Upload folder using huggingface_hub
Browse files- added_tokens.json +4 -0
- checkpoint-9380/config.json +107 -0
- checkpoint-9380/model.safetensors +3 -0
- checkpoint-9380/optimizer.pt +3 -0
- checkpoint-9380/preprocessor_config.json +9 -0
- checkpoint-9380/rng_state.pth +3 -0
- checkpoint-9380/scaler.pt +3 -0
- checkpoint-9380/scheduler.pt +3 -0
- checkpoint-9380/trainer_state.json +322 -0
- checkpoint-9380/training_args.bin +3 -0
- config.json +107 -0
- model.safetensors +3 -0
- preprocessor_config.json +10 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +49 -0
- vocab.json +31 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 30,
|
| 3 |
+
"<s>": 29
|
| 4 |
+
}
|
checkpoint-9380/config.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.0,
|
| 3 |
+
"adapter_attn_dim": null,
|
| 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 |
+
"dtype": "float32",
|
| 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_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": 28,
|
| 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 |
+
"transformers_version": "4.57.1",
|
| 104 |
+
"use_weighted_layer_sum": false,
|
| 105 |
+
"vocab_size": 31,
|
| 106 |
+
"xvector_output_dim": 512
|
| 107 |
+
}
|
checkpoint-9380/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81a39ed9b6ef1eff2fc65ad482d7d0f4ff912a15a453a4b09d001aabaa2fbdd3
|
| 3 |
+
size 1261934580
|
checkpoint-9380/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9303733744e81732c20f7e22ccc64b19a07208c98a2306f01e9213f886a542e0
|
| 3 |
+
size 2524125339
|
checkpoint-9380/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
checkpoint-9380/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:efdd4b94c1e6ad8732d109aef85c971cb156d145bb6ae2654b14f21f32464824
|
| 3 |
+
size 14709
|
checkpoint-9380/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab71dceeceb379e73894555f0e2a4b12b34de46e52e7a0f458508e20186eb0a3
|
| 3 |
+
size 1383
|
checkpoint-9380/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce187e3f555b7fd7243a8f31396ec053a90fed362680018d8b2aa39fb391bb01
|
| 3 |
+
size 1465
|
checkpoint-9380/trainer_state.json
ADDED
|
@@ -0,0 +1,322 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 9000,
|
| 3 |
+
"best_metric": 0.2908235294117647,
|
| 4 |
+
"best_model_checkpoint": "/content/drive/MyDrive/Model-Finetuning/wav2vec2/model/nbl_wav2vec2/checkpoint-9000",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 9380,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.5333333333333333,
|
| 14 |
+
"grad_norm": 2.6857900619506836,
|
| 15 |
+
"learning_rate": 0.00029939999999999996,
|
| 16 |
+
"loss": 5.2283,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.5333333333333333,
|
| 21 |
+
"eval_loss": 2.025707721710205,
|
| 22 |
+
"eval_runtime": 148.4202,
|
| 23 |
+
"eval_samples_per_second": 20.213,
|
| 24 |
+
"eval_steps_per_second": 2.527,
|
| 25 |
+
"eval_wer": 1.0111372549019608,
|
| 26 |
+
"step": 500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0661333333333334,
|
| 30 |
+
"grad_norm": 1.386874794960022,
|
| 31 |
+
"learning_rate": 0.0002831418918918919,
|
| 32 |
+
"loss": 0.9713,
|
| 33 |
+
"step": 1000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.0661333333333334,
|
| 37 |
+
"eval_loss": 0.5391530394554138,
|
| 38 |
+
"eval_runtime": 148.3623,
|
| 39 |
+
"eval_samples_per_second": 20.221,
|
| 40 |
+
"eval_steps_per_second": 2.528,
|
| 41 |
+
"eval_wer": 0.6509019607843137,
|
| 42 |
+
"step": 1000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.5994666666666668,
|
| 46 |
+
"grad_norm": 2.7049543857574463,
|
| 47 |
+
"learning_rate": 0.00026624999999999994,
|
| 48 |
+
"loss": 0.6379,
|
| 49 |
+
"step": 1500
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.5994666666666668,
|
| 53 |
+
"eval_loss": 0.42855554819107056,
|
| 54 |
+
"eval_runtime": 148.347,
|
| 55 |
+
"eval_samples_per_second": 20.223,
|
| 56 |
+
"eval_steps_per_second": 2.528,
|
| 57 |
+
"eval_wer": 0.5332549019607843,
|
| 58 |
+
"step": 1500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 2.1322666666666668,
|
| 62 |
+
"grad_norm": 1.8714250326156616,
|
| 63 |
+
"learning_rate": 0.00024935810810810806,
|
| 64 |
+
"loss": 0.5595,
|
| 65 |
+
"step": 2000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.1322666666666668,
|
| 69 |
+
"eval_loss": 0.39198794960975647,
|
| 70 |
+
"eval_runtime": 149.3713,
|
| 71 |
+
"eval_samples_per_second": 20.084,
|
| 72 |
+
"eval_steps_per_second": 2.511,
|
| 73 |
+
"eval_wer": 0.5109803921568628,
|
| 74 |
+
"step": 2000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.6656,
|
| 78 |
+
"grad_norm": 1.8362187147140503,
|
| 79 |
+
"learning_rate": 0.0002324662162162162,
|
| 80 |
+
"loss": 0.4807,
|
| 81 |
+
"step": 2500
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.6656,
|
| 85 |
+
"eval_loss": 0.3563985228538513,
|
| 86 |
+
"eval_runtime": 149.1684,
|
| 87 |
+
"eval_samples_per_second": 20.112,
|
| 88 |
+
"eval_steps_per_second": 2.514,
|
| 89 |
+
"eval_wer": 0.46658823529411764,
|
| 90 |
+
"step": 2500
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 3.1984,
|
| 94 |
+
"grad_norm": 1.0880553722381592,
|
| 95 |
+
"learning_rate": 0.00021557432432432432,
|
| 96 |
+
"loss": 0.4404,
|
| 97 |
+
"step": 3000
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 3.1984,
|
| 101 |
+
"eval_loss": 0.3388464152812958,
|
| 102 |
+
"eval_runtime": 149.8437,
|
| 103 |
+
"eval_samples_per_second": 20.021,
|
| 104 |
+
"eval_steps_per_second": 2.503,
|
| 105 |
+
"eval_wer": 0.42486274509803923,
|
| 106 |
+
"step": 3000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 3.7317333333333336,
|
| 110 |
+
"grad_norm": 10.404922485351562,
|
| 111 |
+
"learning_rate": 0.00019868243243243244,
|
| 112 |
+
"loss": 0.4022,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.7317333333333336,
|
| 117 |
+
"eval_loss": 0.3252636194229126,
|
| 118 |
+
"eval_runtime": 150.8413,
|
| 119 |
+
"eval_samples_per_second": 19.888,
|
| 120 |
+
"eval_steps_per_second": 2.486,
|
| 121 |
+
"eval_wer": 0.4043921568627451,
|
| 122 |
+
"step": 3500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.2645333333333335,
|
| 126 |
+
"grad_norm": 4.443390846252441,
|
| 127 |
+
"learning_rate": 0.0001817905405405405,
|
| 128 |
+
"loss": 0.3754,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4.2645333333333335,
|
| 133 |
+
"eval_loss": 0.3209179937839508,
|
| 134 |
+
"eval_runtime": 150.4901,
|
| 135 |
+
"eval_samples_per_second": 19.935,
|
| 136 |
+
"eval_steps_per_second": 2.492,
|
| 137 |
+
"eval_wer": 0.4051764705882353,
|
| 138 |
+
"step": 4000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 4.797866666666667,
|
| 142 |
+
"grad_norm": 1.9804564714431763,
|
| 143 |
+
"learning_rate": 0.00016489864864864862,
|
| 144 |
+
"loss": 0.3379,
|
| 145 |
+
"step": 4500
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 4.797866666666667,
|
| 149 |
+
"eval_loss": 0.30479735136032104,
|
| 150 |
+
"eval_runtime": 149.6398,
|
| 151 |
+
"eval_samples_per_second": 20.048,
|
| 152 |
+
"eval_steps_per_second": 2.506,
|
| 153 |
+
"eval_wer": 0.3894901960784314,
|
| 154 |
+
"step": 4500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.330666666666667,
|
| 158 |
+
"grad_norm": 1.1342227458953857,
|
| 159 |
+
"learning_rate": 0.00014800675675675674,
|
| 160 |
+
"loss": 0.3184,
|
| 161 |
+
"step": 5000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.330666666666667,
|
| 165 |
+
"eval_loss": 0.28083688020706177,
|
| 166 |
+
"eval_runtime": 150.4194,
|
| 167 |
+
"eval_samples_per_second": 19.944,
|
| 168 |
+
"eval_steps_per_second": 2.493,
|
| 169 |
+
"eval_wer": 0.3618823529411765,
|
| 170 |
+
"step": 5000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 5.864,
|
| 174 |
+
"grad_norm": 0.6732638478279114,
|
| 175 |
+
"learning_rate": 0.00013111486486486486,
|
| 176 |
+
"loss": 0.3003,
|
| 177 |
+
"step": 5500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 5.864,
|
| 181 |
+
"eval_loss": 0.27100080251693726,
|
| 182 |
+
"eval_runtime": 150.5926,
|
| 183 |
+
"eval_samples_per_second": 19.921,
|
| 184 |
+
"eval_steps_per_second": 2.49,
|
| 185 |
+
"eval_wer": 0.3436078431372549,
|
| 186 |
+
"step": 5500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 6.3968,
|
| 190 |
+
"grad_norm": 4.13991117477417,
|
| 191 |
+
"learning_rate": 0.00011422297297297297,
|
| 192 |
+
"loss": 0.264,
|
| 193 |
+
"step": 6000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 6.3968,
|
| 197 |
+
"eval_loss": 0.264404833316803,
|
| 198 |
+
"eval_runtime": 151.1359,
|
| 199 |
+
"eval_samples_per_second": 19.85,
|
| 200 |
+
"eval_steps_per_second": 2.481,
|
| 201 |
+
"eval_wer": 0.33356862745098037,
|
| 202 |
+
"step": 6000
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 6.930133333333333,
|
| 206 |
+
"grad_norm": 3.0504963397979736,
|
| 207 |
+
"learning_rate": 9.733108108108107e-05,
|
| 208 |
+
"loss": 0.2574,
|
| 209 |
+
"step": 6500
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 6.930133333333333,
|
| 213 |
+
"eval_loss": 0.24905848503112793,
|
| 214 |
+
"eval_runtime": 151.2904,
|
| 215 |
+
"eval_samples_per_second": 19.829,
|
| 216 |
+
"eval_steps_per_second": 2.479,
|
| 217 |
+
"eval_wer": 0.3235294117647059,
|
| 218 |
+
"step": 6500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.462933333333333,
|
| 222 |
+
"grad_norm": 1.2022693157196045,
|
| 223 |
+
"learning_rate": 8.043918918918918e-05,
|
| 224 |
+
"loss": 0.2319,
|
| 225 |
+
"step": 7000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 7.462933333333333,
|
| 229 |
+
"eval_loss": 0.26013123989105225,
|
| 230 |
+
"eval_runtime": 150.0311,
|
| 231 |
+
"eval_samples_per_second": 19.996,
|
| 232 |
+
"eval_steps_per_second": 2.499,
|
| 233 |
+
"eval_wer": 0.31301960784313726,
|
| 234 |
+
"step": 7000
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 7.996266666666667,
|
| 238 |
+
"grad_norm": 0.9905935525894165,
|
| 239 |
+
"learning_rate": 6.354729729729729e-05,
|
| 240 |
+
"loss": 0.2194,
|
| 241 |
+
"step": 7500
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 7.996266666666667,
|
| 245 |
+
"eval_loss": 0.24855607748031616,
|
| 246 |
+
"eval_runtime": 151.4319,
|
| 247 |
+
"eval_samples_per_second": 19.811,
|
| 248 |
+
"eval_steps_per_second": 2.476,
|
| 249 |
+
"eval_wer": 0.3060392156862745,
|
| 250 |
+
"step": 7500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 8.529066666666667,
|
| 254 |
+
"grad_norm": 1.2119730710983276,
|
| 255 |
+
"learning_rate": 4.66554054054054e-05,
|
| 256 |
+
"loss": 0.1958,
|
| 257 |
+
"step": 8000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 8.529066666666667,
|
| 261 |
+
"eval_loss": 0.24899743497371674,
|
| 262 |
+
"eval_runtime": 151.042,
|
| 263 |
+
"eval_samples_per_second": 19.862,
|
| 264 |
+
"eval_steps_per_second": 2.483,
|
| 265 |
+
"eval_wer": 0.30227450980392156,
|
| 266 |
+
"step": 8000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.061866666666667,
|
| 270 |
+
"grad_norm": 0.6335813999176025,
|
| 271 |
+
"learning_rate": 2.976351351351351e-05,
|
| 272 |
+
"loss": 0.1912,
|
| 273 |
+
"step": 8500
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 9.061866666666667,
|
| 277 |
+
"eval_loss": 0.23850028216838837,
|
| 278 |
+
"eval_runtime": 152.7072,
|
| 279 |
+
"eval_samples_per_second": 19.645,
|
| 280 |
+
"eval_steps_per_second": 2.456,
|
| 281 |
+
"eval_wer": 0.2931764705882353,
|
| 282 |
+
"step": 8500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 9.5952,
|
| 286 |
+
"grad_norm": 2.9868931770324707,
|
| 287 |
+
"learning_rate": 1.287162162162162e-05,
|
| 288 |
+
"loss": 0.1749,
|
| 289 |
+
"step": 9000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 9.5952,
|
| 293 |
+
"eval_loss": 0.2415107637643814,
|
| 294 |
+
"eval_runtime": 150.7897,
|
| 295 |
+
"eval_samples_per_second": 19.895,
|
| 296 |
+
"eval_steps_per_second": 2.487,
|
| 297 |
+
"eval_wer": 0.2908235294117647,
|
| 298 |
+
"step": 9000
|
| 299 |
+
}
|
| 300 |
+
],
|
| 301 |
+
"logging_steps": 500,
|
| 302 |
+
"max_steps": 9380,
|
| 303 |
+
"num_input_tokens_seen": 0,
|
| 304 |
+
"num_train_epochs": 10,
|
| 305 |
+
"save_steps": 500,
|
| 306 |
+
"stateful_callbacks": {
|
| 307 |
+
"TrainerControl": {
|
| 308 |
+
"args": {
|
| 309 |
+
"should_epoch_stop": false,
|
| 310 |
+
"should_evaluate": false,
|
| 311 |
+
"should_log": false,
|
| 312 |
+
"should_save": true,
|
| 313 |
+
"should_training_stop": true
|
| 314 |
+
},
|
| 315 |
+
"attributes": {}
|
| 316 |
+
}
|
| 317 |
+
},
|
| 318 |
+
"total_flos": 2.270554818805736e+19,
|
| 319 |
+
"train_batch_size": 8,
|
| 320 |
+
"trial_name": null,
|
| 321 |
+
"trial_params": null
|
| 322 |
+
}
|
checkpoint-9380/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b91ded7996d03b56dbf17d6ce4ee531269fb027966ed5de989ededffd5815f0b
|
| 3 |
+
size 5905
|
config.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.0,
|
| 3 |
+
"adapter_attn_dim": null,
|
| 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 |
+
"dtype": "float32",
|
| 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_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": 28,
|
| 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 |
+
"transformers_version": "4.57.1",
|
| 104 |
+
"use_weighted_layer_sum": false,
|
| 105 |
+
"vocab_size": 31,
|
| 106 |
+
"xvector_output_dim": 512
|
| 107 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:442e4ab7ae7b03e53ee3782d674b347fd8d036f2bc22a61c404784e59e1eeeea
|
| 3 |
+
size 1261934580
|
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 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"27": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": true,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": false
|
| 10 |
+
},
|
| 11 |
+
"28": {
|
| 12 |
+
"content": "[PAD]",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": true,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": false
|
| 18 |
+
},
|
| 19 |
+
"29": {
|
| 20 |
+
"content": "<s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"30": {
|
| 28 |
+
"content": "</s>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 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 |
+
}
|
vocab.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 28,
|
| 3 |
+
"[UNK]": 27,
|
| 4 |
+
"a": 1,
|
| 5 |
+
"b": 2,
|
| 6 |
+
"c": 3,
|
| 7 |
+
"d": 4,
|
| 8 |
+
"e": 5,
|
| 9 |
+
"f": 6,
|
| 10 |
+
"g": 7,
|
| 11 |
+
"h": 8,
|
| 12 |
+
"i": 9,
|
| 13 |
+
"j": 10,
|
| 14 |
+
"k": 11,
|
| 15 |
+
"l": 12,
|
| 16 |
+
"m": 13,
|
| 17 |
+
"n": 14,
|
| 18 |
+
"o": 15,
|
| 19 |
+
"p": 16,
|
| 20 |
+
"q": 17,
|
| 21 |
+
"r": 18,
|
| 22 |
+
"s": 19,
|
| 23 |
+
"t": 20,
|
| 24 |
+
"u": 21,
|
| 25 |
+
"v": 22,
|
| 26 |
+
"w": 23,
|
| 27 |
+
"x": 24,
|
| 28 |
+
"y": 25,
|
| 29 |
+
"z": 26,
|
| 30 |
+
"|": 0
|
| 31 |
+
}
|