Initial Model Commit
Browse files- checkpoint-14500/config.json +108 -0
- checkpoint-14500/optimizer.pt +3 -0
- checkpoint-14500/preprocessor_config.json +9 -0
- checkpoint-14500/pytorch_model.bin +3 -0
- checkpoint-14500/rng_state.pth +3 -0
- checkpoint-14500/scheduler.pt +3 -0
- checkpoint-14500/trainer_state.json +451 -0
- checkpoint-14500/training_args.bin +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- runs/Nov16_04-00-40_843f7eaa2874/1668571325.2454884/events.out.tfevents.1668571325.843f7eaa2874.84.1 +3 -0
- runs/Nov16_04-00-40_843f7eaa2874/events.out.tfevents.1668571325.843f7eaa2874.84.0 +3 -0
- runs/config.json +108 -0
- scheduler.pt +3 -0
- trainer_state.json +466 -0
- training_args.bin +3 -0
checkpoint-14500/config.json
ADDED
|
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "yip-i/uaspeech-pretrained",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 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": 0.0,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 256,
|
| 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": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.0,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"hidden_act": "gelu",
|
| 56 |
+
"hidden_dropout": 0.0,
|
| 57 |
+
"hidden_dropout_prob": 0.0,
|
| 58 |
+
"hidden_size": 768,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 3072,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 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": 0.65,
|
| 69 |
+
"model_type": "wav2vec2",
|
| 70 |
+
"num_adapter_layers": 3,
|
| 71 |
+
"num_attention_heads": 12,
|
| 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": 12,
|
| 78 |
+
"num_negatives": 100,
|
| 79 |
+
"output_hidden_size": 768,
|
| 80 |
+
"pad_token_id": 29,
|
| 81 |
+
"proj_codevector_dim": 256,
|
| 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 |
+
"torch_dtype": "float32",
|
| 104 |
+
"transformers_version": "4.23.1",
|
| 105 |
+
"use_weighted_layer_sum": false,
|
| 106 |
+
"vocab_size": 32,
|
| 107 |
+
"xvector_output_dim": 512
|
| 108 |
+
}
|
checkpoint-14500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a640733ad51beb91270ea0536a1e8b60c0e7ba83052ee9a77833fdd5ba418b2e
|
| 3 |
+
size 721685265
|
checkpoint-14500/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 |
+
}
|
checkpoint-14500/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:894a4fc68547b8d492bec3687b19a9a700d28e0aaf55184eb91c16dfca79a5c6
|
| 3 |
+
size 377702321
|
checkpoint-14500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dfb1d7904fe3d94af77b08186753b59b403c45eb06d0ba26b7c3baba2441ecd2
|
| 3 |
+
size 14567
|
checkpoint-14500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03479f5b2672c6bd21f75445486085d2c0e539ec574be6ce7a5264868a95803d
|
| 3 |
+
size 623
|
checkpoint-14500/trainer_state.json
ADDED
|
@@ -0,0 +1,451 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 28.155339805825243,
|
| 5 |
+
"global_step": 14500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.97,
|
| 12 |
+
"learning_rate": 5e-05,
|
| 13 |
+
"loss": 8.8221,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.97,
|
| 18 |
+
"eval_loss": 3.6322524547576904,
|
| 19 |
+
"eval_runtime": 46.46,
|
| 20 |
+
"eval_samples_per_second": 21.739,
|
| 21 |
+
"eval_steps_per_second": 2.734,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.94,
|
| 27 |
+
"learning_rate": 0.0001,
|
| 28 |
+
"loss": 3.4315,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.94,
|
| 33 |
+
"eval_loss": 3.614928722381592,
|
| 34 |
+
"eval_runtime": 46.1252,
|
| 35 |
+
"eval_samples_per_second": 21.897,
|
| 36 |
+
"eval_steps_per_second": 2.753,
|
| 37 |
+
"eval_wer": 1.001953125,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.91,
|
| 42 |
+
"learning_rate": 9.653979238754325e-05,
|
| 43 |
+
"loss": 3.2825,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.91,
|
| 48 |
+
"eval_loss": 4.486703395843506,
|
| 49 |
+
"eval_runtime": 46.2406,
|
| 50 |
+
"eval_samples_per_second": 21.842,
|
| 51 |
+
"eval_steps_per_second": 2.747,
|
| 52 |
+
"eval_wer": 1.0,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.88,
|
| 57 |
+
"learning_rate": 9.307958477508652e-05,
|
| 58 |
+
"loss": 3.0082,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.88,
|
| 63 |
+
"eval_loss": 4.054150104522705,
|
| 64 |
+
"eval_runtime": 46.7848,
|
| 65 |
+
"eval_samples_per_second": 21.588,
|
| 66 |
+
"eval_steps_per_second": 2.715,
|
| 67 |
+
"eval_wer": 1.069921875,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.85,
|
| 72 |
+
"learning_rate": 8.961937716262977e-05,
|
| 73 |
+
"loss": 2.9323,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.85,
|
| 78 |
+
"eval_loss": 3.1641268730163574,
|
| 79 |
+
"eval_runtime": 46.5642,
|
| 80 |
+
"eval_samples_per_second": 21.69,
|
| 81 |
+
"eval_steps_per_second": 2.727,
|
| 82 |
+
"eval_wer": 1.29140625,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 5.83,
|
| 87 |
+
"learning_rate": 8.615916955017301e-05,
|
| 88 |
+
"loss": 2.9131,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.83,
|
| 93 |
+
"eval_loss": 3.072361707687378,
|
| 94 |
+
"eval_runtime": 46.5467,
|
| 95 |
+
"eval_samples_per_second": 21.699,
|
| 96 |
+
"eval_steps_per_second": 2.728,
|
| 97 |
+
"eval_wer": 1.29140625,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.8,
|
| 102 |
+
"learning_rate": 8.269896193771626e-05,
|
| 103 |
+
"loss": 2.8623,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.8,
|
| 108 |
+
"eval_loss": 3.455599308013916,
|
| 109 |
+
"eval_runtime": 46.5142,
|
| 110 |
+
"eval_samples_per_second": 21.714,
|
| 111 |
+
"eval_steps_per_second": 2.73,
|
| 112 |
+
"eval_wer": 1.29140625,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 7.77,
|
| 117 |
+
"learning_rate": 7.923875432525953e-05,
|
| 118 |
+
"loss": 2.8501,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 7.77,
|
| 123 |
+
"eval_loss": 3.1545495986938477,
|
| 124 |
+
"eval_runtime": 46.4787,
|
| 125 |
+
"eval_samples_per_second": 21.73,
|
| 126 |
+
"eval_steps_per_second": 2.732,
|
| 127 |
+
"eval_wer": 1.29140625,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.74,
|
| 132 |
+
"learning_rate": 7.577854671280276e-05,
|
| 133 |
+
"loss": 2.8556,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 8.74,
|
| 138 |
+
"eval_loss": 3.0713276863098145,
|
| 139 |
+
"eval_runtime": 46.5645,
|
| 140 |
+
"eval_samples_per_second": 21.69,
|
| 141 |
+
"eval_steps_per_second": 2.727,
|
| 142 |
+
"eval_wer": 1.29140625,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 9.71,
|
| 147 |
+
"learning_rate": 7.231833910034602e-05,
|
| 148 |
+
"loss": 3.216,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 9.71,
|
| 153 |
+
"eval_loss": 3.619907855987549,
|
| 154 |
+
"eval_runtime": 46.8857,
|
| 155 |
+
"eval_samples_per_second": 21.542,
|
| 156 |
+
"eval_steps_per_second": 2.709,
|
| 157 |
+
"eval_wer": 1.290234375,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 10.68,
|
| 162 |
+
"learning_rate": 6.885813148788927e-05,
|
| 163 |
+
"loss": 3.0364,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 10.68,
|
| 168 |
+
"eval_loss": 3.2613348960876465,
|
| 169 |
+
"eval_runtime": 46.689,
|
| 170 |
+
"eval_samples_per_second": 21.633,
|
| 171 |
+
"eval_steps_per_second": 2.72,
|
| 172 |
+
"eval_wer": 1.290625,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 11.65,
|
| 177 |
+
"learning_rate": 6.539792387543253e-05,
|
| 178 |
+
"loss": 3.0891,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 11.65,
|
| 183 |
+
"eval_loss": 8.374248504638672,
|
| 184 |
+
"eval_runtime": 46.7038,
|
| 185 |
+
"eval_samples_per_second": 21.626,
|
| 186 |
+
"eval_steps_per_second": 2.719,
|
| 187 |
+
"eval_wer": 1.19453125,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.62,
|
| 192 |
+
"learning_rate": 6.193771626297578e-05,
|
| 193 |
+
"loss": 3.2234,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.62,
|
| 198 |
+
"eval_loss": 3.33834171295166,
|
| 199 |
+
"eval_runtime": 46.7037,
|
| 200 |
+
"eval_samples_per_second": 21.626,
|
| 201 |
+
"eval_steps_per_second": 2.719,
|
| 202 |
+
"eval_wer": 1.290234375,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 13.59,
|
| 207 |
+
"learning_rate": 5.847750865051903e-05,
|
| 208 |
+
"loss": 3.0869,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 13.59,
|
| 213 |
+
"eval_loss": 3.750485420227051,
|
| 214 |
+
"eval_runtime": 46.6392,
|
| 215 |
+
"eval_samples_per_second": 21.656,
|
| 216 |
+
"eval_steps_per_second": 2.723,
|
| 217 |
+
"eval_wer": 1.29140625,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 14.56,
|
| 222 |
+
"learning_rate": 5.501730103806229e-05,
|
| 223 |
+
"loss": 3.1178,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.56,
|
| 228 |
+
"eval_loss": 3.240990400314331,
|
| 229 |
+
"eval_runtime": 46.8161,
|
| 230 |
+
"eval_samples_per_second": 21.574,
|
| 231 |
+
"eval_steps_per_second": 2.713,
|
| 232 |
+
"eval_wer": 1.29140625,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.53,
|
| 237 |
+
"learning_rate": 5.1557093425605544e-05,
|
| 238 |
+
"loss": 3.119,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.53,
|
| 243 |
+
"eval_loss": 3.2141737937927246,
|
| 244 |
+
"eval_runtime": 46.8628,
|
| 245 |
+
"eval_samples_per_second": 21.552,
|
| 246 |
+
"eval_steps_per_second": 2.71,
|
| 247 |
+
"eval_wer": 1.01015625,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 16.5,
|
| 252 |
+
"learning_rate": 4.809688581314879e-05,
|
| 253 |
+
"loss": 3.2652,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 16.5,
|
| 258 |
+
"eval_loss": 3.2599737644195557,
|
| 259 |
+
"eval_runtime": 46.771,
|
| 260 |
+
"eval_samples_per_second": 21.595,
|
| 261 |
+
"eval_steps_per_second": 2.715,
|
| 262 |
+
"eval_wer": 1.29140625,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 17.48,
|
| 267 |
+
"learning_rate": 4.463667820069204e-05,
|
| 268 |
+
"loss": 2.8827,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 17.48,
|
| 273 |
+
"eval_loss": 3.091489791870117,
|
| 274 |
+
"eval_runtime": 47.0421,
|
| 275 |
+
"eval_samples_per_second": 21.47,
|
| 276 |
+
"eval_steps_per_second": 2.7,
|
| 277 |
+
"eval_wer": 1.29140625,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 18.45,
|
| 282 |
+
"learning_rate": 4.11764705882353e-05,
|
| 283 |
+
"loss": 2.8601,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.45,
|
| 288 |
+
"eval_loss": 3.061061143875122,
|
| 289 |
+
"eval_runtime": 46.9034,
|
| 290 |
+
"eval_samples_per_second": 21.534,
|
| 291 |
+
"eval_steps_per_second": 2.708,
|
| 292 |
+
"eval_wer": 1.29140625,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 19.42,
|
| 297 |
+
"learning_rate": 3.7716262975778546e-05,
|
| 298 |
+
"loss": 2.8083,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 19.42,
|
| 303 |
+
"eval_loss": 3.1701200008392334,
|
| 304 |
+
"eval_runtime": 47.1199,
|
| 305 |
+
"eval_samples_per_second": 21.435,
|
| 306 |
+
"eval_steps_per_second": 2.695,
|
| 307 |
+
"eval_wer": 1.281640625,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 20.39,
|
| 312 |
+
"learning_rate": 3.4256055363321795e-05,
|
| 313 |
+
"loss": 2.771,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 20.39,
|
| 318 |
+
"eval_loss": 3.059804916381836,
|
| 319 |
+
"eval_runtime": 47.1428,
|
| 320 |
+
"eval_samples_per_second": 21.424,
|
| 321 |
+
"eval_steps_per_second": 2.694,
|
| 322 |
+
"eval_wer": 1.29140625,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 21.36,
|
| 327 |
+
"learning_rate": 3.079584775086505e-05,
|
| 328 |
+
"loss": 2.7693,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.36,
|
| 333 |
+
"eval_loss": 3.0247151851654053,
|
| 334 |
+
"eval_runtime": 46.8824,
|
| 335 |
+
"eval_samples_per_second": 21.543,
|
| 336 |
+
"eval_steps_per_second": 2.709,
|
| 337 |
+
"eval_wer": 1.29140625,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 22.33,
|
| 342 |
+
"learning_rate": 2.733564013840831e-05,
|
| 343 |
+
"loss": 2.7861,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 22.33,
|
| 348 |
+
"eval_loss": 3.054908514022827,
|
| 349 |
+
"eval_runtime": 47.1813,
|
| 350 |
+
"eval_samples_per_second": 21.407,
|
| 351 |
+
"eval_steps_per_second": 2.692,
|
| 352 |
+
"eval_wer": 1.29140625,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 23.3,
|
| 357 |
+
"learning_rate": 2.387543252595156e-05,
|
| 358 |
+
"loss": 2.9903,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 23.3,
|
| 363 |
+
"eval_loss": 3.235363245010376,
|
| 364 |
+
"eval_runtime": 47.3429,
|
| 365 |
+
"eval_samples_per_second": 21.334,
|
| 366 |
+
"eval_steps_per_second": 2.683,
|
| 367 |
+
"eval_wer": 1.29140625,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 24.27,
|
| 372 |
+
"learning_rate": 2.041522491349481e-05,
|
| 373 |
+
"loss": 2.8883,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 24.27,
|
| 378 |
+
"eval_loss": 3.0525033473968506,
|
| 379 |
+
"eval_runtime": 46.4887,
|
| 380 |
+
"eval_samples_per_second": 21.726,
|
| 381 |
+
"eval_steps_per_second": 2.732,
|
| 382 |
+
"eval_wer": 1.29140625,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 25.24,
|
| 387 |
+
"learning_rate": 1.6955017301038063e-05,
|
| 388 |
+
"loss": 2.743,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 25.24,
|
| 393 |
+
"eval_loss": 3.0735745429992676,
|
| 394 |
+
"eval_runtime": 46.8938,
|
| 395 |
+
"eval_samples_per_second": 21.538,
|
| 396 |
+
"eval_steps_per_second": 2.708,
|
| 397 |
+
"eval_wer": 1.29140625,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 26.21,
|
| 402 |
+
"learning_rate": 1.3494809688581317e-05,
|
| 403 |
+
"loss": 2.7344,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 26.21,
|
| 408 |
+
"eval_loss": 3.0599749088287354,
|
| 409 |
+
"eval_runtime": 46.6796,
|
| 410 |
+
"eval_samples_per_second": 21.637,
|
| 411 |
+
"eval_steps_per_second": 2.721,
|
| 412 |
+
"eval_wer": 1.29140625,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 27.18,
|
| 417 |
+
"learning_rate": 1.0034602076124568e-05,
|
| 418 |
+
"loss": 2.697,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 27.18,
|
| 423 |
+
"eval_loss": 3.055352210998535,
|
| 424 |
+
"eval_runtime": 47.0781,
|
| 425 |
+
"eval_samples_per_second": 21.454,
|
| 426 |
+
"eval_steps_per_second": 2.698,
|
| 427 |
+
"eval_wer": 1.29140625,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 28.16,
|
| 432 |
+
"learning_rate": 6.5743944636678194e-06,
|
| 433 |
+
"loss": 2.7185,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 28.16,
|
| 438 |
+
"eval_loss": 3.130082607269287,
|
| 439 |
+
"eval_runtime": 46.9873,
|
| 440 |
+
"eval_samples_per_second": 21.495,
|
| 441 |
+
"eval_steps_per_second": 2.703,
|
| 442 |
+
"eval_wer": 1.29140625,
|
| 443 |
+
"step": 14500
|
| 444 |
+
}
|
| 445 |
+
],
|
| 446 |
+
"max_steps": 15450,
|
| 447 |
+
"num_train_epochs": 30,
|
| 448 |
+
"total_flos": 3.394197082480326e+18,
|
| 449 |
+
"trial_name": null,
|
| 450 |
+
"trial_params": null
|
| 451 |
+
}
|
checkpoint-14500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edd1494f81c01057f3d9d26d89ba7465c98ae4654ba92ee91478ee50711d9d8a
|
| 3 |
+
size 3375
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a7e9c25e7397388e1ec40c070c197655f75af4a4b653d019d3e8a782f27259e
|
| 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 sha256:4f3e124c9663ae5f60f043e4b500ab9a73bc6a111cbaa73aa6fedbfe598da36b
|
| 3 |
+
size 377702321
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6b4dd5805bfcf3a135735be94e4ca93b90101c2c4f2f220406ab3c95704e98b
|
| 3 |
+
size 14567
|
runs/Nov16_04-00-40_843f7eaa2874/1668571325.2454884/events.out.tfevents.1668571325.843f7eaa2874.84.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4638bee2308ad8a8749a9943b1509c4452d08b565a2f8c2dc1f7878e17c295ee
|
| 3 |
+
size 5471
|
runs/Nov16_04-00-40_843f7eaa2874/events.out.tfevents.1668571325.843f7eaa2874.84.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ca0a07e4c80b84b24a3f94854f94c71275019aa9d041604603dbb7906c4f33f
|
| 3 |
+
size 19721
|
runs/config.json
ADDED
|
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "yip-i/uaspeech-pretrained",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 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": 0.0,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 256,
|
| 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": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.0,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"hidden_act": "gelu",
|
| 56 |
+
"hidden_dropout": 0.0,
|
| 57 |
+
"hidden_dropout_prob": 0.0,
|
| 58 |
+
"hidden_size": 768,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 3072,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.0,
|
| 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": 0.65,
|
| 69 |
+
"model_type": "wav2vec2",
|
| 70 |
+
"num_adapter_layers": 3,
|
| 71 |
+
"num_attention_heads": 12,
|
| 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": 12,
|
| 78 |
+
"num_negatives": 100,
|
| 79 |
+
"output_hidden_size": 768,
|
| 80 |
+
"pad_token_id": 29,
|
| 81 |
+
"proj_codevector_dim": 256,
|
| 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 |
+
"torch_dtype": "float32",
|
| 104 |
+
"transformers_version": "4.23.1",
|
| 105 |
+
"use_weighted_layer_sum": false,
|
| 106 |
+
"vocab_size": 32,
|
| 107 |
+
"xvector_output_dim": 512
|
| 108 |
+
}
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56a25998e21c0b1ef3726cfa370f352a3517e332cba5746497dc922fb89907d1
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,466 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.12621359223301,
|
| 5 |
+
"global_step": 15000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.97,
|
| 12 |
+
"learning_rate": 5e-05,
|
| 13 |
+
"loss": 8.8221,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.97,
|
| 18 |
+
"eval_loss": 3.6322524547576904,
|
| 19 |
+
"eval_runtime": 46.46,
|
| 20 |
+
"eval_samples_per_second": 21.739,
|
| 21 |
+
"eval_steps_per_second": 2.734,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.94,
|
| 27 |
+
"learning_rate": 0.0001,
|
| 28 |
+
"loss": 3.4315,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.94,
|
| 33 |
+
"eval_loss": 3.614928722381592,
|
| 34 |
+
"eval_runtime": 46.1252,
|
| 35 |
+
"eval_samples_per_second": 21.897,
|
| 36 |
+
"eval_steps_per_second": 2.753,
|
| 37 |
+
"eval_wer": 1.001953125,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.91,
|
| 42 |
+
"learning_rate": 9.653979238754325e-05,
|
| 43 |
+
"loss": 3.2825,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.91,
|
| 48 |
+
"eval_loss": 4.486703395843506,
|
| 49 |
+
"eval_runtime": 46.2406,
|
| 50 |
+
"eval_samples_per_second": 21.842,
|
| 51 |
+
"eval_steps_per_second": 2.747,
|
| 52 |
+
"eval_wer": 1.0,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.88,
|
| 57 |
+
"learning_rate": 9.307958477508652e-05,
|
| 58 |
+
"loss": 3.0082,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.88,
|
| 63 |
+
"eval_loss": 4.054150104522705,
|
| 64 |
+
"eval_runtime": 46.7848,
|
| 65 |
+
"eval_samples_per_second": 21.588,
|
| 66 |
+
"eval_steps_per_second": 2.715,
|
| 67 |
+
"eval_wer": 1.069921875,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.85,
|
| 72 |
+
"learning_rate": 8.961937716262977e-05,
|
| 73 |
+
"loss": 2.9323,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.85,
|
| 78 |
+
"eval_loss": 3.1641268730163574,
|
| 79 |
+
"eval_runtime": 46.5642,
|
| 80 |
+
"eval_samples_per_second": 21.69,
|
| 81 |
+
"eval_steps_per_second": 2.727,
|
| 82 |
+
"eval_wer": 1.29140625,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 5.83,
|
| 87 |
+
"learning_rate": 8.615916955017301e-05,
|
| 88 |
+
"loss": 2.9131,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.83,
|
| 93 |
+
"eval_loss": 3.072361707687378,
|
| 94 |
+
"eval_runtime": 46.5467,
|
| 95 |
+
"eval_samples_per_second": 21.699,
|
| 96 |
+
"eval_steps_per_second": 2.728,
|
| 97 |
+
"eval_wer": 1.29140625,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.8,
|
| 102 |
+
"learning_rate": 8.269896193771626e-05,
|
| 103 |
+
"loss": 2.8623,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.8,
|
| 108 |
+
"eval_loss": 3.455599308013916,
|
| 109 |
+
"eval_runtime": 46.5142,
|
| 110 |
+
"eval_samples_per_second": 21.714,
|
| 111 |
+
"eval_steps_per_second": 2.73,
|
| 112 |
+
"eval_wer": 1.29140625,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 7.77,
|
| 117 |
+
"learning_rate": 7.923875432525953e-05,
|
| 118 |
+
"loss": 2.8501,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 7.77,
|
| 123 |
+
"eval_loss": 3.1545495986938477,
|
| 124 |
+
"eval_runtime": 46.4787,
|
| 125 |
+
"eval_samples_per_second": 21.73,
|
| 126 |
+
"eval_steps_per_second": 2.732,
|
| 127 |
+
"eval_wer": 1.29140625,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.74,
|
| 132 |
+
"learning_rate": 7.577854671280276e-05,
|
| 133 |
+
"loss": 2.8556,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 8.74,
|
| 138 |
+
"eval_loss": 3.0713276863098145,
|
| 139 |
+
"eval_runtime": 46.5645,
|
| 140 |
+
"eval_samples_per_second": 21.69,
|
| 141 |
+
"eval_steps_per_second": 2.727,
|
| 142 |
+
"eval_wer": 1.29140625,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 9.71,
|
| 147 |
+
"learning_rate": 7.231833910034602e-05,
|
| 148 |
+
"loss": 3.216,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 9.71,
|
| 153 |
+
"eval_loss": 3.619907855987549,
|
| 154 |
+
"eval_runtime": 46.8857,
|
| 155 |
+
"eval_samples_per_second": 21.542,
|
| 156 |
+
"eval_steps_per_second": 2.709,
|
| 157 |
+
"eval_wer": 1.290234375,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 10.68,
|
| 162 |
+
"learning_rate": 6.885813148788927e-05,
|
| 163 |
+
"loss": 3.0364,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 10.68,
|
| 168 |
+
"eval_loss": 3.2613348960876465,
|
| 169 |
+
"eval_runtime": 46.689,
|
| 170 |
+
"eval_samples_per_second": 21.633,
|
| 171 |
+
"eval_steps_per_second": 2.72,
|
| 172 |
+
"eval_wer": 1.290625,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 11.65,
|
| 177 |
+
"learning_rate": 6.539792387543253e-05,
|
| 178 |
+
"loss": 3.0891,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 11.65,
|
| 183 |
+
"eval_loss": 8.374248504638672,
|
| 184 |
+
"eval_runtime": 46.7038,
|
| 185 |
+
"eval_samples_per_second": 21.626,
|
| 186 |
+
"eval_steps_per_second": 2.719,
|
| 187 |
+
"eval_wer": 1.19453125,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.62,
|
| 192 |
+
"learning_rate": 6.193771626297578e-05,
|
| 193 |
+
"loss": 3.2234,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.62,
|
| 198 |
+
"eval_loss": 3.33834171295166,
|
| 199 |
+
"eval_runtime": 46.7037,
|
| 200 |
+
"eval_samples_per_second": 21.626,
|
| 201 |
+
"eval_steps_per_second": 2.719,
|
| 202 |
+
"eval_wer": 1.290234375,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 13.59,
|
| 207 |
+
"learning_rate": 5.847750865051903e-05,
|
| 208 |
+
"loss": 3.0869,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 13.59,
|
| 213 |
+
"eval_loss": 3.750485420227051,
|
| 214 |
+
"eval_runtime": 46.6392,
|
| 215 |
+
"eval_samples_per_second": 21.656,
|
| 216 |
+
"eval_steps_per_second": 2.723,
|
| 217 |
+
"eval_wer": 1.29140625,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 14.56,
|
| 222 |
+
"learning_rate": 5.501730103806229e-05,
|
| 223 |
+
"loss": 3.1178,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.56,
|
| 228 |
+
"eval_loss": 3.240990400314331,
|
| 229 |
+
"eval_runtime": 46.8161,
|
| 230 |
+
"eval_samples_per_second": 21.574,
|
| 231 |
+
"eval_steps_per_second": 2.713,
|
| 232 |
+
"eval_wer": 1.29140625,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.53,
|
| 237 |
+
"learning_rate": 5.1557093425605544e-05,
|
| 238 |
+
"loss": 3.119,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 15.53,
|
| 243 |
+
"eval_loss": 3.2141737937927246,
|
| 244 |
+
"eval_runtime": 46.8628,
|
| 245 |
+
"eval_samples_per_second": 21.552,
|
| 246 |
+
"eval_steps_per_second": 2.71,
|
| 247 |
+
"eval_wer": 1.01015625,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 16.5,
|
| 252 |
+
"learning_rate": 4.809688581314879e-05,
|
| 253 |
+
"loss": 3.2652,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 16.5,
|
| 258 |
+
"eval_loss": 3.2599737644195557,
|
| 259 |
+
"eval_runtime": 46.771,
|
| 260 |
+
"eval_samples_per_second": 21.595,
|
| 261 |
+
"eval_steps_per_second": 2.715,
|
| 262 |
+
"eval_wer": 1.29140625,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 17.48,
|
| 267 |
+
"learning_rate": 4.463667820069204e-05,
|
| 268 |
+
"loss": 2.8827,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 17.48,
|
| 273 |
+
"eval_loss": 3.091489791870117,
|
| 274 |
+
"eval_runtime": 47.0421,
|
| 275 |
+
"eval_samples_per_second": 21.47,
|
| 276 |
+
"eval_steps_per_second": 2.7,
|
| 277 |
+
"eval_wer": 1.29140625,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 18.45,
|
| 282 |
+
"learning_rate": 4.11764705882353e-05,
|
| 283 |
+
"loss": 2.8601,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.45,
|
| 288 |
+
"eval_loss": 3.061061143875122,
|
| 289 |
+
"eval_runtime": 46.9034,
|
| 290 |
+
"eval_samples_per_second": 21.534,
|
| 291 |
+
"eval_steps_per_second": 2.708,
|
| 292 |
+
"eval_wer": 1.29140625,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 19.42,
|
| 297 |
+
"learning_rate": 3.7716262975778546e-05,
|
| 298 |
+
"loss": 2.8083,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 19.42,
|
| 303 |
+
"eval_loss": 3.1701200008392334,
|
| 304 |
+
"eval_runtime": 47.1199,
|
| 305 |
+
"eval_samples_per_second": 21.435,
|
| 306 |
+
"eval_steps_per_second": 2.695,
|
| 307 |
+
"eval_wer": 1.281640625,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 20.39,
|
| 312 |
+
"learning_rate": 3.4256055363321795e-05,
|
| 313 |
+
"loss": 2.771,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 20.39,
|
| 318 |
+
"eval_loss": 3.059804916381836,
|
| 319 |
+
"eval_runtime": 47.1428,
|
| 320 |
+
"eval_samples_per_second": 21.424,
|
| 321 |
+
"eval_steps_per_second": 2.694,
|
| 322 |
+
"eval_wer": 1.29140625,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 21.36,
|
| 327 |
+
"learning_rate": 3.079584775086505e-05,
|
| 328 |
+
"loss": 2.7693,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.36,
|
| 333 |
+
"eval_loss": 3.0247151851654053,
|
| 334 |
+
"eval_runtime": 46.8824,
|
| 335 |
+
"eval_samples_per_second": 21.543,
|
| 336 |
+
"eval_steps_per_second": 2.709,
|
| 337 |
+
"eval_wer": 1.29140625,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 22.33,
|
| 342 |
+
"learning_rate": 2.733564013840831e-05,
|
| 343 |
+
"loss": 2.7861,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 22.33,
|
| 348 |
+
"eval_loss": 3.054908514022827,
|
| 349 |
+
"eval_runtime": 47.1813,
|
| 350 |
+
"eval_samples_per_second": 21.407,
|
| 351 |
+
"eval_steps_per_second": 2.692,
|
| 352 |
+
"eval_wer": 1.29140625,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 23.3,
|
| 357 |
+
"learning_rate": 2.387543252595156e-05,
|
| 358 |
+
"loss": 2.9903,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 23.3,
|
| 363 |
+
"eval_loss": 3.235363245010376,
|
| 364 |
+
"eval_runtime": 47.3429,
|
| 365 |
+
"eval_samples_per_second": 21.334,
|
| 366 |
+
"eval_steps_per_second": 2.683,
|
| 367 |
+
"eval_wer": 1.29140625,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 24.27,
|
| 372 |
+
"learning_rate": 2.041522491349481e-05,
|
| 373 |
+
"loss": 2.8883,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 24.27,
|
| 378 |
+
"eval_loss": 3.0525033473968506,
|
| 379 |
+
"eval_runtime": 46.4887,
|
| 380 |
+
"eval_samples_per_second": 21.726,
|
| 381 |
+
"eval_steps_per_second": 2.732,
|
| 382 |
+
"eval_wer": 1.29140625,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 25.24,
|
| 387 |
+
"learning_rate": 1.6955017301038063e-05,
|
| 388 |
+
"loss": 2.743,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 25.24,
|
| 393 |
+
"eval_loss": 3.0735745429992676,
|
| 394 |
+
"eval_runtime": 46.8938,
|
| 395 |
+
"eval_samples_per_second": 21.538,
|
| 396 |
+
"eval_steps_per_second": 2.708,
|
| 397 |
+
"eval_wer": 1.29140625,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 26.21,
|
| 402 |
+
"learning_rate": 1.3494809688581317e-05,
|
| 403 |
+
"loss": 2.7344,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 26.21,
|
| 408 |
+
"eval_loss": 3.0599749088287354,
|
| 409 |
+
"eval_runtime": 46.6796,
|
| 410 |
+
"eval_samples_per_second": 21.637,
|
| 411 |
+
"eval_steps_per_second": 2.721,
|
| 412 |
+
"eval_wer": 1.29140625,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 27.18,
|
| 417 |
+
"learning_rate": 1.0034602076124568e-05,
|
| 418 |
+
"loss": 2.697,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 27.18,
|
| 423 |
+
"eval_loss": 3.055352210998535,
|
| 424 |
+
"eval_runtime": 47.0781,
|
| 425 |
+
"eval_samples_per_second": 21.454,
|
| 426 |
+
"eval_steps_per_second": 2.698,
|
| 427 |
+
"eval_wer": 1.29140625,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 28.16,
|
| 432 |
+
"learning_rate": 6.5743944636678194e-06,
|
| 433 |
+
"loss": 2.7185,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 28.16,
|
| 438 |
+
"eval_loss": 3.130082607269287,
|
| 439 |
+
"eval_runtime": 46.9873,
|
| 440 |
+
"eval_samples_per_second": 21.495,
|
| 441 |
+
"eval_steps_per_second": 2.703,
|
| 442 |
+
"eval_wer": 1.29140625,
|
| 443 |
+
"step": 14500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 29.13,
|
| 447 |
+
"learning_rate": 3.1141868512110726e-06,
|
| 448 |
+
"loss": 2.6788,
|
| 449 |
+
"step": 15000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 29.13,
|
| 453 |
+
"eval_loss": 3.2066450119018555,
|
| 454 |
+
"eval_runtime": 46.7382,
|
| 455 |
+
"eval_samples_per_second": 21.61,
|
| 456 |
+
"eval_steps_per_second": 2.717,
|
| 457 |
+
"eval_wer": 1.29140625,
|
| 458 |
+
"step": 15000
|
| 459 |
+
}
|
| 460 |
+
],
|
| 461 |
+
"max_steps": 15450,
|
| 462 |
+
"num_train_epochs": 30,
|
| 463 |
+
"total_flos": 3.5108035253147873e+18,
|
| 464 |
+
"trial_name": null,
|
| 465 |
+
"trial_params": null
|
| 466 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edd1494f81c01057f3d9d26d89ba7465c98ae4654ba92ee91478ee50711d9d8a
|
| 3 |
+
size 3375
|