Sourabh Gupta
commited on
Commit
·
ec9768a
1
Parent(s):
aef86cb
first commit
Browse files- config.json +86 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +496 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-base",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"apply_spec_augment": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Wav2Vec2ForCTC"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"codevector_dim": 256,
|
| 11 |
+
"contrastive_logits_temperature": 0.1,
|
| 12 |
+
"conv_bias": false,
|
| 13 |
+
"conv_dim": [
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512
|
| 21 |
+
],
|
| 22 |
+
"conv_kernel": [
|
| 23 |
+
10,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
3,
|
| 28 |
+
2,
|
| 29 |
+
2
|
| 30 |
+
],
|
| 31 |
+
"conv_stride": [
|
| 32 |
+
5,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2,
|
| 38 |
+
2
|
| 39 |
+
],
|
| 40 |
+
"ctc_loss_reduction": "mean",
|
| 41 |
+
"ctc_zero_infinity": false,
|
| 42 |
+
"diversity_loss_weight": 0.1,
|
| 43 |
+
"do_stable_layer_norm": false,
|
| 44 |
+
"eos_token_id": 2,
|
| 45 |
+
"feat_extract_activation": "gelu",
|
| 46 |
+
"feat_extract_norm": "group",
|
| 47 |
+
"feat_proj_dropout": 0.1,
|
| 48 |
+
"feat_quantizer_dropout": 0.0,
|
| 49 |
+
"final_dropout": 0.0,
|
| 50 |
+
"freeze_feat_extract_train": true,
|
| 51 |
+
"gradient_checkpointing": true,
|
| 52 |
+
"hidden_act": "gelu",
|
| 53 |
+
"hidden_dropout": 0.1,
|
| 54 |
+
"hidden_size": 768,
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"layer_norm_eps": 1e-05,
|
| 58 |
+
"layerdrop": 0.05,
|
| 59 |
+
"mask_channel_length": 10,
|
| 60 |
+
"mask_channel_min_space": 1,
|
| 61 |
+
"mask_channel_other": 0.0,
|
| 62 |
+
"mask_channel_prob": 0.0,
|
| 63 |
+
"mask_channel_selection": "static",
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_prob": 0.0,
|
| 66 |
+
"mask_time_length": 10,
|
| 67 |
+
"mask_time_min_space": 1,
|
| 68 |
+
"mask_time_other": 0.0,
|
| 69 |
+
"mask_time_prob": 0.05,
|
| 70 |
+
"mask_time_selection": "static",
|
| 71 |
+
"model_type": "wav2vec2",
|
| 72 |
+
"no_mask_channel_overlap": false,
|
| 73 |
+
"no_mask_time_overlap": false,
|
| 74 |
+
"num_attention_heads": 12,
|
| 75 |
+
"num_codevector_groups": 2,
|
| 76 |
+
"num_codevectors_per_group": 320,
|
| 77 |
+
"num_conv_pos_embedding_groups": 16,
|
| 78 |
+
"num_conv_pos_embeddings": 128,
|
| 79 |
+
"num_feat_extract_layers": 7,
|
| 80 |
+
"num_hidden_layers": 12,
|
| 81 |
+
"num_negatives": 100,
|
| 82 |
+
"pad_token_id": 29,
|
| 83 |
+
"proj_codevector_dim": 256,
|
| 84 |
+
"transformers_version": "4.8.2",
|
| 85 |
+
"vocab_size": 32
|
| 86 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98fde8e74d7a8b2b7dc5f15703537251d5609c0ac4e0e4dee376a2d8961afde4
|
| 3 |
+
size 721685265
|
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": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3386ee4cecd97af2b1d2a197bcc8c4609f8a2acc4839229264bce3e58e4e948a
|
| 3 |
+
size 377670039
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:431d1733cb5e5c7b76a527aac0a7a9cd041d1b0b07239c1ccc6cf97c77a085aa
|
| 3 |
+
size 17563
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:855e90b86493eac22e9016356f64ba6930858b1ed321024ff004ddf9cce9c4a5
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2dfd2e36c120141d95b66394d25c45930f4c27eb388087627bc3ec813348ab7a
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,496 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.906542056074766,
|
| 5 |
+
"global_step": 3200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.93,
|
| 12 |
+
"learning_rate": 9.7e-06,
|
| 13 |
+
"loss": 9.9876,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.93,
|
| 18 |
+
"eval_loss": 4.772322177886963,
|
| 19 |
+
"eval_runtime": 43.7501,
|
| 20 |
+
"eval_samples_per_second": 19.497,
|
| 21 |
+
"eval_steps_per_second": 0.617,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.87,
|
| 27 |
+
"learning_rate": 1.97e-05,
|
| 28 |
+
"loss": 3.6526,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.87,
|
| 33 |
+
"eval_loss": 3.8356621265411377,
|
| 34 |
+
"eval_runtime": 43.2583,
|
| 35 |
+
"eval_samples_per_second": 19.719,
|
| 36 |
+
"eval_steps_per_second": 0.624,
|
| 37 |
+
"eval_wer": 1.0,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.8,
|
| 42 |
+
"learning_rate": 2.96e-05,
|
| 43 |
+
"loss": 3.1832,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.8,
|
| 48 |
+
"eval_loss": 3.42309832572937,
|
| 49 |
+
"eval_runtime": 43.8662,
|
| 50 |
+
"eval_samples_per_second": 19.446,
|
| 51 |
+
"eval_steps_per_second": 0.616,
|
| 52 |
+
"eval_wer": 1.0,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.74,
|
| 57 |
+
"learning_rate": 3.960000000000001e-05,
|
| 58 |
+
"loss": 3.0394,
|
| 59 |
+
"step": 400
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.74,
|
| 63 |
+
"eval_loss": 3.35982608795166,
|
| 64 |
+
"eval_runtime": 43.5204,
|
| 65 |
+
"eval_samples_per_second": 19.6,
|
| 66 |
+
"eval_steps_per_second": 0.62,
|
| 67 |
+
"eval_wer": 1.0,
|
| 68 |
+
"step": 400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.67,
|
| 72 |
+
"learning_rate": 4.96e-05,
|
| 73 |
+
"loss": 2.9964,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.67,
|
| 78 |
+
"eval_loss": 3.0924816131591797,
|
| 79 |
+
"eval_runtime": 42.6866,
|
| 80 |
+
"eval_samples_per_second": 19.983,
|
| 81 |
+
"eval_steps_per_second": 0.633,
|
| 82 |
+
"eval_wer": 1.0,
|
| 83 |
+
"step": 500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 5.61,
|
| 87 |
+
"learning_rate": 5.96e-05,
|
| 88 |
+
"loss": 2.9881,
|
| 89 |
+
"step": 600
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.61,
|
| 93 |
+
"eval_loss": 3.0199828147888184,
|
| 94 |
+
"eval_runtime": 44.2234,
|
| 95 |
+
"eval_samples_per_second": 19.288,
|
| 96 |
+
"eval_steps_per_second": 0.611,
|
| 97 |
+
"eval_wer": 1.0,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.54,
|
| 102 |
+
"learning_rate": 6.96e-05,
|
| 103 |
+
"loss": 2.9678,
|
| 104 |
+
"step": 700
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.54,
|
| 108 |
+
"eval_loss": 3.0672409534454346,
|
| 109 |
+
"eval_runtime": 44.282,
|
| 110 |
+
"eval_samples_per_second": 19.263,
|
| 111 |
+
"eval_steps_per_second": 0.61,
|
| 112 |
+
"eval_wer": 1.0,
|
| 113 |
+
"step": 700
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 7.48,
|
| 117 |
+
"learning_rate": 7.960000000000001e-05,
|
| 118 |
+
"loss": 2.7541,
|
| 119 |
+
"step": 800
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 7.48,
|
| 123 |
+
"eval_loss": 2.397754669189453,
|
| 124 |
+
"eval_runtime": 43.7165,
|
| 125 |
+
"eval_samples_per_second": 19.512,
|
| 126 |
+
"eval_steps_per_second": 0.618,
|
| 127 |
+
"eval_wer": 0.9788053949903661,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.41,
|
| 132 |
+
"learning_rate": 8.960000000000001e-05,
|
| 133 |
+
"loss": 1.876,
|
| 134 |
+
"step": 900
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 8.41,
|
| 138 |
+
"eval_loss": 1.5307629108428955,
|
| 139 |
+
"eval_runtime": 43.4064,
|
| 140 |
+
"eval_samples_per_second": 19.651,
|
| 141 |
+
"eval_steps_per_second": 0.622,
|
| 142 |
+
"eval_wer": 0.7509098694069792,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 9.35,
|
| 147 |
+
"learning_rate": 9.960000000000001e-05,
|
| 148 |
+
"loss": 1.4334,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 9.35,
|
| 153 |
+
"eval_loss": 1.435613751411438,
|
| 154 |
+
"eval_runtime": 43.9507,
|
| 155 |
+
"eval_samples_per_second": 19.408,
|
| 156 |
+
"eval_steps_per_second": 0.614,
|
| 157 |
+
"eval_wer": 0.620316848640548,
|
| 158 |
+
"step": 1000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 10.28,
|
| 162 |
+
"learning_rate": 9.565610859728508e-05,
|
| 163 |
+
"loss": 1.223,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 10.28,
|
| 168 |
+
"eval_loss": 1.0539987087249756,
|
| 169 |
+
"eval_runtime": 43.8182,
|
| 170 |
+
"eval_samples_per_second": 19.467,
|
| 171 |
+
"eval_steps_per_second": 0.616,
|
| 172 |
+
"eval_wer": 0.5885249411260972,
|
| 173 |
+
"step": 1100
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 11.21,
|
| 177 |
+
"learning_rate": 9.113122171945702e-05,
|
| 178 |
+
"loss": 1.0139,
|
| 179 |
+
"step": 1200
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 11.21,
|
| 183 |
+
"eval_loss": 0.9824701547622681,
|
| 184 |
+
"eval_runtime": 43.7265,
|
| 185 |
+
"eval_samples_per_second": 19.508,
|
| 186 |
+
"eval_steps_per_second": 0.617,
|
| 187 |
+
"eval_wer": 0.4964675658317277,
|
| 188 |
+
"step": 1200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.15,
|
| 192 |
+
"learning_rate": 8.660633484162897e-05,
|
| 193 |
+
"loss": 0.9349,
|
| 194 |
+
"step": 1300
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.15,
|
| 198 |
+
"eval_loss": 0.981073796749115,
|
| 199 |
+
"eval_runtime": 43.357,
|
| 200 |
+
"eval_samples_per_second": 19.674,
|
| 201 |
+
"eval_steps_per_second": 0.623,
|
| 202 |
+
"eval_wer": 0.4774138300149861,
|
| 203 |
+
"step": 1300
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 13.08,
|
| 207 |
+
"learning_rate": 8.20814479638009e-05,
|
| 208 |
+
"loss": 0.8395,
|
| 209 |
+
"step": 1400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 13.08,
|
| 213 |
+
"eval_loss": 0.9597522616386414,
|
| 214 |
+
"eval_runtime": 44.326,
|
| 215 |
+
"eval_samples_per_second": 19.244,
|
| 216 |
+
"eval_steps_per_second": 0.609,
|
| 217 |
+
"eval_wer": 0.44166131449368445,
|
| 218 |
+
"step": 1400
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 14.02,
|
| 222 |
+
"learning_rate": 7.755656108597285e-05,
|
| 223 |
+
"loss": 0.8417,
|
| 224 |
+
"step": 1500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.02,
|
| 228 |
+
"eval_loss": 0.9240782856941223,
|
| 229 |
+
"eval_runtime": 44.9894,
|
| 230 |
+
"eval_samples_per_second": 18.96,
|
| 231 |
+
"eval_steps_per_second": 0.6,
|
| 232 |
+
"eval_wer": 0.43513166345536286,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.95,
|
| 237 |
+
"learning_rate": 7.30316742081448e-05,
|
| 238 |
+
"loss": 0.7091,
|
| 239 |
+
"step": 1600
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 14.95,
|
| 243 |
+
"eval_loss": 0.9342209100723267,
|
| 244 |
+
"eval_runtime": 43.4211,
|
| 245 |
+
"eval_samples_per_second": 19.645,
|
| 246 |
+
"eval_steps_per_second": 0.622,
|
| 247 |
+
"eval_wer": 0.4319203596660244,
|
| 248 |
+
"step": 1600
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.89,
|
| 252 |
+
"learning_rate": 6.850678733031674e-05,
|
| 253 |
+
"loss": 0.6622,
|
| 254 |
+
"step": 1700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.89,
|
| 258 |
+
"eval_loss": 0.895724892616272,
|
| 259 |
+
"eval_runtime": 43.4542,
|
| 260 |
+
"eval_samples_per_second": 19.63,
|
| 261 |
+
"eval_steps_per_second": 0.621,
|
| 262 |
+
"eval_wer": 0.41629201455791054,
|
| 263 |
+
"step": 1700
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 16.82,
|
| 267 |
+
"learning_rate": 6.398190045248869e-05,
|
| 268 |
+
"loss": 0.6762,
|
| 269 |
+
"step": 1800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 16.82,
|
| 273 |
+
"eval_loss": 0.9993765354156494,
|
| 274 |
+
"eval_runtime": 44.1607,
|
| 275 |
+
"eval_samples_per_second": 19.316,
|
| 276 |
+
"eval_steps_per_second": 0.611,
|
| 277 |
+
"eval_wer": 0.40023549561121813,
|
| 278 |
+
"step": 1800
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 17.76,
|
| 282 |
+
"learning_rate": 5.945701357466064e-05,
|
| 283 |
+
"loss": 0.6325,
|
| 284 |
+
"step": 1900
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 17.76,
|
| 288 |
+
"eval_loss": 0.9341434240341187,
|
| 289 |
+
"eval_runtime": 42.9623,
|
| 290 |
+
"eval_samples_per_second": 19.855,
|
| 291 |
+
"eval_steps_per_second": 0.628,
|
| 292 |
+
"eval_wer": 0.41168914579319205,
|
| 293 |
+
"step": 1900
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 18.69,
|
| 297 |
+
"learning_rate": 5.4932126696832586e-05,
|
| 298 |
+
"loss": 0.5829,
|
| 299 |
+
"step": 2000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 18.69,
|
| 303 |
+
"eval_loss": 1.0207712650299072,
|
| 304 |
+
"eval_runtime": 42.8374,
|
| 305 |
+
"eval_samples_per_second": 19.913,
|
| 306 |
+
"eval_steps_per_second": 0.63,
|
| 307 |
+
"eval_wer": 0.3980946264183258,
|
| 308 |
+
"step": 2000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 19.63,
|
| 312 |
+
"learning_rate": 5.0407239819004526e-05,
|
| 313 |
+
"loss": 0.5598,
|
| 314 |
+
"step": 2100
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 19.63,
|
| 318 |
+
"eval_loss": 1.0339125394821167,
|
| 319 |
+
"eval_runtime": 42.5183,
|
| 320 |
+
"eval_samples_per_second": 20.062,
|
| 321 |
+
"eval_steps_per_second": 0.635,
|
| 322 |
+
"eval_wer": 0.39359880111325196,
|
| 323 |
+
"step": 2100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 20.56,
|
| 327 |
+
"learning_rate": 4.588235294117647e-05,
|
| 328 |
+
"loss": 0.5637,
|
| 329 |
+
"step": 2200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 20.56,
|
| 333 |
+
"eval_loss": 0.9660681486129761,
|
| 334 |
+
"eval_runtime": 44.6662,
|
| 335 |
+
"eval_samples_per_second": 19.097,
|
| 336 |
+
"eval_steps_per_second": 0.604,
|
| 337 |
+
"eval_wer": 0.399165061014772,
|
| 338 |
+
"step": 2200
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 21.5,
|
| 342 |
+
"learning_rate": 4.1357466063348414e-05,
|
| 343 |
+
"loss": 0.5396,
|
| 344 |
+
"step": 2300
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 21.5,
|
| 348 |
+
"eval_loss": 1.05593740940094,
|
| 349 |
+
"eval_runtime": 44.132,
|
| 350 |
+
"eval_samples_per_second": 19.328,
|
| 351 |
+
"eval_steps_per_second": 0.612,
|
| 352 |
+
"eval_wer": 0.3923142795975166,
|
| 353 |
+
"step": 2300
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 22.43,
|
| 357 |
+
"learning_rate": 3.683257918552037e-05,
|
| 358 |
+
"loss": 0.5001,
|
| 359 |
+
"step": 2400
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 22.43,
|
| 363 |
+
"eval_loss": 0.9710575938224792,
|
| 364 |
+
"eval_runtime": 43.7084,
|
| 365 |
+
"eval_samples_per_second": 19.516,
|
| 366 |
+
"eval_steps_per_second": 0.618,
|
| 367 |
+
"eval_wer": 0.38589167201883967,
|
| 368 |
+
"step": 2400
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 23.36,
|
| 372 |
+
"learning_rate": 3.230769230769231e-05,
|
| 373 |
+
"loss": 0.4797,
|
| 374 |
+
"step": 2500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 23.36,
|
| 378 |
+
"eval_loss": 0.9755498766899109,
|
| 379 |
+
"eval_runtime": 43.0821,
|
| 380 |
+
"eval_samples_per_second": 19.799,
|
| 381 |
+
"eval_steps_per_second": 0.627,
|
| 382 |
+
"eval_wer": 0.3775422821665596,
|
| 383 |
+
"step": 2500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 24.3,
|
| 387 |
+
"learning_rate": 2.7782805429864255e-05,
|
| 388 |
+
"loss": 0.4993,
|
| 389 |
+
"step": 2600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 24.3,
|
| 393 |
+
"eval_loss": 0.9937364459037781,
|
| 394 |
+
"eval_runtime": 43.0983,
|
| 395 |
+
"eval_samples_per_second": 19.792,
|
| 396 |
+
"eval_steps_per_second": 0.626,
|
| 397 |
+
"eval_wer": 0.37315350032113037,
|
| 398 |
+
"step": 2600
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 25.23,
|
| 402 |
+
"learning_rate": 2.3257918552036202e-05,
|
| 403 |
+
"loss": 0.4728,
|
| 404 |
+
"step": 2700
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 25.23,
|
| 408 |
+
"eval_loss": 1.028990387916565,
|
| 409 |
+
"eval_runtime": 44.5142,
|
| 410 |
+
"eval_samples_per_second": 19.162,
|
| 411 |
+
"eval_steps_per_second": 0.607,
|
| 412 |
+
"eval_wer": 0.37229715264397345,
|
| 413 |
+
"step": 2700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 26.17,
|
| 417 |
+
"learning_rate": 1.8733031674208146e-05,
|
| 418 |
+
"loss": 0.4479,
|
| 419 |
+
"step": 2800
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 26.17,
|
| 423 |
+
"eval_loss": 0.9995871186256409,
|
| 424 |
+
"eval_runtime": 43.2227,
|
| 425 |
+
"eval_samples_per_second": 19.735,
|
| 426 |
+
"eval_steps_per_second": 0.625,
|
| 427 |
+
"eval_wer": 0.36897880539499034,
|
| 428 |
+
"step": 2800
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 27.1,
|
| 432 |
+
"learning_rate": 1.4208144796380091e-05,
|
| 433 |
+
"loss": 0.4675,
|
| 434 |
+
"step": 2900
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 27.1,
|
| 438 |
+
"eval_loss": 1.0029648542404175,
|
| 439 |
+
"eval_runtime": 42.9456,
|
| 440 |
+
"eval_samples_per_second": 19.862,
|
| 441 |
+
"eval_steps_per_second": 0.629,
|
| 442 |
+
"eval_wer": 0.37101263112823807,
|
| 443 |
+
"step": 2900
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 28.04,
|
| 447 |
+
"learning_rate": 9.683257918552037e-06,
|
| 448 |
+
"loss": 0.4454,
|
| 449 |
+
"step": 3000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 28.04,
|
| 453 |
+
"eval_loss": 1.0169341564178467,
|
| 454 |
+
"eval_runtime": 43.9547,
|
| 455 |
+
"eval_samples_per_second": 19.406,
|
| 456 |
+
"eval_steps_per_second": 0.614,
|
| 457 |
+
"eval_wer": 0.37037037037037035,
|
| 458 |
+
"step": 3000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 28.97,
|
| 462 |
+
"learning_rate": 5.1583710407239815e-06,
|
| 463 |
+
"loss": 0.4473,
|
| 464 |
+
"step": 3100
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 28.97,
|
| 468 |
+
"eval_loss": 1.0175889730453491,
|
| 469 |
+
"eval_runtime": 43.8487,
|
| 470 |
+
"eval_samples_per_second": 19.453,
|
| 471 |
+
"eval_steps_per_second": 0.616,
|
| 472 |
+
"eval_wer": 0.3707985442089488,
|
| 473 |
+
"step": 3100
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 29.91,
|
| 477 |
+
"learning_rate": 6.334841628959276e-07,
|
| 478 |
+
"loss": 0.4005,
|
| 479 |
+
"step": 3200
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 29.91,
|
| 483 |
+
"eval_loss": 1.0105210542678833,
|
| 484 |
+
"eval_runtime": 42.6213,
|
| 485 |
+
"eval_samples_per_second": 20.013,
|
| 486 |
+
"eval_steps_per_second": 0.633,
|
| 487 |
+
"eval_wer": 0.36951402269321343,
|
| 488 |
+
"step": 3200
|
| 489 |
+
}
|
| 490 |
+
],
|
| 491 |
+
"max_steps": 3210,
|
| 492 |
+
"num_train_epochs": 30,
|
| 493 |
+
"total_flos": 4.853367218401499e+18,
|
| 494 |
+
"trial_name": null,
|
| 495 |
+
"trial_params": null
|
| 496 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cc93a3edab83807f4e8aa596b6a8205774da23b84143b518db4caa278252946
|
| 3 |
+
size 2671
|