
| 368 |
+
"trial_params": null
|
| 369 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7aa28fd00175efe967014fe4050d5f6631387ca3983cba58afca6e40e3f57ee4
|
| 3 |
+
size 5304
|
vocab.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 40,
|
| 3 |
+
"[UNK]": 39,
|
| 4 |
+
"a": 10,
|
| 5 |
+
"b": 23,
|
| 6 |
+
"d": 12,
|
| 7 |
+
"e": 21,
|
| 8 |
+
"f": 37,
|
| 9 |
+
"h": 30,
|
| 10 |
+
"i": 22,
|
| 11 |
+
"j": 0,
|
| 12 |
+
"k": 28,
|
| 13 |
+
"l": 35,
|
| 14 |
+
"m": 1,
|
| 15 |
+
"n": 6,
|
| 16 |
+
"o": 25,
|
| 17 |
+
"p": 11,
|
| 18 |
+
"s": 14,
|
| 19 |
+
"t": 18,
|
| 20 |
+
"u": 5,
|
| 21 |
+
"v": 2,
|
| 22 |
+
"w": 27,
|
| 23 |
+
"z": 20,
|
| 24 |
+
"|": 31,
|
| 25 |
+
"æ": 29,
|
| 26 |
+
"ð": 34,
|
| 27 |
+
"ŋ": 33,
|
| 28 |
+
"ɑ": 38,
|
| 29 |
+
"ɔ": 19,
|
| 30 |
+
"ə": 13,
|
| 31 |
+
"ɚ": 15,
|
| 32 |
+
"ɛ": 17,
|
| 33 |
+
"ɡ": 8,
|
| 34 |
+
"ɪ": 24,
|
| 35 |
+
"ɹ": 36,
|
| 36 |
+
"ʃ": 7,
|
| 37 |
+
"ʊ": 4,
|
| 38 |
+
"ʌ": 9,
|
| 39 |
+
"ʒ": 26,
|
| 40 |
+
"͡": 3,
|
| 41 |
+
"θ": 16
|
| 42 |
+
}
|