leecho
commited on
Commit
•
e412287
1
Parent(s):
e36bb34
commit from fleek
Browse files- config.json +76 -0
- optimizer.pt +3 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- trainer_state.json +628 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 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 |
+
"conv_bias": true,
|
| 11 |
+
"conv_dim": [
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512
|
| 19 |
+
],
|
| 20 |
+
"conv_kernel": [
|
| 21 |
+
10,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
2,
|
| 27 |
+
2
|
| 28 |
+
],
|
| 29 |
+
"conv_stride": [
|
| 30 |
+
5,
|
| 31 |
+
2,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2
|
| 37 |
+
],
|
| 38 |
+
"ctc_loss_reduction": "mean",
|
| 39 |
+
"ctc_zero_infinity": false,
|
| 40 |
+
"do_stable_layer_norm": true,
|
| 41 |
+
"eos_token_id": 2,
|
| 42 |
+
"feat_extract_activation": "gelu",
|
| 43 |
+
"feat_extract_dropout": 0.0,
|
| 44 |
+
"feat_extract_norm": "layer",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"final_dropout": 0.0,
|
| 47 |
+
"gradient_checkpointing": true,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 1024,
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 4096,
|
| 53 |
+
"layer_norm_eps": 1e-05,
|
| 54 |
+
"layerdrop": 0.1,
|
| 55 |
+
"mask_channel_length": 10,
|
| 56 |
+
"mask_channel_min_space": 1,
|
| 57 |
+
"mask_channel_other": 0.0,
|
| 58 |
+
"mask_channel_prob": 0.0,
|
| 59 |
+
"mask_channel_selection": "static",
|
| 60 |
+
"mask_feature_length": 10,
|
| 61 |
+
"mask_feature_prob": 0.0,
|
| 62 |
+
"mask_time_length": 10,
|
| 63 |
+
"mask_time_min_space": 1,
|
| 64 |
+
"mask_time_other": 0.0,
|
| 65 |
+
"mask_time_prob": 0.05,
|
| 66 |
+
"mask_time_selection": "static",
|
| 67 |
+
"model_type": "wav2vec2",
|
| 68 |
+
"num_attention_heads": 16,
|
| 69 |
+
"num_conv_pos_embedding_groups": 16,
|
| 70 |
+
"num_conv_pos_embeddings": 128,
|
| 71 |
+
"num_feat_extract_layers": 7,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"pad_token_id": 49,
|
| 74 |
+
"transformers_version": "4.4.0",
|
| 75 |
+
"vocab_size": 50
|
| 76 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:843ef1c076d692fbab55c293fa2d7b7abb30a66da9ba17af87118705b66a1b5a
|
| 3 |
+
size 2490487175
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_size": 1,
|
| 4 |
+
"padding_side": "right",
|
| 5 |
+
"padding_value": 0.0,
|
| 6 |
+
"return_attention_mask": false,
|
| 7 |
+
"sampling_rate": 16000
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5cff7c083c2d21af1ae70f91a3628e930e121f213c8db1dd9208aaca3458734f
|
| 3 |
+
size 1262138839
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bb8b86bc4e378631af9b45c4f21eb4a7919ab5a185958820da51aa1f8703a89
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,628 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 27.272727272727273,
|
| 5 |
+
"global_step": 3600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.3,
|
| 12 |
+
"learning_rate": 2.3999999999999997e-05,
|
| 13 |
+
"loss": 17.8657,
|
| 14 |
+
"step": 40
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.61,
|
| 18 |
+
"learning_rate": 4.7999999999999994e-05,
|
| 19 |
+
"loss": 12.104,
|
| 20 |
+
"step": 80
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.91,
|
| 24 |
+
"learning_rate": 7.199999999999999e-05,
|
| 25 |
+
"loss": 4.7542,
|
| 26 |
+
"step": 120
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.21,
|
| 30 |
+
"learning_rate": 9.599999999999999e-05,
|
| 31 |
+
"loss": 3.5251,
|
| 32 |
+
"step": 160
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.52,
|
| 36 |
+
"learning_rate": 0.00011999999999999999,
|
| 37 |
+
"loss": 3.2968,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.82,
|
| 42 |
+
"learning_rate": 0.00014399999999999998,
|
| 43 |
+
"loss": 3.1885,
|
| 44 |
+
"step": 240
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.12,
|
| 48 |
+
"learning_rate": 0.000168,
|
| 49 |
+
"loss": 3.173,
|
| 50 |
+
"step": 280
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.42,
|
| 54 |
+
"learning_rate": 0.00019199999999999998,
|
| 55 |
+
"loss": 3.1139,
|
| 56 |
+
"step": 320
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.73,
|
| 60 |
+
"learning_rate": 0.00021599999999999996,
|
| 61 |
+
"loss": 3.103,
|
| 62 |
+
"step": 360
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.03,
|
| 66 |
+
"learning_rate": 0.00023999999999999998,
|
| 67 |
+
"loss": 3.0862,
|
| 68 |
+
"step": 400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.03,
|
| 72 |
+
"eval_loss": 3.177677869796753,
|
| 73 |
+
"eval_runtime": 53.6086,
|
| 74 |
+
"eval_samples_per_second": 22.403,
|
| 75 |
+
"eval_wer": 1.0,
|
| 76 |
+
"step": 400
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.33,
|
| 80 |
+
"learning_rate": 0.00026399999999999997,
|
| 81 |
+
"loss": 3.0227,
|
| 82 |
+
"step": 440
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 3.64,
|
| 86 |
+
"learning_rate": 0.00028799999999999995,
|
| 87 |
+
"loss": 2.6934,
|
| 88 |
+
"step": 480
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.94,
|
| 92 |
+
"learning_rate": 0.0002982658959537572,
|
| 93 |
+
"loss": 2.1587,
|
| 94 |
+
"step": 520
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 4.24,
|
| 98 |
+
"learning_rate": 0.00029479768786127165,
|
| 99 |
+
"loss": 1.7134,
|
| 100 |
+
"step": 560
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.55,
|
| 104 |
+
"learning_rate": 0.0002913294797687861,
|
| 105 |
+
"loss": 1.4758,
|
| 106 |
+
"step": 600
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.85,
|
| 110 |
+
"learning_rate": 0.00028786127167630053,
|
| 111 |
+
"loss": 1.311,
|
| 112 |
+
"step": 640
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 5.15,
|
| 116 |
+
"learning_rate": 0.000284393063583815,
|
| 117 |
+
"loss": 1.1944,
|
| 118 |
+
"step": 680
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.45,
|
| 122 |
+
"learning_rate": 0.00028092485549132947,
|
| 123 |
+
"loss": 1.082,
|
| 124 |
+
"step": 720
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.76,
|
| 128 |
+
"learning_rate": 0.0002774566473988439,
|
| 129 |
+
"loss": 1.024,
|
| 130 |
+
"step": 760
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 6.06,
|
| 134 |
+
"learning_rate": 0.00027398843930635835,
|
| 135 |
+
"loss": 0.9718,
|
| 136 |
+
"step": 800
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 6.06,
|
| 140 |
+
"eval_loss": 1.0074659585952759,
|
| 141 |
+
"eval_runtime": 53.837,
|
| 142 |
+
"eval_samples_per_second": 22.308,
|
| 143 |
+
"eval_wer": 0.8955532574974147,
|
| 144 |
+
"step": 800
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.36,
|
| 148 |
+
"learning_rate": 0.0002705202312138728,
|
| 149 |
+
"loss": 0.8674,
|
| 150 |
+
"step": 840
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.67,
|
| 154 |
+
"learning_rate": 0.0002670520231213873,
|
| 155 |
+
"loss": 0.8369,
|
| 156 |
+
"step": 880
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 6.97,
|
| 160 |
+
"learning_rate": 0.0002635838150289017,
|
| 161 |
+
"loss": 0.8437,
|
| 162 |
+
"step": 920
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 7.27,
|
| 166 |
+
"learning_rate": 0.00026011560693641616,
|
| 167 |
+
"loss": 0.7578,
|
| 168 |
+
"step": 960
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.58,
|
| 172 |
+
"learning_rate": 0.00025664739884393063,
|
| 173 |
+
"loss": 0.7275,
|
| 174 |
+
"step": 1000
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 7.88,
|
| 178 |
+
"learning_rate": 0.00025317919075144504,
|
| 179 |
+
"loss": 0.7207,
|
| 180 |
+
"step": 1040
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 8.18,
|
| 184 |
+
"learning_rate": 0.0002497109826589595,
|
| 185 |
+
"loss": 0.6415,
|
| 186 |
+
"step": 1080
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 8.48,
|
| 190 |
+
"learning_rate": 0.000246242774566474,
|
| 191 |
+
"loss": 0.6446,
|
| 192 |
+
"step": 1120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 8.79,
|
| 196 |
+
"learning_rate": 0.00024277456647398842,
|
| 197 |
+
"loss": 0.6243,
|
| 198 |
+
"step": 1160
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 9.09,
|
| 202 |
+
"learning_rate": 0.0002393063583815029,
|
| 203 |
+
"loss": 0.6085,
|
| 204 |
+
"step": 1200
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 9.09,
|
| 208 |
+
"eval_loss": 0.8931847214698792,
|
| 209 |
+
"eval_runtime": 53.9711,
|
| 210 |
+
"eval_samples_per_second": 22.253,
|
| 211 |
+
"eval_wer": 0.797015807357069,
|
| 212 |
+
"step": 1200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 9.39,
|
| 216 |
+
"learning_rate": 0.00023583815028901733,
|
| 217 |
+
"loss": 0.5279,
|
| 218 |
+
"step": 1240
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.7,
|
| 222 |
+
"learning_rate": 0.00023236994219653174,
|
| 223 |
+
"loss": 0.5279,
|
| 224 |
+
"step": 1280
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 10.0,
|
| 228 |
+
"learning_rate": 0.0002289017341040462,
|
| 229 |
+
"loss": 0.5421,
|
| 230 |
+
"step": 1320
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 10.3,
|
| 234 |
+
"learning_rate": 0.00022543352601156065,
|
| 235 |
+
"loss": 0.4856,
|
| 236 |
+
"step": 1360
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 10.61,
|
| 240 |
+
"learning_rate": 0.00022196531791907512,
|
| 241 |
+
"loss": 0.5182,
|
| 242 |
+
"step": 1400
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 10.91,
|
| 246 |
+
"learning_rate": 0.00021849710982658956,
|
| 247 |
+
"loss": 0.4756,
|
| 248 |
+
"step": 1440
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 11.21,
|
| 252 |
+
"learning_rate": 0.00021502890173410403,
|
| 253 |
+
"loss": 0.479,
|
| 254 |
+
"step": 1480
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 11.52,
|
| 258 |
+
"learning_rate": 0.00021156069364161847,
|
| 259 |
+
"loss": 0.4419,
|
| 260 |
+
"step": 1520
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 11.82,
|
| 264 |
+
"learning_rate": 0.00020809248554913294,
|
| 265 |
+
"loss": 0.448,
|
| 266 |
+
"step": 1560
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 12.12,
|
| 270 |
+
"learning_rate": 0.00020462427745664738,
|
| 271 |
+
"loss": 0.4295,
|
| 272 |
+
"step": 1600
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 12.12,
|
| 276 |
+
"eval_loss": 0.9030922055244446,
|
| 277 |
+
"eval_runtime": 54.1136,
|
| 278 |
+
"eval_samples_per_second": 22.194,
|
| 279 |
+
"eval_wer": 0.7748559609986704,
|
| 280 |
+
"step": 1600
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 12.42,
|
| 284 |
+
"learning_rate": 0.00020115606936416184,
|
| 285 |
+
"loss": 0.3976,
|
| 286 |
+
"step": 1640
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 12.73,
|
| 290 |
+
"learning_rate": 0.00019768786127167629,
|
| 291 |
+
"loss": 0.3699,
|
| 292 |
+
"step": 1680
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 13.03,
|
| 296 |
+
"learning_rate": 0.00019421965317919073,
|
| 297 |
+
"loss": 0.4159,
|
| 298 |
+
"step": 1720
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 13.33,
|
| 302 |
+
"learning_rate": 0.0001907514450867052,
|
| 303 |
+
"loss": 0.3221,
|
| 304 |
+
"step": 1760
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 13.64,
|
| 308 |
+
"learning_rate": 0.00018728323699421963,
|
| 309 |
+
"loss": 0.3642,
|
| 310 |
+
"step": 1800
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 13.94,
|
| 314 |
+
"learning_rate": 0.0001838150289017341,
|
| 315 |
+
"loss": 0.3832,
|
| 316 |
+
"step": 1840
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 14.24,
|
| 320 |
+
"learning_rate": 0.00018034682080924854,
|
| 321 |
+
"loss": 0.3367,
|
| 322 |
+
"step": 1880
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 14.55,
|
| 326 |
+
"learning_rate": 0.000176878612716763,
|
| 327 |
+
"loss": 0.3152,
|
| 328 |
+
"step": 1920
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 14.85,
|
| 332 |
+
"learning_rate": 0.00017341040462427745,
|
| 333 |
+
"loss": 0.3246,
|
| 334 |
+
"step": 1960
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 15.15,
|
| 338 |
+
"learning_rate": 0.00016994219653179192,
|
| 339 |
+
"loss": 0.3012,
|
| 340 |
+
"step": 2000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 15.15,
|
| 344 |
+
"eval_loss": 0.9842168688774109,
|
| 345 |
+
"eval_runtime": 54.3687,
|
| 346 |
+
"eval_samples_per_second": 22.09,
|
| 347 |
+
"eval_wer": 0.7636283055104152,
|
| 348 |
+
"step": 2000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 15.45,
|
| 352 |
+
"learning_rate": 0.00016647398843930633,
|
| 353 |
+
"loss": 0.3051,
|
| 354 |
+
"step": 2040
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 15.76,
|
| 358 |
+
"learning_rate": 0.00016300578034682077,
|
| 359 |
+
"loss": 0.2966,
|
| 360 |
+
"step": 2080
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 16.06,
|
| 364 |
+
"learning_rate": 0.00015953757225433524,
|
| 365 |
+
"loss": 0.309,
|
| 366 |
+
"step": 2120
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 16.36,
|
| 370 |
+
"learning_rate": 0.00015606936416184968,
|
| 371 |
+
"loss": 0.2553,
|
| 372 |
+
"step": 2160
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 16.67,
|
| 376 |
+
"learning_rate": 0.00015260115606936415,
|
| 377 |
+
"loss": 0.279,
|
| 378 |
+
"step": 2200
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 16.97,
|
| 382 |
+
"learning_rate": 0.0001491329479768786,
|
| 383 |
+
"loss": 0.2962,
|
| 384 |
+
"step": 2240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 17.27,
|
| 388 |
+
"learning_rate": 0.00014566473988439306,
|
| 389 |
+
"loss": 0.2829,
|
| 390 |
+
"step": 2280
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 17.58,
|
| 394 |
+
"learning_rate": 0.0001421965317919075,
|
| 395 |
+
"loss": 0.2428,
|
| 396 |
+
"step": 2320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 17.88,
|
| 400 |
+
"learning_rate": 0.00013872832369942194,
|
| 401 |
+
"loss": 0.2434,
|
| 402 |
+
"step": 2360
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 18.18,
|
| 406 |
+
"learning_rate": 0.0001352601156069364,
|
| 407 |
+
"loss": 0.2513,
|
| 408 |
+
"step": 2400
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 18.18,
|
| 412 |
+
"eval_loss": 1.0579547882080078,
|
| 413 |
+
"eval_runtime": 84.4423,
|
| 414 |
+
"eval_samples_per_second": 14.223,
|
| 415 |
+
"eval_wer": 0.7682080070911508,
|
| 416 |
+
"step": 2400
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 18.48,
|
| 420 |
+
"learning_rate": 0.00013179190751445085,
|
| 421 |
+
"loss": 0.2331,
|
| 422 |
+
"step": 2440
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 18.79,
|
| 426 |
+
"learning_rate": 0.00012832369942196532,
|
| 427 |
+
"loss": 0.2275,
|
| 428 |
+
"step": 2480
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 19.09,
|
| 432 |
+
"learning_rate": 0.00012485549132947976,
|
| 433 |
+
"loss": 0.2275,
|
| 434 |
+
"step": 2520
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 19.39,
|
| 438 |
+
"learning_rate": 0.00012138728323699421,
|
| 439 |
+
"loss": 0.2174,
|
| 440 |
+
"step": 2560
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 19.7,
|
| 444 |
+
"learning_rate": 0.00011791907514450866,
|
| 445 |
+
"loss": 0.2202,
|
| 446 |
+
"step": 2600
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 20.0,
|
| 450 |
+
"learning_rate": 0.0001144508670520231,
|
| 451 |
+
"loss": 0.2291,
|
| 452 |
+
"step": 2640
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 20.3,
|
| 456 |
+
"learning_rate": 0.00011098265895953756,
|
| 457 |
+
"loss": 0.1909,
|
| 458 |
+
"step": 2680
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 20.61,
|
| 462 |
+
"learning_rate": 0.00010751445086705201,
|
| 463 |
+
"loss": 0.1897,
|
| 464 |
+
"step": 2720
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 20.91,
|
| 468 |
+
"learning_rate": 0.00010404624277456647,
|
| 469 |
+
"loss": 0.22,
|
| 470 |
+
"step": 2760
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 21.21,
|
| 474 |
+
"learning_rate": 0.00010057803468208092,
|
| 475 |
+
"loss": 0.194,
|
| 476 |
+
"step": 2800
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 21.21,
|
| 480 |
+
"eval_loss": 1.0953465700149536,
|
| 481 |
+
"eval_runtime": 79.9574,
|
| 482 |
+
"eval_samples_per_second": 15.02,
|
| 483 |
+
"eval_wer": 0.7524006500221598,
|
| 484 |
+
"step": 2800
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 21.52,
|
| 488 |
+
"learning_rate": 9.710982658959536e-05,
|
| 489 |
+
"loss": 0.1862,
|
| 490 |
+
"step": 2840
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 21.82,
|
| 494 |
+
"learning_rate": 9.364161849710982e-05,
|
| 495 |
+
"loss": 0.2012,
|
| 496 |
+
"step": 2880
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 22.12,
|
| 500 |
+
"learning_rate": 9.017341040462427e-05,
|
| 501 |
+
"loss": 0.1823,
|
| 502 |
+
"step": 2920
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 22.42,
|
| 506 |
+
"learning_rate": 8.670520231213873e-05,
|
| 507 |
+
"loss": 0.1885,
|
| 508 |
+
"step": 2960
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 22.73,
|
| 512 |
+
"learning_rate": 8.323699421965317e-05,
|
| 513 |
+
"loss": 0.1616,
|
| 514 |
+
"step": 3000
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 23.03,
|
| 518 |
+
"learning_rate": 7.976878612716762e-05,
|
| 519 |
+
"loss": 0.1893,
|
| 520 |
+
"step": 3040
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 23.33,
|
| 524 |
+
"learning_rate": 7.630057803468207e-05,
|
| 525 |
+
"loss": 0.1879,
|
| 526 |
+
"step": 3080
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 23.64,
|
| 530 |
+
"learning_rate": 7.283236994219653e-05,
|
| 531 |
+
"loss": 0.1759,
|
| 532 |
+
"step": 3120
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 23.94,
|
| 536 |
+
"learning_rate": 6.936416184971097e-05,
|
| 537 |
+
"loss": 0.1703,
|
| 538 |
+
"step": 3160
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 24.24,
|
| 542 |
+
"learning_rate": 6.589595375722542e-05,
|
| 543 |
+
"loss": 0.1679,
|
| 544 |
+
"step": 3200
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 24.24,
|
| 548 |
+
"eval_loss": 1.1073198318481445,
|
| 549 |
+
"eval_runtime": 79.6701,
|
| 550 |
+
"eval_samples_per_second": 15.075,
|
| 551 |
+
"eval_wer": 0.7289112128822574,
|
| 552 |
+
"step": 3200
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 24.55,
|
| 556 |
+
"learning_rate": 6.242774566473988e-05,
|
| 557 |
+
"loss": 0.1663,
|
| 558 |
+
"step": 3240
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 24.85,
|
| 562 |
+
"learning_rate": 5.895953757225433e-05,
|
| 563 |
+
"loss": 0.1726,
|
| 564 |
+
"step": 3280
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 25.15,
|
| 568 |
+
"learning_rate": 5.549132947976878e-05,
|
| 569 |
+
"loss": 0.1733,
|
| 570 |
+
"step": 3320
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 25.45,
|
| 574 |
+
"learning_rate": 5.2023121387283234e-05,
|
| 575 |
+
"loss": 0.1764,
|
| 576 |
+
"step": 3360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 25.76,
|
| 580 |
+
"learning_rate": 4.855491329479768e-05,
|
| 581 |
+
"loss": 0.14,
|
| 582 |
+
"step": 3400
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 26.06,
|
| 586 |
+
"learning_rate": 4.5086705202312136e-05,
|
| 587 |
+
"loss": 0.164,
|
| 588 |
+
"step": 3440
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 26.36,
|
| 592 |
+
"learning_rate": 4.161849710982658e-05,
|
| 593 |
+
"loss": 0.1569,
|
| 594 |
+
"step": 3480
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 26.67,
|
| 598 |
+
"learning_rate": 3.815028901734104e-05,
|
| 599 |
+
"loss": 0.1674,
|
| 600 |
+
"step": 3520
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 26.97,
|
| 604 |
+
"learning_rate": 3.4682080924855485e-05,
|
| 605 |
+
"loss": 0.1635,
|
| 606 |
+
"step": 3560
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 27.27,
|
| 610 |
+
"learning_rate": 3.121387283236994e-05,
|
| 611 |
+
"loss": 0.1372,
|
| 612 |
+
"step": 3600
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 27.27,
|
| 616 |
+
"eval_loss": 1.149294376373291,
|
| 617 |
+
"eval_runtime": 54.3637,
|
| 618 |
+
"eval_samples_per_second": 22.092,
|
| 619 |
+
"eval_wer": 0.7225587235928498,
|
| 620 |
+
"step": 3600
|
| 621 |
+
}
|
| 622 |
+
],
|
| 623 |
+
"max_steps": 3960,
|
| 624 |
+
"num_train_epochs": 30,
|
| 625 |
+
"total_flos": 1.107485248885501e+19,
|
| 626 |
+
"trial_name": null,
|
| 627 |
+
"trial_params": null
|
| 628 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e299b5eff76b2f0965206b7254308074518f08e05f5b99d0dc57f79e50d7bd4
|
| 3 |
+
size 2287
|