Ayen
commited on
Commit
•
c63997e
1
Parent(s):
d937792
Balikas
Browse files- new_iloko_model/added_tokens.json +1 -0
- new_iloko_model/checkpoint-2600/config.json +87 -0
- new_iloko_model/checkpoint-2600/preprocessor_config.json +9 -0
- new_iloko_model/checkpoint-2600/pytorch_model.bin +3 -0
- new_iloko_model/checkpoint-2600/rng_state.pth +3 -0
- new_iloko_model/checkpoint-2600/scheduler.pt +3 -0
- new_iloko_model/checkpoint-2600/trainer_state.json +1810 -0
- new_iloko_model/checkpoint-2600/training_args.bin +3 -0
- new_iloko_model/checkpoint-2700/config.json +87 -0
- new_iloko_model/checkpoint-2700/preprocessor_config.json +9 -0
- new_iloko_model/checkpoint-2700/pytorch_model.bin +3 -0
- new_iloko_model/checkpoint-2700/rng_state.pth +3 -0
- new_iloko_model/checkpoint-2700/scheduler.pt +3 -0
- new_iloko_model/checkpoint-2700/trainer_state.json +1879 -0
- new_iloko_model/checkpoint-2700/training_args.bin +3 -0
- new_iloko_model/config.json +87 -0
- new_iloko_model/preprocessor_config.json +9 -0
- new_iloko_model/pytorch_model.bin +3 -0
- new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634895404.4292953/events.out.tfevents.1634895404.e109e789bd2a.91.1 +3 -0
- new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634896681.5475419/events.out.tfevents.1634896681.e109e789bd2a.91.2 +3 -0
- new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634896893.7200775/events.out.tfevents.1634896893.e109e789bd2a.91.3 +3 -0
- new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634897154.445655/events.out.tfevents.1634897154.e109e789bd2a.91.4 +3 -0
- new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/events.out.tfevents.1634895404.e109e789bd2a.91.0 +3 -0
- new_iloko_model/runs/Oct22_10-50-27_8cab4545d814/1634899887.187254/events.out.tfevents.1634899887.8cab4545d814.91.1 +3 -0
- new_iloko_model/runs/Oct22_10-50-27_8cab4545d814/events.out.tfevents.1634899887.8cab4545d814.91.0 +3 -0
- new_iloko_model/runs/Oct22_13-59-07_9090242a067f/1634911877.7595568/events.out.tfevents.1634911877.9090242a067f.90.1 +3 -0
- new_iloko_model/runs/Oct22_13-59-07_9090242a067f/events.out.tfevents.1634911877.9090242a067f.90.0 +3 -0
- new_iloko_model/runs/Oct22_14-44-05_33dae6ed61d6/1634914033.677216/events.out.tfevents.1634914033.33dae6ed61d6.92.1 +3 -0
- new_iloko_model/runs/Oct22_14-44-05_33dae6ed61d6/events.out.tfevents.1634914033.33dae6ed61d6.92.0 +3 -0
- new_iloko_model/special_tokens_map.json +1 -0
- new_iloko_model/tokenizer_config.json +1 -0
- new_iloko_model/training_args.bin +3 -0
- new_iloko_model/vocab.json +1 -0
- optimizer-004.pt +3 -0
- optimizer-005.pt +3 -0
new_iloko_model/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"<s>": 26, "</s>": 27}
|
new_iloko_model/checkpoint-2600/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"codevector_dim": 768,
|
| 12 |
+
"contrastive_logits_temperature": 0.1,
|
| 13 |
+
"conv_bias": true,
|
| 14 |
+
"conv_dim": [
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512
|
| 22 |
+
],
|
| 23 |
+
"conv_kernel": [
|
| 24 |
+
10,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
3,
|
| 28 |
+
3,
|
| 29 |
+
2,
|
| 30 |
+
2
|
| 31 |
+
],
|
| 32 |
+
"conv_stride": [
|
| 33 |
+
5,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2
|
| 40 |
+
],
|
| 41 |
+
"ctc_loss_reduction": "mean",
|
| 42 |
+
"ctc_zero_infinity": false,
|
| 43 |
+
"diversity_loss_weight": 0.1,
|
| 44 |
+
"do_stable_layer_norm": true,
|
| 45 |
+
"eos_token_id": 2,
|
| 46 |
+
"feat_extract_activation": "gelu",
|
| 47 |
+
"feat_extract_dropout": 0.0,
|
| 48 |
+
"feat_extract_norm": "layer",
|
| 49 |
+
"feat_proj_dropout": 0.0,
|
| 50 |
+
"feat_quantizer_dropout": 0.0,
|
| 51 |
+
"final_dropout": 0.0,
|
| 52 |
+
"gradient_checkpointing": false,
|
| 53 |
+
"hidden_act": "gelu",
|
| 54 |
+
"hidden_dropout": 0.1,
|
| 55 |
+
"hidden_size": 1024,
|
| 56 |
+
"initializer_range": 0.02,
|
| 57 |
+
"intermediate_size": 4096,
|
| 58 |
+
"layer_norm_eps": 1e-05,
|
| 59 |
+
"layerdrop": 0.1,
|
| 60 |
+
"mask_channel_length": 10,
|
| 61 |
+
"mask_channel_min_space": 1,
|
| 62 |
+
"mask_channel_other": 0.0,
|
| 63 |
+
"mask_channel_prob": 0.0,
|
| 64 |
+
"mask_channel_selection": "static",
|
| 65 |
+
"mask_feature_length": 10,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_space": 1,
|
| 69 |
+
"mask_time_other": 0.0,
|
| 70 |
+
"mask_time_prob": 0.05,
|
| 71 |
+
"mask_time_selection": "static",
|
| 72 |
+
"model_type": "wav2vec2",
|
| 73 |
+
"num_attention_heads": 16,
|
| 74 |
+
"num_codevector_groups": 2,
|
| 75 |
+
"num_codevectors_per_group": 320,
|
| 76 |
+
"num_conv_pos_embedding_groups": 16,
|
| 77 |
+
"num_conv_pos_embeddings": 128,
|
| 78 |
+
"num_feat_extract_layers": 7,
|
| 79 |
+
"num_hidden_layers": 24,
|
| 80 |
+
"num_negatives": 100,
|
| 81 |
+
"pad_token_id": 25,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"torch_dtype": "float32",
|
| 84 |
+
"transformers_version": "4.11.3",
|
| 85 |
+
"use_weighted_layer_sum": false,
|
| 86 |
+
"vocab_size": 26
|
| 87 |
+
}
|
new_iloko_model/checkpoint-2600/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
new_iloko_model/checkpoint-2600/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73cb2ad85e45850d25e08d226c55ca63c0534cca806d3117303ce54dfe09b73f
|
| 3 |
+
size 1262030257
|
new_iloko_model/checkpoint-2600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a32bc25cfd995c9205f71152ba3995df836d5d2a908ef5137b5412532486e19
|
| 3 |
+
size 14567
|
new_iloko_model/checkpoint-2600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6f9f3e1814f82a481b73f197daecaa840ac28285ea2c4e3c1e8204e2c08f9ed
|
| 3 |
+
size 623
|
new_iloko_model/checkpoint-2600/trainer_state.json
ADDED
|
@@ -0,0 +1,1810 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 28.88888888888889,
|
| 5 |
+
"global_step": 2600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.11,
|
| 12 |
+
"learning_rate": 5.999999999999999e-06,
|
| 13 |
+
"loss": 12.0901,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.22,
|
| 18 |
+
"learning_rate": 1.1999999999999999e-05,
|
| 19 |
+
"loss": 11.5786,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.33,
|
| 24 |
+
"learning_rate": 1.7999999999999997e-05,
|
| 25 |
+
"loss": 12.4996,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.44,
|
| 30 |
+
"learning_rate": 2.3999999999999997e-05,
|
| 31 |
+
"loss": 12.1575,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.56,
|
| 36 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 37 |
+
"loss": 11.6027,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.67,
|
| 42 |
+
"learning_rate": 3.5999999999999994e-05,
|
| 43 |
+
"loss": 9.5766,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.78,
|
| 48 |
+
"learning_rate": 4.2e-05,
|
| 49 |
+
"loss": 7.3941,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.89,
|
| 54 |
+
"learning_rate": 4.7999999999999994e-05,
|
| 55 |
+
"loss": 4.7536,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.0,
|
| 60 |
+
"learning_rate": 5.399999999999999e-05,
|
| 61 |
+
"loss": 3.7347,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.11,
|
| 66 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 67 |
+
"loss": 3.1484,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.11,
|
| 72 |
+
"eval_loss": 2.928152322769165,
|
| 73 |
+
"eval_runtime": 166.6108,
|
| 74 |
+
"eval_samples_per_second": 4.309,
|
| 75 |
+
"eval_steps_per_second": 0.54,
|
| 76 |
+
"eval_wer": 1.0,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.22,
|
| 81 |
+
"learning_rate": 6.599999999999999e-05,
|
| 82 |
+
"loss": 2.8961,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.33,
|
| 87 |
+
"learning_rate": 7.199999999999999e-05,
|
| 88 |
+
"loss": 2.7846,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.44,
|
| 93 |
+
"learning_rate": 7.8e-05,
|
| 94 |
+
"loss": 2.7632,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.56,
|
| 99 |
+
"learning_rate": 8.4e-05,
|
| 100 |
+
"loss": 2.6963,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.67,
|
| 105 |
+
"learning_rate": 8.999999999999999e-05,
|
| 106 |
+
"loss": 2.7072,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.78,
|
| 111 |
+
"learning_rate": 9.599999999999999e-05,
|
| 112 |
+
"loss": 2.6889,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.89,
|
| 117 |
+
"learning_rate": 0.000102,
|
| 118 |
+
"loss": 2.6548,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.0,
|
| 123 |
+
"learning_rate": 0.00010799999999999998,
|
| 124 |
+
"loss": 2.7483,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.11,
|
| 129 |
+
"learning_rate": 0.00011399999999999999,
|
| 130 |
+
"loss": 2.6067,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.22,
|
| 135 |
+
"learning_rate": 0.00011999999999999999,
|
| 136 |
+
"loss": 2.69,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.22,
|
| 141 |
+
"eval_loss": 2.6755106449127197,
|
| 142 |
+
"eval_runtime": 166.1686,
|
| 143 |
+
"eval_samples_per_second": 4.321,
|
| 144 |
+
"eval_steps_per_second": 0.542,
|
| 145 |
+
"eval_wer": 1.0,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.33,
|
| 150 |
+
"learning_rate": 0.00012599999999999997,
|
| 151 |
+
"loss": 2.6435,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.44,
|
| 156 |
+
"learning_rate": 0.00013199999999999998,
|
| 157 |
+
"loss": 2.6126,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.56,
|
| 162 |
+
"learning_rate": 0.000138,
|
| 163 |
+
"loss": 2.6488,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.67,
|
| 168 |
+
"learning_rate": 0.00014399999999999998,
|
| 169 |
+
"loss": 2.6612,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.78,
|
| 174 |
+
"learning_rate": 0.00015,
|
| 175 |
+
"loss": 2.6615,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.89,
|
| 180 |
+
"learning_rate": 0.000156,
|
| 181 |
+
"loss": 2.6005,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.0,
|
| 186 |
+
"learning_rate": 0.000162,
|
| 187 |
+
"loss": 2.7924,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.11,
|
| 192 |
+
"learning_rate": 0.000168,
|
| 193 |
+
"loss": 2.5926,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.22,
|
| 198 |
+
"learning_rate": 0.00017399999999999997,
|
| 199 |
+
"loss": 2.6799,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.33,
|
| 204 |
+
"learning_rate": 0.00017999999999999998,
|
| 205 |
+
"loss": 2.6133,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.33,
|
| 210 |
+
"eval_loss": 2.637242078781128,
|
| 211 |
+
"eval_runtime": 154.1101,
|
| 212 |
+
"eval_samples_per_second": 4.659,
|
| 213 |
+
"eval_steps_per_second": 0.584,
|
| 214 |
+
"eval_wer": 1.0,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.44,
|
| 219 |
+
"learning_rate": 0.000186,
|
| 220 |
+
"loss": 2.6176,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.56,
|
| 225 |
+
"learning_rate": 0.00019199999999999998,
|
| 226 |
+
"loss": 2.6436,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.67,
|
| 231 |
+
"learning_rate": 0.000198,
|
| 232 |
+
"loss": 2.6309,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.78,
|
| 237 |
+
"learning_rate": 0.000204,
|
| 238 |
+
"loss": 2.6305,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.89,
|
| 243 |
+
"learning_rate": 0.00020999999999999998,
|
| 244 |
+
"loss": 2.5714,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.0,
|
| 249 |
+
"learning_rate": 0.00021599999999999996,
|
| 250 |
+
"loss": 2.6415,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.11,
|
| 255 |
+
"learning_rate": 0.00022199999999999998,
|
| 256 |
+
"loss": 2.5131,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 4.22,
|
| 261 |
+
"learning_rate": 0.00022799999999999999,
|
| 262 |
+
"loss": 2.4668,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 4.33,
|
| 267 |
+
"learning_rate": 0.000234,
|
| 268 |
+
"loss": 2.4911,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 4.44,
|
| 273 |
+
"learning_rate": 0.00023999999999999998,
|
| 274 |
+
"loss": 2.2501,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 4.44,
|
| 279 |
+
"eval_loss": 2.055128335952759,
|
| 280 |
+
"eval_runtime": 153.8268,
|
| 281 |
+
"eval_samples_per_second": 4.668,
|
| 282 |
+
"eval_steps_per_second": 0.585,
|
| 283 |
+
"eval_wer": 1.0,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.56,
|
| 288 |
+
"learning_rate": 0.00024599999999999996,
|
| 289 |
+
"loss": 2.0839,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.67,
|
| 294 |
+
"learning_rate": 0.00025199999999999995,
|
| 295 |
+
"loss": 1.8988,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.78,
|
| 300 |
+
"learning_rate": 0.000258,
|
| 301 |
+
"loss": 1.6222,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.89,
|
| 306 |
+
"learning_rate": 0.00026399999999999997,
|
| 307 |
+
"loss": 1.2373,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 5.0,
|
| 312 |
+
"learning_rate": 0.00027,
|
| 313 |
+
"loss": 1.078,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 5.11,
|
| 318 |
+
"learning_rate": 0.000276,
|
| 319 |
+
"loss": 0.866,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.22,
|
| 324 |
+
"learning_rate": 0.00028199999999999997,
|
| 325 |
+
"loss": 0.7399,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.33,
|
| 330 |
+
"learning_rate": 0.00028799999999999995,
|
| 331 |
+
"loss": 0.7751,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 5.44,
|
| 336 |
+
"learning_rate": 0.000294,
|
| 337 |
+
"loss": 0.6117,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 5.56,
|
| 342 |
+
"learning_rate": 0.0003,
|
| 343 |
+
"loss": 0.6642,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 5.56,
|
| 348 |
+
"eval_loss": 0.3055833578109741,
|
| 349 |
+
"eval_runtime": 154.0205,
|
| 350 |
+
"eval_samples_per_second": 4.662,
|
| 351 |
+
"eval_steps_per_second": 0.584,
|
| 352 |
+
"eval_wer": 0.39202622673434856,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 5.67,
|
| 357 |
+
"learning_rate": 0.00029249999999999995,
|
| 358 |
+
"loss": 0.469,
|
| 359 |
+
"step": 510
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 5.78,
|
| 363 |
+
"learning_rate": 0.000285,
|
| 364 |
+
"loss": 0.4959,
|
| 365 |
+
"step": 520
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 5.89,
|
| 369 |
+
"learning_rate": 0.00027749999999999997,
|
| 370 |
+
"loss": 0.4317,
|
| 371 |
+
"step": 530
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 6.0,
|
| 375 |
+
"learning_rate": 0.00027,
|
| 376 |
+
"loss": 0.3656,
|
| 377 |
+
"step": 540
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 6.11,
|
| 381 |
+
"learning_rate": 0.0002625,
|
| 382 |
+
"loss": 0.3608,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 6.22,
|
| 387 |
+
"learning_rate": 0.00025499999999999996,
|
| 388 |
+
"loss": 0.2853,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 6.33,
|
| 393 |
+
"learning_rate": 0.00024749999999999994,
|
| 394 |
+
"loss": 0.3534,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 6.44,
|
| 399 |
+
"learning_rate": 0.00023999999999999998,
|
| 400 |
+
"loss": 0.2887,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 6.56,
|
| 405 |
+
"learning_rate": 0.00023249999999999999,
|
| 406 |
+
"loss": 0.3348,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.67,
|
| 411 |
+
"learning_rate": 0.000225,
|
| 412 |
+
"loss": 0.2755,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.67,
|
| 417 |
+
"eval_loss": 0.1520056277513504,
|
| 418 |
+
"eval_runtime": 153.988,
|
| 419 |
+
"eval_samples_per_second": 4.663,
|
| 420 |
+
"eval_steps_per_second": 0.584,
|
| 421 |
+
"eval_wer": 0.2118231810490694,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.78,
|
| 426 |
+
"learning_rate": 0.00021749999999999997,
|
| 427 |
+
"loss": 0.2901,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.89,
|
| 432 |
+
"learning_rate": 0.00020999999999999998,
|
| 433 |
+
"loss": 0.2485,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 7.0,
|
| 438 |
+
"learning_rate": 0.0002025,
|
| 439 |
+
"loss": 0.2533,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 7.11,
|
| 444 |
+
"learning_rate": 0.000195,
|
| 445 |
+
"loss": 0.2502,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 7.22,
|
| 450 |
+
"learning_rate": 0.00018749999999999998,
|
| 451 |
+
"loss": 0.1834,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 7.33,
|
| 456 |
+
"learning_rate": 0.00017999999999999998,
|
| 457 |
+
"loss": 0.2201,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 7.44,
|
| 462 |
+
"learning_rate": 0.00017249999999999996,
|
| 463 |
+
"loss": 0.2079,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.56,
|
| 468 |
+
"learning_rate": 0.000165,
|
| 469 |
+
"loss": 0.226,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 7.67,
|
| 474 |
+
"learning_rate": 0.00015749999999999998,
|
| 475 |
+
"loss": 0.1831,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.78,
|
| 480 |
+
"learning_rate": 0.00015,
|
| 481 |
+
"loss": 0.1749,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 7.78,
|
| 486 |
+
"eval_loss": 0.10200744867324829,
|
| 487 |
+
"eval_runtime": 153.705,
|
| 488 |
+
"eval_samples_per_second": 4.671,
|
| 489 |
+
"eval_steps_per_second": 0.586,
|
| 490 |
+
"eval_wer": 0.1515439932318105,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.89,
|
| 495 |
+
"learning_rate": 0.0001425,
|
| 496 |
+
"loss": 0.163,
|
| 497 |
+
"step": 710
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 8.0,
|
| 501 |
+
"learning_rate": 0.000135,
|
| 502 |
+
"loss": 0.1567,
|
| 503 |
+
"step": 720
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 8.11,
|
| 507 |
+
"learning_rate": 0.00012749999999999998,
|
| 508 |
+
"loss": 0.1884,
|
| 509 |
+
"step": 730
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 8.22,
|
| 513 |
+
"learning_rate": 0.00011999999999999999,
|
| 514 |
+
"loss": 0.1245,
|
| 515 |
+
"step": 740
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.33,
|
| 519 |
+
"learning_rate": 0.0001125,
|
| 520 |
+
"loss": 0.145,
|
| 521 |
+
"step": 750
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.44,
|
| 525 |
+
"learning_rate": 0.00010499999999999999,
|
| 526 |
+
"loss": 0.1196,
|
| 527 |
+
"step": 760
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 8.56,
|
| 531 |
+
"learning_rate": 9.75e-05,
|
| 532 |
+
"loss": 0.1875,
|
| 533 |
+
"step": 770
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 8.67,
|
| 537 |
+
"learning_rate": 8.999999999999999e-05,
|
| 538 |
+
"loss": 0.1603,
|
| 539 |
+
"step": 780
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 8.78,
|
| 543 |
+
"learning_rate": 8.25e-05,
|
| 544 |
+
"loss": 0.1459,
|
| 545 |
+
"step": 790
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 8.89,
|
| 549 |
+
"learning_rate": 7.5e-05,
|
| 550 |
+
"loss": 0.1404,
|
| 551 |
+
"step": 800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 8.89,
|
| 555 |
+
"eval_loss": 0.08183874934911728,
|
| 556 |
+
"eval_runtime": 154.0029,
|
| 557 |
+
"eval_samples_per_second": 4.662,
|
| 558 |
+
"eval_steps_per_second": 0.584,
|
| 559 |
+
"eval_wer": 0.13293147208121828,
|
| 560 |
+
"step": 800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 9.0,
|
| 564 |
+
"learning_rate": 6.75e-05,
|
| 565 |
+
"loss": 0.11,
|
| 566 |
+
"step": 810
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.11,
|
| 570 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 571 |
+
"loss": 0.1284,
|
| 572 |
+
"step": 820
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 9.22,
|
| 576 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 577 |
+
"loss": 0.1055,
|
| 578 |
+
"step": 830
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.33,
|
| 582 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 583 |
+
"loss": 0.1459,
|
| 584 |
+
"step": 840
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 9.44,
|
| 588 |
+
"learning_rate": 3.75e-05,
|
| 589 |
+
"loss": 0.1028,
|
| 590 |
+
"step": 850
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 9.56,
|
| 594 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 595 |
+
"loss": 0.1635,
|
| 596 |
+
"step": 860
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 9.67,
|
| 600 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 601 |
+
"loss": 0.1315,
|
| 602 |
+
"step": 870
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 9.78,
|
| 606 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 607 |
+
"loss": 0.1215,
|
| 608 |
+
"step": 880
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.89,
|
| 612 |
+
"learning_rate": 7.499999999999999e-06,
|
| 613 |
+
"loss": 0.157,
|
| 614 |
+
"step": 890
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.0,
|
| 618 |
+
"learning_rate": 0.0,
|
| 619 |
+
"loss": 0.1064,
|
| 620 |
+
"step": 900
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 10.0,
|
| 624 |
+
"eval_loss": 0.07435603439807892,
|
| 625 |
+
"eval_runtime": 154.0414,
|
| 626 |
+
"eval_samples_per_second": 4.661,
|
| 627 |
+
"eval_steps_per_second": 0.584,
|
| 628 |
+
"eval_wer": 0.12531725888324874,
|
| 629 |
+
"step": 900
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 10.11,
|
| 633 |
+
"learning_rate": 0.00024409090909090905,
|
| 634 |
+
"loss": 6.6354,
|
| 635 |
+
"step": 910
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 10.22,
|
| 639 |
+
"learning_rate": 0.0002427272727272727,
|
| 640 |
+
"loss": 3.0269,
|
| 641 |
+
"step": 920
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 10.33,
|
| 645 |
+
"learning_rate": 0.00024136363636363635,
|
| 646 |
+
"loss": 2.7006,
|
| 647 |
+
"step": 930
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 10.44,
|
| 651 |
+
"learning_rate": 0.00023999999999999998,
|
| 652 |
+
"loss": 2.6422,
|
| 653 |
+
"step": 940
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 10.56,
|
| 657 |
+
"learning_rate": 0.0002386363636363636,
|
| 658 |
+
"loss": 2.6196,
|
| 659 |
+
"step": 950
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 10.67,
|
| 663 |
+
"learning_rate": 0.00023727272727272724,
|
| 664 |
+
"loss": 2.5681,
|
| 665 |
+
"step": 960
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 10.78,
|
| 669 |
+
"learning_rate": 0.0002359090909090909,
|
| 670 |
+
"loss": 2.4665,
|
| 671 |
+
"step": 970
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 10.89,
|
| 675 |
+
"learning_rate": 0.00023454545454545454,
|
| 676 |
+
"loss": 2.0804,
|
| 677 |
+
"step": 980
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 11.0,
|
| 681 |
+
"learning_rate": 0.00023318181818181814,
|
| 682 |
+
"loss": 1.3389,
|
| 683 |
+
"step": 990
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 11.11,
|
| 687 |
+
"learning_rate": 0.0002318181818181818,
|
| 688 |
+
"loss": 0.5882,
|
| 689 |
+
"step": 1000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 11.11,
|
| 693 |
+
"eval_loss": 0.21222124993801117,
|
| 694 |
+
"eval_runtime": 150.6754,
|
| 695 |
+
"eval_samples_per_second": 4.765,
|
| 696 |
+
"eval_steps_per_second": 0.597,
|
| 697 |
+
"eval_wer": 0.2306472081218274,
|
| 698 |
+
"step": 1000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 11.22,
|
| 702 |
+
"learning_rate": 0.00023045454545454544,
|
| 703 |
+
"loss": 0.2554,
|
| 704 |
+
"step": 1010
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 11.33,
|
| 708 |
+
"learning_rate": 0.00022909090909090907,
|
| 709 |
+
"loss": 0.2722,
|
| 710 |
+
"step": 1020
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 11.44,
|
| 714 |
+
"learning_rate": 0.0002277272727272727,
|
| 715 |
+
"loss": 0.2067,
|
| 716 |
+
"step": 1030
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 11.56,
|
| 720 |
+
"learning_rate": 0.00022636363636363633,
|
| 721 |
+
"loss": 0.2437,
|
| 722 |
+
"step": 1040
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 11.67,
|
| 726 |
+
"learning_rate": 0.000225,
|
| 727 |
+
"loss": 0.1675,
|
| 728 |
+
"step": 1050
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 11.78,
|
| 732 |
+
"learning_rate": 0.00022363636363636363,
|
| 733 |
+
"loss": 0.181,
|
| 734 |
+
"step": 1060
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 11.89,
|
| 738 |
+
"learning_rate": 0.00022227272727272726,
|
| 739 |
+
"loss": 0.1562,
|
| 740 |
+
"step": 1070
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 12.0,
|
| 744 |
+
"learning_rate": 0.0002209090909090909,
|
| 745 |
+
"loss": 0.1805,
|
| 746 |
+
"step": 1080
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 12.11,
|
| 750 |
+
"learning_rate": 0.00021954545454545452,
|
| 751 |
+
"loss": 0.2045,
|
| 752 |
+
"step": 1090
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 12.22,
|
| 756 |
+
"learning_rate": 0.00021818181818181816,
|
| 757 |
+
"loss": 0.1813,
|
| 758 |
+
"step": 1100
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 12.22,
|
| 762 |
+
"eval_loss": 0.0888812392950058,
|
| 763 |
+
"eval_runtime": 150.0493,
|
| 764 |
+
"eval_samples_per_second": 4.785,
|
| 765 |
+
"eval_steps_per_second": 0.6,
|
| 766 |
+
"eval_wer": 0.14075719120135363,
|
| 767 |
+
"step": 1100
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 12.33,
|
| 771 |
+
"learning_rate": 0.00021681818181818182,
|
| 772 |
+
"loss": 0.2353,
|
| 773 |
+
"step": 1110
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 12.44,
|
| 777 |
+
"learning_rate": 0.00021545454545454542,
|
| 778 |
+
"loss": 0.1769,
|
| 779 |
+
"step": 1120
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 12.56,
|
| 783 |
+
"learning_rate": 0.00021409090909090906,
|
| 784 |
+
"loss": 0.1727,
|
| 785 |
+
"step": 1130
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 12.67,
|
| 789 |
+
"learning_rate": 0.00021272727272727272,
|
| 790 |
+
"loss": 0.1213,
|
| 791 |
+
"step": 1140
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 12.78,
|
| 795 |
+
"learning_rate": 0.00021136363636363635,
|
| 796 |
+
"loss": 0.1829,
|
| 797 |
+
"step": 1150
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 12.89,
|
| 801 |
+
"learning_rate": 0.00020999999999999998,
|
| 802 |
+
"loss": 0.1414,
|
| 803 |
+
"step": 1160
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 13.0,
|
| 807 |
+
"learning_rate": 0.00020863636363636361,
|
| 808 |
+
"loss": 0.132,
|
| 809 |
+
"step": 1170
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 13.11,
|
| 813 |
+
"learning_rate": 0.00020727272727272725,
|
| 814 |
+
"loss": 0.1344,
|
| 815 |
+
"step": 1180
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 13.22,
|
| 819 |
+
"learning_rate": 0.0002059090909090909,
|
| 820 |
+
"loss": 0.1064,
|
| 821 |
+
"step": 1190
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 13.33,
|
| 825 |
+
"learning_rate": 0.0002045454545454545,
|
| 826 |
+
"loss": 0.1166,
|
| 827 |
+
"step": 1200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.33,
|
| 831 |
+
"eval_loss": 0.06592676788568497,
|
| 832 |
+
"eval_runtime": 149.9746,
|
| 833 |
+
"eval_samples_per_second": 4.787,
|
| 834 |
+
"eval_steps_per_second": 0.6,
|
| 835 |
+
"eval_wer": 0.11928934010152284,
|
| 836 |
+
"step": 1200
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 13.44,
|
| 840 |
+
"learning_rate": 0.00020318181818181815,
|
| 841 |
+
"loss": 0.1118,
|
| 842 |
+
"step": 1210
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 13.56,
|
| 846 |
+
"learning_rate": 0.0002018181818181818,
|
| 847 |
+
"loss": 0.1629,
|
| 848 |
+
"step": 1220
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 13.67,
|
| 852 |
+
"learning_rate": 0.00020045454545454544,
|
| 853 |
+
"loss": 0.1198,
|
| 854 |
+
"step": 1230
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 13.78,
|
| 858 |
+
"learning_rate": 0.0001990909090909091,
|
| 859 |
+
"loss": 0.1352,
|
| 860 |
+
"step": 1240
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 13.89,
|
| 864 |
+
"learning_rate": 0.0001977272727272727,
|
| 865 |
+
"loss": 0.132,
|
| 866 |
+
"step": 1250
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 14.0,
|
| 870 |
+
"learning_rate": 0.00019636363636363634,
|
| 871 |
+
"loss": 0.1223,
|
| 872 |
+
"step": 1260
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 14.11,
|
| 876 |
+
"learning_rate": 0.000195,
|
| 877 |
+
"loss": 0.1334,
|
| 878 |
+
"step": 1270
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 14.22,
|
| 882 |
+
"learning_rate": 0.00019363636363636363,
|
| 883 |
+
"loss": 0.1102,
|
| 884 |
+
"step": 1280
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.33,
|
| 888 |
+
"learning_rate": 0.00019227272727272723,
|
| 889 |
+
"loss": 0.1369,
|
| 890 |
+
"step": 1290
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 14.44,
|
| 894 |
+
"learning_rate": 0.0001909090909090909,
|
| 895 |
+
"loss": 0.0871,
|
| 896 |
+
"step": 1300
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 14.44,
|
| 900 |
+
"eval_loss": 0.05411907657980919,
|
| 901 |
+
"eval_runtime": 149.6406,
|
| 902 |
+
"eval_samples_per_second": 4.798,
|
| 903 |
+
"eval_steps_per_second": 0.601,
|
| 904 |
+
"eval_wer": 0.11072335025380711,
|
| 905 |
+
"step": 1300
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 14.56,
|
| 909 |
+
"learning_rate": 0.00018954545454545453,
|
| 910 |
+
"loss": 0.1193,
|
| 911 |
+
"step": 1310
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 14.67,
|
| 915 |
+
"learning_rate": 0.0001881818181818182,
|
| 916 |
+
"loss": 0.0963,
|
| 917 |
+
"step": 1320
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 14.78,
|
| 921 |
+
"learning_rate": 0.0001868181818181818,
|
| 922 |
+
"loss": 0.1135,
|
| 923 |
+
"step": 1330
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 14.89,
|
| 927 |
+
"learning_rate": 0.00018545454545454543,
|
| 928 |
+
"loss": 0.0891,
|
| 929 |
+
"step": 1340
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 15.0,
|
| 933 |
+
"learning_rate": 0.00018409090909090909,
|
| 934 |
+
"loss": 0.0724,
|
| 935 |
+
"step": 1350
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 15.11,
|
| 939 |
+
"learning_rate": 0.00018272727272727272,
|
| 940 |
+
"loss": 0.1289,
|
| 941 |
+
"step": 1360
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 15.22,
|
| 945 |
+
"learning_rate": 0.00018136363636363632,
|
| 946 |
+
"loss": 0.0639,
|
| 947 |
+
"step": 1370
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 15.33,
|
| 951 |
+
"learning_rate": 0.00017999999999999998,
|
| 952 |
+
"loss": 0.0868,
|
| 953 |
+
"step": 1380
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 15.44,
|
| 957 |
+
"learning_rate": 0.00017863636363636362,
|
| 958 |
+
"loss": 0.0749,
|
| 959 |
+
"step": 1390
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 15.56,
|
| 963 |
+
"learning_rate": 0.00017727272727272728,
|
| 964 |
+
"loss": 0.091,
|
| 965 |
+
"step": 1400
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 15.56,
|
| 969 |
+
"eval_loss": 0.046893950551748276,
|
| 970 |
+
"eval_runtime": 149.6721,
|
| 971 |
+
"eval_samples_per_second": 4.797,
|
| 972 |
+
"eval_steps_per_second": 0.601,
|
| 973 |
+
"eval_wer": 0.10046531302876481,
|
| 974 |
+
"step": 1400
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 15.67,
|
| 978 |
+
"learning_rate": 0.00017590909090909088,
|
| 979 |
+
"loss": 0.0819,
|
| 980 |
+
"step": 1410
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 15.78,
|
| 984 |
+
"learning_rate": 0.00017454545454545452,
|
| 985 |
+
"loss": 0.0966,
|
| 986 |
+
"step": 1420
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 15.89,
|
| 990 |
+
"learning_rate": 0.00017318181818181818,
|
| 991 |
+
"loss": 0.0823,
|
| 992 |
+
"step": 1430
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 16.0,
|
| 996 |
+
"learning_rate": 0.0001718181818181818,
|
| 997 |
+
"loss": 0.0958,
|
| 998 |
+
"step": 1440
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 16.11,
|
| 1002 |
+
"learning_rate": 0.00017045454545454547,
|
| 1003 |
+
"loss": 0.1276,
|
| 1004 |
+
"step": 1450
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 16.22,
|
| 1008 |
+
"learning_rate": 0.00016909090909090907,
|
| 1009 |
+
"loss": 0.0855,
|
| 1010 |
+
"step": 1460
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 16.33,
|
| 1014 |
+
"learning_rate": 0.0001677272727272727,
|
| 1015 |
+
"loss": 0.1032,
|
| 1016 |
+
"step": 1470
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 16.44,
|
| 1020 |
+
"learning_rate": 0.00016636363636363637,
|
| 1021 |
+
"loss": 0.0683,
|
| 1022 |
+
"step": 1480
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 16.56,
|
| 1026 |
+
"learning_rate": 0.000165,
|
| 1027 |
+
"loss": 0.0988,
|
| 1028 |
+
"step": 1490
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 16.67,
|
| 1032 |
+
"learning_rate": 0.0001636363636363636,
|
| 1033 |
+
"loss": 0.0551,
|
| 1034 |
+
"step": 1500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 16.67,
|
| 1038 |
+
"eval_loss": 0.043081216514110565,
|
| 1039 |
+
"eval_runtime": 150.166,
|
| 1040 |
+
"eval_samples_per_second": 4.781,
|
| 1041 |
+
"eval_steps_per_second": 0.599,
|
| 1042 |
+
"eval_wer": 0.10268612521150593,
|
| 1043 |
+
"step": 1500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 16.78,
|
| 1047 |
+
"learning_rate": 0.00016227272727272726,
|
| 1048 |
+
"loss": 0.067,
|
| 1049 |
+
"step": 1510
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 16.89,
|
| 1053 |
+
"learning_rate": 0.0001609090909090909,
|
| 1054 |
+
"loss": 0.0667,
|
| 1055 |
+
"step": 1520
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 17.0,
|
| 1059 |
+
"learning_rate": 0.00015954545454545453,
|
| 1060 |
+
"loss": 0.0639,
|
| 1061 |
+
"step": 1530
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 17.11,
|
| 1065 |
+
"learning_rate": 0.00015818181818181816,
|
| 1066 |
+
"loss": 0.0994,
|
| 1067 |
+
"step": 1540
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 17.22,
|
| 1071 |
+
"learning_rate": 0.0001568181818181818,
|
| 1072 |
+
"loss": 0.0515,
|
| 1073 |
+
"step": 1550
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 17.33,
|
| 1077 |
+
"learning_rate": 0.00015545454545454546,
|
| 1078 |
+
"loss": 0.0911,
|
| 1079 |
+
"step": 1560
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 17.44,
|
| 1083 |
+
"learning_rate": 0.0001540909090909091,
|
| 1084 |
+
"loss": 0.0771,
|
| 1085 |
+
"step": 1570
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 17.56,
|
| 1089 |
+
"learning_rate": 0.0001527272727272727,
|
| 1090 |
+
"loss": 0.0706,
|
| 1091 |
+
"step": 1580
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 17.67,
|
| 1095 |
+
"learning_rate": 0.00015136363636363635,
|
| 1096 |
+
"loss": 0.0716,
|
| 1097 |
+
"step": 1590
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 17.78,
|
| 1101 |
+
"learning_rate": 0.00015,
|
| 1102 |
+
"loss": 0.0679,
|
| 1103 |
+
"step": 1600
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 17.78,
|
| 1107 |
+
"eval_loss": 0.043587520718574524,
|
| 1108 |
+
"eval_runtime": 149.9856,
|
| 1109 |
+
"eval_samples_per_second": 4.787,
|
| 1110 |
+
"eval_steps_per_second": 0.6,
|
| 1111 |
+
"eval_wer": 0.09898477157360407,
|
| 1112 |
+
"step": 1600
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 17.89,
|
| 1116 |
+
"learning_rate": 0.00014863636363636362,
|
| 1117 |
+
"loss": 0.0822,
|
| 1118 |
+
"step": 1610
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 18.0,
|
| 1122 |
+
"learning_rate": 0.00014727272727272725,
|
| 1123 |
+
"loss": 0.0655,
|
| 1124 |
+
"step": 1620
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 18.11,
|
| 1128 |
+
"learning_rate": 0.0001459090909090909,
|
| 1129 |
+
"loss": 0.1081,
|
| 1130 |
+
"step": 1630
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 18.22,
|
| 1134 |
+
"learning_rate": 0.00014454545454545452,
|
| 1135 |
+
"loss": 0.0488,
|
| 1136 |
+
"step": 1640
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 18.33,
|
| 1140 |
+
"learning_rate": 0.00014318181818181818,
|
| 1141 |
+
"loss": 0.0739,
|
| 1142 |
+
"step": 1650
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 18.44,
|
| 1146 |
+
"learning_rate": 0.0001418181818181818,
|
| 1147 |
+
"loss": 0.0711,
|
| 1148 |
+
"step": 1660
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 18.56,
|
| 1152 |
+
"learning_rate": 0.00014045454545454544,
|
| 1153 |
+
"loss": 0.0893,
|
| 1154 |
+
"step": 1670
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 18.67,
|
| 1158 |
+
"learning_rate": 0.00013909090909090908,
|
| 1159 |
+
"loss": 0.0419,
|
| 1160 |
+
"step": 1680
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 18.78,
|
| 1164 |
+
"learning_rate": 0.0001377272727272727,
|
| 1165 |
+
"loss": 0.0625,
|
| 1166 |
+
"step": 1690
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 18.89,
|
| 1170 |
+
"learning_rate": 0.00013636363636363634,
|
| 1171 |
+
"loss": 0.0497,
|
| 1172 |
+
"step": 1700
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 18.89,
|
| 1176 |
+
"eval_loss": 0.030962202697992325,
|
| 1177 |
+
"eval_runtime": 150.4238,
|
| 1178 |
+
"eval_samples_per_second": 4.773,
|
| 1179 |
+
"eval_steps_per_second": 0.598,
|
| 1180 |
+
"eval_wer": 0.09443739424703891,
|
| 1181 |
+
"step": 1700
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 19.0,
|
| 1185 |
+
"learning_rate": 0.000135,
|
| 1186 |
+
"loss": 0.0405,
|
| 1187 |
+
"step": 1710
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 19.11,
|
| 1191 |
+
"learning_rate": 0.0001336363636363636,
|
| 1192 |
+
"loss": 0.0697,
|
| 1193 |
+
"step": 1720
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 19.22,
|
| 1197 |
+
"learning_rate": 0.00013227272727272727,
|
| 1198 |
+
"loss": 0.0434,
|
| 1199 |
+
"step": 1730
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 19.33,
|
| 1203 |
+
"learning_rate": 0.0001309090909090909,
|
| 1204 |
+
"loss": 0.0696,
|
| 1205 |
+
"step": 1740
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 19.44,
|
| 1209 |
+
"learning_rate": 0.00012954545454545453,
|
| 1210 |
+
"loss": 0.0538,
|
| 1211 |
+
"step": 1750
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 19.56,
|
| 1215 |
+
"learning_rate": 0.00012818181818181817,
|
| 1216 |
+
"loss": 0.0951,
|
| 1217 |
+
"step": 1760
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 19.67,
|
| 1221 |
+
"learning_rate": 0.0001268181818181818,
|
| 1222 |
+
"loss": 0.0545,
|
| 1223 |
+
"step": 1770
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 19.78,
|
| 1227 |
+
"learning_rate": 0.00012545454545454543,
|
| 1228 |
+
"loss": 0.0622,
|
| 1229 |
+
"step": 1780
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 19.89,
|
| 1233 |
+
"learning_rate": 0.0001240909090909091,
|
| 1234 |
+
"loss": 0.0884,
|
| 1235 |
+
"step": 1790
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 20.0,
|
| 1239 |
+
"learning_rate": 0.00012272727272727272,
|
| 1240 |
+
"loss": 0.0669,
|
| 1241 |
+
"step": 1800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 20.0,
|
| 1245 |
+
"eval_loss": 0.026656057685613632,
|
| 1246 |
+
"eval_runtime": 150.2929,
|
| 1247 |
+
"eval_samples_per_second": 4.777,
|
| 1248 |
+
"eval_steps_per_second": 0.599,
|
| 1249 |
+
"eval_wer": 0.0942258883248731,
|
| 1250 |
+
"step": 1800
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 20.11,
|
| 1254 |
+
"learning_rate": 0.00012136363636363636,
|
| 1255 |
+
"loss": 0.066,
|
| 1256 |
+
"step": 1810
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 20.22,
|
| 1260 |
+
"learning_rate": 0.00011999999999999999,
|
| 1261 |
+
"loss": 0.0337,
|
| 1262 |
+
"step": 1820
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 20.33,
|
| 1266 |
+
"learning_rate": 0.00011863636363636362,
|
| 1267 |
+
"loss": 0.0766,
|
| 1268 |
+
"step": 1830
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 20.44,
|
| 1272 |
+
"learning_rate": 0.00011727272727272727,
|
| 1273 |
+
"loss": 0.0625,
|
| 1274 |
+
"step": 1840
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 20.56,
|
| 1278 |
+
"learning_rate": 0.0001159090909090909,
|
| 1279 |
+
"loss": 0.0539,
|
| 1280 |
+
"step": 1850
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 20.67,
|
| 1284 |
+
"learning_rate": 0.00011454545454545453,
|
| 1285 |
+
"loss": 0.0605,
|
| 1286 |
+
"step": 1860
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 20.78,
|
| 1290 |
+
"learning_rate": 0.00011318181818181817,
|
| 1291 |
+
"loss": 0.0465,
|
| 1292 |
+
"step": 1870
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 20.89,
|
| 1296 |
+
"learning_rate": 0.00011181818181818181,
|
| 1297 |
+
"loss": 0.0642,
|
| 1298 |
+
"step": 1880
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 21.0,
|
| 1302 |
+
"learning_rate": 0.00011045454545454545,
|
| 1303 |
+
"loss": 0.047,
|
| 1304 |
+
"step": 1890
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 21.11,
|
| 1308 |
+
"learning_rate": 0.00010909090909090908,
|
| 1309 |
+
"loss": 0.0532,
|
| 1310 |
+
"step": 1900
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 21.11,
|
| 1314 |
+
"eval_loss": 0.02508886530995369,
|
| 1315 |
+
"eval_runtime": 150.8935,
|
| 1316 |
+
"eval_samples_per_second": 4.758,
|
| 1317 |
+
"eval_steps_per_second": 0.596,
|
| 1318 |
+
"eval_wer": 0.09105329949238579,
|
| 1319 |
+
"step": 1900
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 21.22,
|
| 1323 |
+
"learning_rate": 0.00010772727272727271,
|
| 1324 |
+
"loss": 0.0245,
|
| 1325 |
+
"step": 1910
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 21.33,
|
| 1329 |
+
"learning_rate": 0.00010636363636363636,
|
| 1330 |
+
"loss": 0.0614,
|
| 1331 |
+
"step": 1920
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 21.44,
|
| 1335 |
+
"learning_rate": 0.00010499999999999999,
|
| 1336 |
+
"loss": 0.0333,
|
| 1337 |
+
"step": 1930
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 21.56,
|
| 1341 |
+
"learning_rate": 0.00010363636363636362,
|
| 1342 |
+
"loss": 0.0463,
|
| 1343 |
+
"step": 1940
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 21.67,
|
| 1347 |
+
"learning_rate": 0.00010227272727272726,
|
| 1348 |
+
"loss": 0.0509,
|
| 1349 |
+
"step": 1950
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 21.78,
|
| 1353 |
+
"learning_rate": 0.0001009090909090909,
|
| 1354 |
+
"loss": 0.0522,
|
| 1355 |
+
"step": 1960
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 21.89,
|
| 1359 |
+
"learning_rate": 9.954545454545455e-05,
|
| 1360 |
+
"loss": 0.032,
|
| 1361 |
+
"step": 1970
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 22.0,
|
| 1365 |
+
"learning_rate": 9.818181818181817e-05,
|
| 1366 |
+
"loss": 0.0491,
|
| 1367 |
+
"step": 1980
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 22.11,
|
| 1371 |
+
"learning_rate": 9.681818181818181e-05,
|
| 1372 |
+
"loss": 0.0603,
|
| 1373 |
+
"step": 1990
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 22.22,
|
| 1377 |
+
"learning_rate": 9.545454545454545e-05,
|
| 1378 |
+
"loss": 0.0269,
|
| 1379 |
+
"step": 2000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 22.22,
|
| 1383 |
+
"eval_loss": 0.02213282324373722,
|
| 1384 |
+
"eval_runtime": 150.9545,
|
| 1385 |
+
"eval_samples_per_second": 4.756,
|
| 1386 |
+
"eval_steps_per_second": 0.596,
|
| 1387 |
+
"eval_wer": 0.09020727580372251,
|
| 1388 |
+
"step": 2000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 22.33,
|
| 1392 |
+
"learning_rate": 9.40909090909091e-05,
|
| 1393 |
+
"loss": 0.0544,
|
| 1394 |
+
"step": 2010
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 22.44,
|
| 1398 |
+
"learning_rate": 9.272727272727271e-05,
|
| 1399 |
+
"loss": 0.0331,
|
| 1400 |
+
"step": 2020
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 22.56,
|
| 1404 |
+
"learning_rate": 9.136363636363636e-05,
|
| 1405 |
+
"loss": 0.0502,
|
| 1406 |
+
"step": 2030
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 22.67,
|
| 1410 |
+
"learning_rate": 8.999999999999999e-05,
|
| 1411 |
+
"loss": 0.0493,
|
| 1412 |
+
"step": 2040
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 22.78,
|
| 1416 |
+
"learning_rate": 8.863636363636364e-05,
|
| 1417 |
+
"loss": 0.0579,
|
| 1418 |
+
"step": 2050
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 22.89,
|
| 1422 |
+
"learning_rate": 8.727272727272726e-05,
|
| 1423 |
+
"loss": 0.0392,
|
| 1424 |
+
"step": 2060
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 23.0,
|
| 1428 |
+
"learning_rate": 8.59090909090909e-05,
|
| 1429 |
+
"loss": 0.0423,
|
| 1430 |
+
"step": 2070
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 23.11,
|
| 1434 |
+
"learning_rate": 8.454545454545454e-05,
|
| 1435 |
+
"loss": 0.0696,
|
| 1436 |
+
"step": 2080
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 23.22,
|
| 1440 |
+
"learning_rate": 8.318181818181818e-05,
|
| 1441 |
+
"loss": 0.0304,
|
| 1442 |
+
"step": 2090
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 23.33,
|
| 1446 |
+
"learning_rate": 8.18181818181818e-05,
|
| 1447 |
+
"loss": 0.0686,
|
| 1448 |
+
"step": 2100
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 23.33,
|
| 1452 |
+
"eval_loss": 0.020356198772788048,
|
| 1453 |
+
"eval_runtime": 151.0382,
|
| 1454 |
+
"eval_samples_per_second": 4.754,
|
| 1455 |
+
"eval_steps_per_second": 0.596,
|
| 1456 |
+
"eval_wer": 0.08745769881556684,
|
| 1457 |
+
"step": 2100
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 23.44,
|
| 1461 |
+
"learning_rate": 8.045454545454545e-05,
|
| 1462 |
+
"loss": 0.0523,
|
| 1463 |
+
"step": 2110
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 23.56,
|
| 1467 |
+
"learning_rate": 7.909090909090908e-05,
|
| 1468 |
+
"loss": 0.0583,
|
| 1469 |
+
"step": 2120
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 23.67,
|
| 1473 |
+
"learning_rate": 7.772727272727273e-05,
|
| 1474 |
+
"loss": 0.0495,
|
| 1475 |
+
"step": 2130
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 23.78,
|
| 1479 |
+
"learning_rate": 7.636363636363635e-05,
|
| 1480 |
+
"loss": 0.0476,
|
| 1481 |
+
"step": 2140
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 23.89,
|
| 1485 |
+
"learning_rate": 7.5e-05,
|
| 1486 |
+
"loss": 0.0321,
|
| 1487 |
+
"step": 2150
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 24.0,
|
| 1491 |
+
"learning_rate": 7.363636363636363e-05,
|
| 1492 |
+
"loss": 0.0393,
|
| 1493 |
+
"step": 2160
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 24.11,
|
| 1497 |
+
"learning_rate": 7.227272727272726e-05,
|
| 1498 |
+
"loss": 0.0425,
|
| 1499 |
+
"step": 2170
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 24.22,
|
| 1503 |
+
"learning_rate": 7.09090909090909e-05,
|
| 1504 |
+
"loss": 0.064,
|
| 1505 |
+
"step": 2180
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 24.33,
|
| 1509 |
+
"learning_rate": 6.954545454545454e-05,
|
| 1510 |
+
"loss": 0.048,
|
| 1511 |
+
"step": 2190
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 24.44,
|
| 1515 |
+
"learning_rate": 6.818181818181817e-05,
|
| 1516 |
+
"loss": 0.0242,
|
| 1517 |
+
"step": 2200
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 24.44,
|
| 1521 |
+
"eval_loss": 0.018909653648734093,
|
| 1522 |
+
"eval_runtime": 151.2527,
|
| 1523 |
+
"eval_samples_per_second": 4.747,
|
| 1524 |
+
"eval_steps_per_second": 0.595,
|
| 1525 |
+
"eval_wer": 0.08724619289340102,
|
| 1526 |
+
"step": 2200
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 24.56,
|
| 1530 |
+
"learning_rate": 6.68181818181818e-05,
|
| 1531 |
+
"loss": 0.0481,
|
| 1532 |
+
"step": 2210
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 24.67,
|
| 1536 |
+
"learning_rate": 6.545454545454545e-05,
|
| 1537 |
+
"loss": 0.0442,
|
| 1538 |
+
"step": 2220
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 24.78,
|
| 1542 |
+
"learning_rate": 6.409090909090908e-05,
|
| 1543 |
+
"loss": 0.05,
|
| 1544 |
+
"step": 2230
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 24.89,
|
| 1548 |
+
"learning_rate": 6.272727272727272e-05,
|
| 1549 |
+
"loss": 0.0474,
|
| 1550 |
+
"step": 2240
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 25.0,
|
| 1554 |
+
"learning_rate": 6.136363636363636e-05,
|
| 1555 |
+
"loss": 0.0391,
|
| 1556 |
+
"step": 2250
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 25.11,
|
| 1560 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 1561 |
+
"loss": 0.068,
|
| 1562 |
+
"step": 2260
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 25.22,
|
| 1566 |
+
"learning_rate": 5.8636363636363634e-05,
|
| 1567 |
+
"loss": 0.0189,
|
| 1568 |
+
"step": 2270
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 25.33,
|
| 1572 |
+
"learning_rate": 5.727272727272727e-05,
|
| 1573 |
+
"loss": 0.0396,
|
| 1574 |
+
"step": 2280
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 25.44,
|
| 1578 |
+
"learning_rate": 5.590909090909091e-05,
|
| 1579 |
+
"loss": 0.0317,
|
| 1580 |
+
"step": 2290
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 25.56,
|
| 1584 |
+
"learning_rate": 5.454545454545454e-05,
|
| 1585 |
+
"loss": 0.0454,
|
| 1586 |
+
"step": 2300
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 25.56,
|
| 1590 |
+
"eval_loss": 0.01684204488992691,
|
| 1591 |
+
"eval_runtime": 150.5397,
|
| 1592 |
+
"eval_samples_per_second": 4.77,
|
| 1593 |
+
"eval_steps_per_second": 0.598,
|
| 1594 |
+
"eval_wer": 0.08513113367174281,
|
| 1595 |
+
"step": 2300
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 25.67,
|
| 1599 |
+
"learning_rate": 5.318181818181818e-05,
|
| 1600 |
+
"loss": 0.0317,
|
| 1601 |
+
"step": 2310
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 25.78,
|
| 1605 |
+
"learning_rate": 5.181818181818181e-05,
|
| 1606 |
+
"loss": 0.0315,
|
| 1607 |
+
"step": 2320
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 25.89,
|
| 1611 |
+
"learning_rate": 5.045454545454545e-05,
|
| 1612 |
+
"loss": 0.0302,
|
| 1613 |
+
"step": 2330
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 26.0,
|
| 1617 |
+
"learning_rate": 4.9090909090909084e-05,
|
| 1618 |
+
"loss": 0.0359,
|
| 1619 |
+
"step": 2340
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 26.11,
|
| 1623 |
+
"learning_rate": 4.7727272727272724e-05,
|
| 1624 |
+
"loss": 0.0433,
|
| 1625 |
+
"step": 2350
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 26.22,
|
| 1629 |
+
"learning_rate": 4.6363636363636356e-05,
|
| 1630 |
+
"loss": 0.0222,
|
| 1631 |
+
"step": 2360
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 26.33,
|
| 1635 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 1636 |
+
"loss": 0.0474,
|
| 1637 |
+
"step": 2370
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 26.44,
|
| 1641 |
+
"learning_rate": 4.363636363636363e-05,
|
| 1642 |
+
"loss": 0.026,
|
| 1643 |
+
"step": 2380
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 26.56,
|
| 1647 |
+
"learning_rate": 4.227272727272727e-05,
|
| 1648 |
+
"loss": 0.0391,
|
| 1649 |
+
"step": 2390
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 26.67,
|
| 1653 |
+
"learning_rate": 4.09090909090909e-05,
|
| 1654 |
+
"loss": 0.0424,
|
| 1655 |
+
"step": 2400
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 26.67,
|
| 1659 |
+
"eval_loss": 0.016244513913989067,
|
| 1660 |
+
"eval_runtime": 151.2464,
|
| 1661 |
+
"eval_samples_per_second": 4.747,
|
| 1662 |
+
"eval_steps_per_second": 0.595,
|
| 1663 |
+
"eval_wer": 0.08555414551607445,
|
| 1664 |
+
"step": 2400
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 26.78,
|
| 1668 |
+
"learning_rate": 3.954545454545454e-05,
|
| 1669 |
+
"loss": 0.035,
|
| 1670 |
+
"step": 2410
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 26.89,
|
| 1674 |
+
"learning_rate": 3.8181818181818174e-05,
|
| 1675 |
+
"loss": 0.0218,
|
| 1676 |
+
"step": 2420
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 27.0,
|
| 1680 |
+
"learning_rate": 3.681818181818181e-05,
|
| 1681 |
+
"loss": 0.0363,
|
| 1682 |
+
"step": 2430
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 27.11,
|
| 1686 |
+
"learning_rate": 3.545454545454545e-05,
|
| 1687 |
+
"loss": 0.0285,
|
| 1688 |
+
"step": 2440
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 27.22,
|
| 1692 |
+
"learning_rate": 3.4090909090909085e-05,
|
| 1693 |
+
"loss": 0.0257,
|
| 1694 |
+
"step": 2450
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 27.33,
|
| 1698 |
+
"learning_rate": 3.2727272727272725e-05,
|
| 1699 |
+
"loss": 0.0466,
|
| 1700 |
+
"step": 2460
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 27.44,
|
| 1704 |
+
"learning_rate": 3.136363636363636e-05,
|
| 1705 |
+
"loss": 0.0194,
|
| 1706 |
+
"step": 2470
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 27.56,
|
| 1710 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 1711 |
+
"loss": 0.0579,
|
| 1712 |
+
"step": 2480
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 27.67,
|
| 1716 |
+
"learning_rate": 2.8636363636363634e-05,
|
| 1717 |
+
"loss": 0.0414,
|
| 1718 |
+
"step": 2490
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 27.78,
|
| 1722 |
+
"learning_rate": 2.727272727272727e-05,
|
| 1723 |
+
"loss": 0.0641,
|
| 1724 |
+
"step": 2500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 27.78,
|
| 1728 |
+
"eval_loss": 0.01549264881759882,
|
| 1729 |
+
"eval_runtime": 151.2416,
|
| 1730 |
+
"eval_samples_per_second": 4.747,
|
| 1731 |
+
"eval_steps_per_second": 0.595,
|
| 1732 |
+
"eval_wer": 0.0850253807106599,
|
| 1733 |
+
"step": 2500
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 27.89,
|
| 1737 |
+
"learning_rate": 2.5909090909090906e-05,
|
| 1738 |
+
"loss": 0.0307,
|
| 1739 |
+
"step": 2510
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 28.0,
|
| 1743 |
+
"learning_rate": 2.4545454545454542e-05,
|
| 1744 |
+
"loss": 0.0327,
|
| 1745 |
+
"step": 2520
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 28.11,
|
| 1749 |
+
"learning_rate": 2.3181818181818178e-05,
|
| 1750 |
+
"loss": 0.0472,
|
| 1751 |
+
"step": 2530
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 28.22,
|
| 1755 |
+
"learning_rate": 2.1818181818181814e-05,
|
| 1756 |
+
"loss": 0.0312,
|
| 1757 |
+
"step": 2540
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 28.33,
|
| 1761 |
+
"learning_rate": 2.045454545454545e-05,
|
| 1762 |
+
"loss": 0.0543,
|
| 1763 |
+
"step": 2550
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 28.44,
|
| 1767 |
+
"learning_rate": 1.9090909090909087e-05,
|
| 1768 |
+
"loss": 0.0294,
|
| 1769 |
+
"step": 2560
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 28.56,
|
| 1773 |
+
"learning_rate": 1.7727272727272726e-05,
|
| 1774 |
+
"loss": 0.044,
|
| 1775 |
+
"step": 2570
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 28.67,
|
| 1779 |
+
"learning_rate": 1.6363636363636363e-05,
|
| 1780 |
+
"loss": 0.0359,
|
| 1781 |
+
"step": 2580
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 28.78,
|
| 1785 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 1786 |
+
"loss": 0.0294,
|
| 1787 |
+
"step": 2590
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 28.89,
|
| 1791 |
+
"learning_rate": 1.3636363636363635e-05,
|
| 1792 |
+
"loss": 0.0188,
|
| 1793 |
+
"step": 2600
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 28.89,
|
| 1797 |
+
"eval_loss": 0.014889966696500778,
|
| 1798 |
+
"eval_runtime": 151.5825,
|
| 1799 |
+
"eval_samples_per_second": 4.737,
|
| 1800 |
+
"eval_steps_per_second": 0.594,
|
| 1801 |
+
"eval_wer": 0.08491962774957698,
|
| 1802 |
+
"step": 2600
|
| 1803 |
+
}
|
| 1804 |
+
],
|
| 1805 |
+
"max_steps": 2700,
|
| 1806 |
+
"num_train_epochs": 30,
|
| 1807 |
+
"total_flos": 9.754561830239207e+18,
|
| 1808 |
+
"trial_name": null,
|
| 1809 |
+
"trial_params": null
|
| 1810 |
+
}
|
new_iloko_model/checkpoint-2600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df4ceaeb36eed95f90c10e832d109616b001844d347dc46998b11093d52357e2
|
| 3 |
+
size 2799
|
new_iloko_model/checkpoint-2700/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"codevector_dim": 768,
|
| 12 |
+
"contrastive_logits_temperature": 0.1,
|
| 13 |
+
"conv_bias": true,
|
| 14 |
+
"conv_dim": [
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512
|
| 22 |
+
],
|
| 23 |
+
"conv_kernel": [
|
| 24 |
+
10,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
3,
|
| 28 |
+
3,
|
| 29 |
+
2,
|
| 30 |
+
2
|
| 31 |
+
],
|
| 32 |
+
"conv_stride": [
|
| 33 |
+
5,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2
|
| 40 |
+
],
|
| 41 |
+
"ctc_loss_reduction": "mean",
|
| 42 |
+
"ctc_zero_infinity": false,
|
| 43 |
+
"diversity_loss_weight": 0.1,
|
| 44 |
+
"do_stable_layer_norm": true,
|
| 45 |
+
"eos_token_id": 2,
|
| 46 |
+
"feat_extract_activation": "gelu",
|
| 47 |
+
"feat_extract_dropout": 0.0,
|
| 48 |
+
"feat_extract_norm": "layer",
|
| 49 |
+
"feat_proj_dropout": 0.0,
|
| 50 |
+
"feat_quantizer_dropout": 0.0,
|
| 51 |
+
"final_dropout": 0.0,
|
| 52 |
+
"gradient_checkpointing": false,
|
| 53 |
+
"hidden_act": "gelu",
|
| 54 |
+
"hidden_dropout": 0.1,
|
| 55 |
+
"hidden_size": 1024,
|
| 56 |
+
"initializer_range": 0.02,
|
| 57 |
+
"intermediate_size": 4096,
|
| 58 |
+
"layer_norm_eps": 1e-05,
|
| 59 |
+
"layerdrop": 0.1,
|
| 60 |
+
"mask_channel_length": 10,
|
| 61 |
+
"mask_channel_min_space": 1,
|
| 62 |
+
"mask_channel_other": 0.0,
|
| 63 |
+
"mask_channel_prob": 0.0,
|
| 64 |
+
"mask_channel_selection": "static",
|
| 65 |
+
"mask_feature_length": 10,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_space": 1,
|
| 69 |
+
"mask_time_other": 0.0,
|
| 70 |
+
"mask_time_prob": 0.05,
|
| 71 |
+
"mask_time_selection": "static",
|
| 72 |
+
"model_type": "wav2vec2",
|
| 73 |
+
"num_attention_heads": 16,
|
| 74 |
+
"num_codevector_groups": 2,
|
| 75 |
+
"num_codevectors_per_group": 320,
|
| 76 |
+
"num_conv_pos_embedding_groups": 16,
|
| 77 |
+
"num_conv_pos_embeddings": 128,
|
| 78 |
+
"num_feat_extract_layers": 7,
|
| 79 |
+
"num_hidden_layers": 24,
|
| 80 |
+
"num_negatives": 100,
|
| 81 |
+
"pad_token_id": 25,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"torch_dtype": "float32",
|
| 84 |
+
"transformers_version": "4.11.3",
|
| 85 |
+
"use_weighted_layer_sum": false,
|
| 86 |
+
"vocab_size": 26
|
| 87 |
+
}
|
new_iloko_model/checkpoint-2700/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
new_iloko_model/checkpoint-2700/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70e90cd9ed6cdbe5b4cbc33655ab7debe797fbe31c485236430ae9fceed04647
|
| 3 |
+
size 1262030257
|
new_iloko_model/checkpoint-2700/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bd00e891b5c780e0eda53d082a7d6e3e17604dcee1dbfdb08fae167e2bc532d
|
| 3 |
+
size 14503
|
new_iloko_model/checkpoint-2700/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46e9b594a1a572fd3c780f948f76bf38e893d4dd21c6d9854ad3fd1c3756831b
|
| 3 |
+
size 623
|
new_iloko_model/checkpoint-2700/trainer_state.json
ADDED
|
@@ -0,0 +1,1879 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 2700,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.11,
|
| 12 |
+
"learning_rate": 5.999999999999999e-06,
|
| 13 |
+
"loss": 12.0901,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.22,
|
| 18 |
+
"learning_rate": 1.1999999999999999e-05,
|
| 19 |
+
"loss": 11.5786,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.33,
|
| 24 |
+
"learning_rate": 1.7999999999999997e-05,
|
| 25 |
+
"loss": 12.4996,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.44,
|
| 30 |
+
"learning_rate": 2.3999999999999997e-05,
|
| 31 |
+
"loss": 12.1575,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.56,
|
| 36 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 37 |
+
"loss": 11.6027,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.67,
|
| 42 |
+
"learning_rate": 3.5999999999999994e-05,
|
| 43 |
+
"loss": 9.5766,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.78,
|
| 48 |
+
"learning_rate": 4.2e-05,
|
| 49 |
+
"loss": 7.3941,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.89,
|
| 54 |
+
"learning_rate": 4.7999999999999994e-05,
|
| 55 |
+
"loss": 4.7536,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.0,
|
| 60 |
+
"learning_rate": 5.399999999999999e-05,
|
| 61 |
+
"loss": 3.7347,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.11,
|
| 66 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 67 |
+
"loss": 3.1484,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.11,
|
| 72 |
+
"eval_loss": 2.928152322769165,
|
| 73 |
+
"eval_runtime": 166.6108,
|
| 74 |
+
"eval_samples_per_second": 4.309,
|
| 75 |
+
"eval_steps_per_second": 0.54,
|
| 76 |
+
"eval_wer": 1.0,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.22,
|
| 81 |
+
"learning_rate": 6.599999999999999e-05,
|
| 82 |
+
"loss": 2.8961,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.33,
|
| 87 |
+
"learning_rate": 7.199999999999999e-05,
|
| 88 |
+
"loss": 2.7846,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.44,
|
| 93 |
+
"learning_rate": 7.8e-05,
|
| 94 |
+
"loss": 2.7632,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.56,
|
| 99 |
+
"learning_rate": 8.4e-05,
|
| 100 |
+
"loss": 2.6963,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.67,
|
| 105 |
+
"learning_rate": 8.999999999999999e-05,
|
| 106 |
+
"loss": 2.7072,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.78,
|
| 111 |
+
"learning_rate": 9.599999999999999e-05,
|
| 112 |
+
"loss": 2.6889,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.89,
|
| 117 |
+
"learning_rate": 0.000102,
|
| 118 |
+
"loss": 2.6548,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.0,
|
| 123 |
+
"learning_rate": 0.00010799999999999998,
|
| 124 |
+
"loss": 2.7483,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.11,
|
| 129 |
+
"learning_rate": 0.00011399999999999999,
|
| 130 |
+
"loss": 2.6067,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.22,
|
| 135 |
+
"learning_rate": 0.00011999999999999999,
|
| 136 |
+
"loss": 2.69,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.22,
|
| 141 |
+
"eval_loss": 2.6755106449127197,
|
| 142 |
+
"eval_runtime": 166.1686,
|
| 143 |
+
"eval_samples_per_second": 4.321,
|
| 144 |
+
"eval_steps_per_second": 0.542,
|
| 145 |
+
"eval_wer": 1.0,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.33,
|
| 150 |
+
"learning_rate": 0.00012599999999999997,
|
| 151 |
+
"loss": 2.6435,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.44,
|
| 156 |
+
"learning_rate": 0.00013199999999999998,
|
| 157 |
+
"loss": 2.6126,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.56,
|
| 162 |
+
"learning_rate": 0.000138,
|
| 163 |
+
"loss": 2.6488,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.67,
|
| 168 |
+
"learning_rate": 0.00014399999999999998,
|
| 169 |
+
"loss": 2.6612,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.78,
|
| 174 |
+
"learning_rate": 0.00015,
|
| 175 |
+
"loss": 2.6615,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.89,
|
| 180 |
+
"learning_rate": 0.000156,
|
| 181 |
+
"loss": 2.6005,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.0,
|
| 186 |
+
"learning_rate": 0.000162,
|
| 187 |
+
"loss": 2.7924,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.11,
|
| 192 |
+
"learning_rate": 0.000168,
|
| 193 |
+
"loss": 2.5926,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.22,
|
| 198 |
+
"learning_rate": 0.00017399999999999997,
|
| 199 |
+
"loss": 2.6799,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.33,
|
| 204 |
+
"learning_rate": 0.00017999999999999998,
|
| 205 |
+
"loss": 2.6133,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.33,
|
| 210 |
+
"eval_loss": 2.637242078781128,
|
| 211 |
+
"eval_runtime": 154.1101,
|
| 212 |
+
"eval_samples_per_second": 4.659,
|
| 213 |
+
"eval_steps_per_second": 0.584,
|
| 214 |
+
"eval_wer": 1.0,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.44,
|
| 219 |
+
"learning_rate": 0.000186,
|
| 220 |
+
"loss": 2.6176,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.56,
|
| 225 |
+
"learning_rate": 0.00019199999999999998,
|
| 226 |
+
"loss": 2.6436,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.67,
|
| 231 |
+
"learning_rate": 0.000198,
|
| 232 |
+
"loss": 2.6309,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.78,
|
| 237 |
+
"learning_rate": 0.000204,
|
| 238 |
+
"loss": 2.6305,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.89,
|
| 243 |
+
"learning_rate": 0.00020999999999999998,
|
| 244 |
+
"loss": 2.5714,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.0,
|
| 249 |
+
"learning_rate": 0.00021599999999999996,
|
| 250 |
+
"loss": 2.6415,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.11,
|
| 255 |
+
"learning_rate": 0.00022199999999999998,
|
| 256 |
+
"loss": 2.5131,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 4.22,
|
| 261 |
+
"learning_rate": 0.00022799999999999999,
|
| 262 |
+
"loss": 2.4668,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 4.33,
|
| 267 |
+
"learning_rate": 0.000234,
|
| 268 |
+
"loss": 2.4911,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 4.44,
|
| 273 |
+
"learning_rate": 0.00023999999999999998,
|
| 274 |
+
"loss": 2.2501,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 4.44,
|
| 279 |
+
"eval_loss": 2.055128335952759,
|
| 280 |
+
"eval_runtime": 153.8268,
|
| 281 |
+
"eval_samples_per_second": 4.668,
|
| 282 |
+
"eval_steps_per_second": 0.585,
|
| 283 |
+
"eval_wer": 1.0,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.56,
|
| 288 |
+
"learning_rate": 0.00024599999999999996,
|
| 289 |
+
"loss": 2.0839,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.67,
|
| 294 |
+
"learning_rate": 0.00025199999999999995,
|
| 295 |
+
"loss": 1.8988,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.78,
|
| 300 |
+
"learning_rate": 0.000258,
|
| 301 |
+
"loss": 1.6222,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.89,
|
| 306 |
+
"learning_rate": 0.00026399999999999997,
|
| 307 |
+
"loss": 1.2373,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 5.0,
|
| 312 |
+
"learning_rate": 0.00027,
|
| 313 |
+
"loss": 1.078,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 5.11,
|
| 318 |
+
"learning_rate": 0.000276,
|
| 319 |
+
"loss": 0.866,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.22,
|
| 324 |
+
"learning_rate": 0.00028199999999999997,
|
| 325 |
+
"loss": 0.7399,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.33,
|
| 330 |
+
"learning_rate": 0.00028799999999999995,
|
| 331 |
+
"loss": 0.7751,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 5.44,
|
| 336 |
+
"learning_rate": 0.000294,
|
| 337 |
+
"loss": 0.6117,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 5.56,
|
| 342 |
+
"learning_rate": 0.0003,
|
| 343 |
+
"loss": 0.6642,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 5.56,
|
| 348 |
+
"eval_loss": 0.3055833578109741,
|
| 349 |
+
"eval_runtime": 154.0205,
|
| 350 |
+
"eval_samples_per_second": 4.662,
|
| 351 |
+
"eval_steps_per_second": 0.584,
|
| 352 |
+
"eval_wer": 0.39202622673434856,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 5.67,
|
| 357 |
+
"learning_rate": 0.00029249999999999995,
|
| 358 |
+
"loss": 0.469,
|
| 359 |
+
"step": 510
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 5.78,
|
| 363 |
+
"learning_rate": 0.000285,
|
| 364 |
+
"loss": 0.4959,
|
| 365 |
+
"step": 520
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 5.89,
|
| 369 |
+
"learning_rate": 0.00027749999999999997,
|
| 370 |
+
"loss": 0.4317,
|
| 371 |
+
"step": 530
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 6.0,
|
| 375 |
+
"learning_rate": 0.00027,
|
| 376 |
+
"loss": 0.3656,
|
| 377 |
+
"step": 540
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 6.11,
|
| 381 |
+
"learning_rate": 0.0002625,
|
| 382 |
+
"loss": 0.3608,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 6.22,
|
| 387 |
+
"learning_rate": 0.00025499999999999996,
|
| 388 |
+
"loss": 0.2853,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 6.33,
|
| 393 |
+
"learning_rate": 0.00024749999999999994,
|
| 394 |
+
"loss": 0.3534,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 6.44,
|
| 399 |
+
"learning_rate": 0.00023999999999999998,
|
| 400 |
+
"loss": 0.2887,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 6.56,
|
| 405 |
+
"learning_rate": 0.00023249999999999999,
|
| 406 |
+
"loss": 0.3348,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.67,
|
| 411 |
+
"learning_rate": 0.000225,
|
| 412 |
+
"loss": 0.2755,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.67,
|
| 417 |
+
"eval_loss": 0.1520056277513504,
|
| 418 |
+
"eval_runtime": 153.988,
|
| 419 |
+
"eval_samples_per_second": 4.663,
|
| 420 |
+
"eval_steps_per_second": 0.584,
|
| 421 |
+
"eval_wer": 0.2118231810490694,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.78,
|
| 426 |
+
"learning_rate": 0.00021749999999999997,
|
| 427 |
+
"loss": 0.2901,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.89,
|
| 432 |
+
"learning_rate": 0.00020999999999999998,
|
| 433 |
+
"loss": 0.2485,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 7.0,
|
| 438 |
+
"learning_rate": 0.0002025,
|
| 439 |
+
"loss": 0.2533,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 7.11,
|
| 444 |
+
"learning_rate": 0.000195,
|
| 445 |
+
"loss": 0.2502,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 7.22,
|
| 450 |
+
"learning_rate": 0.00018749999999999998,
|
| 451 |
+
"loss": 0.1834,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 7.33,
|
| 456 |
+
"learning_rate": 0.00017999999999999998,
|
| 457 |
+
"loss": 0.2201,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 7.44,
|
| 462 |
+
"learning_rate": 0.00017249999999999996,
|
| 463 |
+
"loss": 0.2079,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.56,
|
| 468 |
+
"learning_rate": 0.000165,
|
| 469 |
+
"loss": 0.226,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 7.67,
|
| 474 |
+
"learning_rate": 0.00015749999999999998,
|
| 475 |
+
"loss": 0.1831,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.78,
|
| 480 |
+
"learning_rate": 0.00015,
|
| 481 |
+
"loss": 0.1749,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 7.78,
|
| 486 |
+
"eval_loss": 0.10200744867324829,
|
| 487 |
+
"eval_runtime": 153.705,
|
| 488 |
+
"eval_samples_per_second": 4.671,
|
| 489 |
+
"eval_steps_per_second": 0.586,
|
| 490 |
+
"eval_wer": 0.1515439932318105,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.89,
|
| 495 |
+
"learning_rate": 0.0001425,
|
| 496 |
+
"loss": 0.163,
|
| 497 |
+
"step": 710
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 8.0,
|
| 501 |
+
"learning_rate": 0.000135,
|
| 502 |
+
"loss": 0.1567,
|
| 503 |
+
"step": 720
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 8.11,
|
| 507 |
+
"learning_rate": 0.00012749999999999998,
|
| 508 |
+
"loss": 0.1884,
|
| 509 |
+
"step": 730
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 8.22,
|
| 513 |
+
"learning_rate": 0.00011999999999999999,
|
| 514 |
+
"loss": 0.1245,
|
| 515 |
+
"step": 740
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.33,
|
| 519 |
+
"learning_rate": 0.0001125,
|
| 520 |
+
"loss": 0.145,
|
| 521 |
+
"step": 750
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.44,
|
| 525 |
+
"learning_rate": 0.00010499999999999999,
|
| 526 |
+
"loss": 0.1196,
|
| 527 |
+
"step": 760
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 8.56,
|
| 531 |
+
"learning_rate": 9.75e-05,
|
| 532 |
+
"loss": 0.1875,
|
| 533 |
+
"step": 770
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 8.67,
|
| 537 |
+
"learning_rate": 8.999999999999999e-05,
|
| 538 |
+
"loss": 0.1603,
|
| 539 |
+
"step": 780
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 8.78,
|
| 543 |
+
"learning_rate": 8.25e-05,
|
| 544 |
+
"loss": 0.1459,
|
| 545 |
+
"step": 790
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 8.89,
|
| 549 |
+
"learning_rate": 7.5e-05,
|
| 550 |
+
"loss": 0.1404,
|
| 551 |
+
"step": 800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 8.89,
|
| 555 |
+
"eval_loss": 0.08183874934911728,
|
| 556 |
+
"eval_runtime": 154.0029,
|
| 557 |
+
"eval_samples_per_second": 4.662,
|
| 558 |
+
"eval_steps_per_second": 0.584,
|
| 559 |
+
"eval_wer": 0.13293147208121828,
|
| 560 |
+
"step": 800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 9.0,
|
| 564 |
+
"learning_rate": 6.75e-05,
|
| 565 |
+
"loss": 0.11,
|
| 566 |
+
"step": 810
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.11,
|
| 570 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 571 |
+
"loss": 0.1284,
|
| 572 |
+
"step": 820
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 9.22,
|
| 576 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 577 |
+
"loss": 0.1055,
|
| 578 |
+
"step": 830
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.33,
|
| 582 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 583 |
+
"loss": 0.1459,
|
| 584 |
+
"step": 840
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 9.44,
|
| 588 |
+
"learning_rate": 3.75e-05,
|
| 589 |
+
"loss": 0.1028,
|
| 590 |
+
"step": 850
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 9.56,
|
| 594 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 595 |
+
"loss": 0.1635,
|
| 596 |
+
"step": 860
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 9.67,
|
| 600 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 601 |
+
"loss": 0.1315,
|
| 602 |
+
"step": 870
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 9.78,
|
| 606 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 607 |
+
"loss": 0.1215,
|
| 608 |
+
"step": 880
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.89,
|
| 612 |
+
"learning_rate": 7.499999999999999e-06,
|
| 613 |
+
"loss": 0.157,
|
| 614 |
+
"step": 890
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.0,
|
| 618 |
+
"learning_rate": 0.0,
|
| 619 |
+
"loss": 0.1064,
|
| 620 |
+
"step": 900
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 10.0,
|
| 624 |
+
"eval_loss": 0.07435603439807892,
|
| 625 |
+
"eval_runtime": 154.0414,
|
| 626 |
+
"eval_samples_per_second": 4.661,
|
| 627 |
+
"eval_steps_per_second": 0.584,
|
| 628 |
+
"eval_wer": 0.12531725888324874,
|
| 629 |
+
"step": 900
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 10.11,
|
| 633 |
+
"learning_rate": 0.00024409090909090905,
|
| 634 |
+
"loss": 6.6354,
|
| 635 |
+
"step": 910
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 10.22,
|
| 639 |
+
"learning_rate": 0.0002427272727272727,
|
| 640 |
+
"loss": 3.0269,
|
| 641 |
+
"step": 920
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 10.33,
|
| 645 |
+
"learning_rate": 0.00024136363636363635,
|
| 646 |
+
"loss": 2.7006,
|
| 647 |
+
"step": 930
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 10.44,
|
| 651 |
+
"learning_rate": 0.00023999999999999998,
|
| 652 |
+
"loss": 2.6422,
|
| 653 |
+
"step": 940
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 10.56,
|
| 657 |
+
"learning_rate": 0.0002386363636363636,
|
| 658 |
+
"loss": 2.6196,
|
| 659 |
+
"step": 950
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 10.67,
|
| 663 |
+
"learning_rate": 0.00023727272727272724,
|
| 664 |
+
"loss": 2.5681,
|
| 665 |
+
"step": 960
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 10.78,
|
| 669 |
+
"learning_rate": 0.0002359090909090909,
|
| 670 |
+
"loss": 2.4665,
|
| 671 |
+
"step": 970
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 10.89,
|
| 675 |
+
"learning_rate": 0.00023454545454545454,
|
| 676 |
+
"loss": 2.0804,
|
| 677 |
+
"step": 980
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 11.0,
|
| 681 |
+
"learning_rate": 0.00023318181818181814,
|
| 682 |
+
"loss": 1.3389,
|
| 683 |
+
"step": 990
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 11.11,
|
| 687 |
+
"learning_rate": 0.0002318181818181818,
|
| 688 |
+
"loss": 0.5882,
|
| 689 |
+
"step": 1000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 11.11,
|
| 693 |
+
"eval_loss": 0.21222124993801117,
|
| 694 |
+
"eval_runtime": 150.6754,
|
| 695 |
+
"eval_samples_per_second": 4.765,
|
| 696 |
+
"eval_steps_per_second": 0.597,
|
| 697 |
+
"eval_wer": 0.2306472081218274,
|
| 698 |
+
"step": 1000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 11.22,
|
| 702 |
+
"learning_rate": 0.00023045454545454544,
|
| 703 |
+
"loss": 0.2554,
|
| 704 |
+
"step": 1010
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 11.33,
|
| 708 |
+
"learning_rate": 0.00022909090909090907,
|
| 709 |
+
"loss": 0.2722,
|
| 710 |
+
"step": 1020
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 11.44,
|
| 714 |
+
"learning_rate": 0.0002277272727272727,
|
| 715 |
+
"loss": 0.2067,
|
| 716 |
+
"step": 1030
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 11.56,
|
| 720 |
+
"learning_rate": 0.00022636363636363633,
|
| 721 |
+
"loss": 0.2437,
|
| 722 |
+
"step": 1040
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 11.67,
|
| 726 |
+
"learning_rate": 0.000225,
|
| 727 |
+
"loss": 0.1675,
|
| 728 |
+
"step": 1050
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 11.78,
|
| 732 |
+
"learning_rate": 0.00022363636363636363,
|
| 733 |
+
"loss": 0.181,
|
| 734 |
+
"step": 1060
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 11.89,
|
| 738 |
+
"learning_rate": 0.00022227272727272726,
|
| 739 |
+
"loss": 0.1562,
|
| 740 |
+
"step": 1070
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 12.0,
|
| 744 |
+
"learning_rate": 0.0002209090909090909,
|
| 745 |
+
"loss": 0.1805,
|
| 746 |
+
"step": 1080
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 12.11,
|
| 750 |
+
"learning_rate": 0.00021954545454545452,
|
| 751 |
+
"loss": 0.2045,
|
| 752 |
+
"step": 1090
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 12.22,
|
| 756 |
+
"learning_rate": 0.00021818181818181816,
|
| 757 |
+
"loss": 0.1813,
|
| 758 |
+
"step": 1100
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 12.22,
|
| 762 |
+
"eval_loss": 0.0888812392950058,
|
| 763 |
+
"eval_runtime": 150.0493,
|
| 764 |
+
"eval_samples_per_second": 4.785,
|
| 765 |
+
"eval_steps_per_second": 0.6,
|
| 766 |
+
"eval_wer": 0.14075719120135363,
|
| 767 |
+
"step": 1100
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 12.33,
|
| 771 |
+
"learning_rate": 0.00021681818181818182,
|
| 772 |
+
"loss": 0.2353,
|
| 773 |
+
"step": 1110
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 12.44,
|
| 777 |
+
"learning_rate": 0.00021545454545454542,
|
| 778 |
+
"loss": 0.1769,
|
| 779 |
+
"step": 1120
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 12.56,
|
| 783 |
+
"learning_rate": 0.00021409090909090906,
|
| 784 |
+
"loss": 0.1727,
|
| 785 |
+
"step": 1130
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 12.67,
|
| 789 |
+
"learning_rate": 0.00021272727272727272,
|
| 790 |
+
"loss": 0.1213,
|
| 791 |
+
"step": 1140
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 12.78,
|
| 795 |
+
"learning_rate": 0.00021136363636363635,
|
| 796 |
+
"loss": 0.1829,
|
| 797 |
+
"step": 1150
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 12.89,
|
| 801 |
+
"learning_rate": 0.00020999999999999998,
|
| 802 |
+
"loss": 0.1414,
|
| 803 |
+
"step": 1160
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 13.0,
|
| 807 |
+
"learning_rate": 0.00020863636363636361,
|
| 808 |
+
"loss": 0.132,
|
| 809 |
+
"step": 1170
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 13.11,
|
| 813 |
+
"learning_rate": 0.00020727272727272725,
|
| 814 |
+
"loss": 0.1344,
|
| 815 |
+
"step": 1180
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 13.22,
|
| 819 |
+
"learning_rate": 0.0002059090909090909,
|
| 820 |
+
"loss": 0.1064,
|
| 821 |
+
"step": 1190
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 13.33,
|
| 825 |
+
"learning_rate": 0.0002045454545454545,
|
| 826 |
+
"loss": 0.1166,
|
| 827 |
+
"step": 1200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.33,
|
| 831 |
+
"eval_loss": 0.06592676788568497,
|
| 832 |
+
"eval_runtime": 149.9746,
|
| 833 |
+
"eval_samples_per_second": 4.787,
|
| 834 |
+
"eval_steps_per_second": 0.6,
|
| 835 |
+
"eval_wer": 0.11928934010152284,
|
| 836 |
+
"step": 1200
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 13.44,
|
| 840 |
+
"learning_rate": 0.00020318181818181815,
|
| 841 |
+
"loss": 0.1118,
|
| 842 |
+
"step": 1210
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 13.56,
|
| 846 |
+
"learning_rate": 0.0002018181818181818,
|
| 847 |
+
"loss": 0.1629,
|
| 848 |
+
"step": 1220
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 13.67,
|
| 852 |
+
"learning_rate": 0.00020045454545454544,
|
| 853 |
+
"loss": 0.1198,
|
| 854 |
+
"step": 1230
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 13.78,
|
| 858 |
+
"learning_rate": 0.0001990909090909091,
|
| 859 |
+
"loss": 0.1352,
|
| 860 |
+
"step": 1240
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 13.89,
|
| 864 |
+
"learning_rate": 0.0001977272727272727,
|
| 865 |
+
"loss": 0.132,
|
| 866 |
+
"step": 1250
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 14.0,
|
| 870 |
+
"learning_rate": 0.00019636363636363634,
|
| 871 |
+
"loss": 0.1223,
|
| 872 |
+
"step": 1260
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 14.11,
|
| 876 |
+
"learning_rate": 0.000195,
|
| 877 |
+
"loss": 0.1334,
|
| 878 |
+
"step": 1270
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 14.22,
|
| 882 |
+
"learning_rate": 0.00019363636363636363,
|
| 883 |
+
"loss": 0.1102,
|
| 884 |
+
"step": 1280
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.33,
|
| 888 |
+
"learning_rate": 0.00019227272727272723,
|
| 889 |
+
"loss": 0.1369,
|
| 890 |
+
"step": 1290
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 14.44,
|
| 894 |
+
"learning_rate": 0.0001909090909090909,
|
| 895 |
+
"loss": 0.0871,
|
| 896 |
+
"step": 1300
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 14.44,
|
| 900 |
+
"eval_loss": 0.05411907657980919,
|
| 901 |
+
"eval_runtime": 149.6406,
|
| 902 |
+
"eval_samples_per_second": 4.798,
|
| 903 |
+
"eval_steps_per_second": 0.601,
|
| 904 |
+
"eval_wer": 0.11072335025380711,
|
| 905 |
+
"step": 1300
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 14.56,
|
| 909 |
+
"learning_rate": 0.00018954545454545453,
|
| 910 |
+
"loss": 0.1193,
|
| 911 |
+
"step": 1310
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 14.67,
|
| 915 |
+
"learning_rate": 0.0001881818181818182,
|
| 916 |
+
"loss": 0.0963,
|
| 917 |
+
"step": 1320
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 14.78,
|
| 921 |
+
"learning_rate": 0.0001868181818181818,
|
| 922 |
+
"loss": 0.1135,
|
| 923 |
+
"step": 1330
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 14.89,
|
| 927 |
+
"learning_rate": 0.00018545454545454543,
|
| 928 |
+
"loss": 0.0891,
|
| 929 |
+
"step": 1340
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 15.0,
|
| 933 |
+
"learning_rate": 0.00018409090909090909,
|
| 934 |
+
"loss": 0.0724,
|
| 935 |
+
"step": 1350
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 15.11,
|
| 939 |
+
"learning_rate": 0.00018272727272727272,
|
| 940 |
+
"loss": 0.1289,
|
| 941 |
+
"step": 1360
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 15.22,
|
| 945 |
+
"learning_rate": 0.00018136363636363632,
|
| 946 |
+
"loss": 0.0639,
|
| 947 |
+
"step": 1370
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 15.33,
|
| 951 |
+
"learning_rate": 0.00017999999999999998,
|
| 952 |
+
"loss": 0.0868,
|
| 953 |
+
"step": 1380
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 15.44,
|
| 957 |
+
"learning_rate": 0.00017863636363636362,
|
| 958 |
+
"loss": 0.0749,
|
| 959 |
+
"step": 1390
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 15.56,
|
| 963 |
+
"learning_rate": 0.00017727272727272728,
|
| 964 |
+
"loss": 0.091,
|
| 965 |
+
"step": 1400
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 15.56,
|
| 969 |
+
"eval_loss": 0.046893950551748276,
|
| 970 |
+
"eval_runtime": 149.6721,
|
| 971 |
+
"eval_samples_per_second": 4.797,
|
| 972 |
+
"eval_steps_per_second": 0.601,
|
| 973 |
+
"eval_wer": 0.10046531302876481,
|
| 974 |
+
"step": 1400
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 15.67,
|
| 978 |
+
"learning_rate": 0.00017590909090909088,
|
| 979 |
+
"loss": 0.0819,
|
| 980 |
+
"step": 1410
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 15.78,
|
| 984 |
+
"learning_rate": 0.00017454545454545452,
|
| 985 |
+
"loss": 0.0966,
|
| 986 |
+
"step": 1420
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 15.89,
|
| 990 |
+
"learning_rate": 0.00017318181818181818,
|
| 991 |
+
"loss": 0.0823,
|
| 992 |
+
"step": 1430
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 16.0,
|
| 996 |
+
"learning_rate": 0.0001718181818181818,
|
| 997 |
+
"loss": 0.0958,
|
| 998 |
+
"step": 1440
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 16.11,
|
| 1002 |
+
"learning_rate": 0.00017045454545454547,
|
| 1003 |
+
"loss": 0.1276,
|
| 1004 |
+
"step": 1450
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 16.22,
|
| 1008 |
+
"learning_rate": 0.00016909090909090907,
|
| 1009 |
+
"loss": 0.0855,
|
| 1010 |
+
"step": 1460
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 16.33,
|
| 1014 |
+
"learning_rate": 0.0001677272727272727,
|
| 1015 |
+
"loss": 0.1032,
|
| 1016 |
+
"step": 1470
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 16.44,
|
| 1020 |
+
"learning_rate": 0.00016636363636363637,
|
| 1021 |
+
"loss": 0.0683,
|
| 1022 |
+
"step": 1480
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 16.56,
|
| 1026 |
+
"learning_rate": 0.000165,
|
| 1027 |
+
"loss": 0.0988,
|
| 1028 |
+
"step": 1490
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 16.67,
|
| 1032 |
+
"learning_rate": 0.0001636363636363636,
|
| 1033 |
+
"loss": 0.0551,
|
| 1034 |
+
"step": 1500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 16.67,
|
| 1038 |
+
"eval_loss": 0.043081216514110565,
|
| 1039 |
+
"eval_runtime": 150.166,
|
| 1040 |
+
"eval_samples_per_second": 4.781,
|
| 1041 |
+
"eval_steps_per_second": 0.599,
|
| 1042 |
+
"eval_wer": 0.10268612521150593,
|
| 1043 |
+
"step": 1500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 16.78,
|
| 1047 |
+
"learning_rate": 0.00016227272727272726,
|
| 1048 |
+
"loss": 0.067,
|
| 1049 |
+
"step": 1510
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 16.89,
|
| 1053 |
+
"learning_rate": 0.0001609090909090909,
|
| 1054 |
+
"loss": 0.0667,
|
| 1055 |
+
"step": 1520
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 17.0,
|
| 1059 |
+
"learning_rate": 0.00015954545454545453,
|
| 1060 |
+
"loss": 0.0639,
|
| 1061 |
+
"step": 1530
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 17.11,
|
| 1065 |
+
"learning_rate": 0.00015818181818181816,
|
| 1066 |
+
"loss": 0.0994,
|
| 1067 |
+
"step": 1540
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 17.22,
|
| 1071 |
+
"learning_rate": 0.0001568181818181818,
|
| 1072 |
+
"loss": 0.0515,
|
| 1073 |
+
"step": 1550
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 17.33,
|
| 1077 |
+
"learning_rate": 0.00015545454545454546,
|
| 1078 |
+
"loss": 0.0911,
|
| 1079 |
+
"step": 1560
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 17.44,
|
| 1083 |
+
"learning_rate": 0.0001540909090909091,
|
| 1084 |
+
"loss": 0.0771,
|
| 1085 |
+
"step": 1570
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 17.56,
|
| 1089 |
+
"learning_rate": 0.0001527272727272727,
|
| 1090 |
+
"loss": 0.0706,
|
| 1091 |
+
"step": 1580
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 17.67,
|
| 1095 |
+
"learning_rate": 0.00015136363636363635,
|
| 1096 |
+
"loss": 0.0716,
|
| 1097 |
+
"step": 1590
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 17.78,
|
| 1101 |
+
"learning_rate": 0.00015,
|
| 1102 |
+
"loss": 0.0679,
|
| 1103 |
+
"step": 1600
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 17.78,
|
| 1107 |
+
"eval_loss": 0.043587520718574524,
|
| 1108 |
+
"eval_runtime": 149.9856,
|
| 1109 |
+
"eval_samples_per_second": 4.787,
|
| 1110 |
+
"eval_steps_per_second": 0.6,
|
| 1111 |
+
"eval_wer": 0.09898477157360407,
|
| 1112 |
+
"step": 1600
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 17.89,
|
| 1116 |
+
"learning_rate": 0.00014863636363636362,
|
| 1117 |
+
"loss": 0.0822,
|
| 1118 |
+
"step": 1610
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 18.0,
|
| 1122 |
+
"learning_rate": 0.00014727272727272725,
|
| 1123 |
+
"loss": 0.0655,
|
| 1124 |
+
"step": 1620
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 18.11,
|
| 1128 |
+
"learning_rate": 0.0001459090909090909,
|
| 1129 |
+
"loss": 0.1081,
|
| 1130 |
+
"step": 1630
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 18.22,
|
| 1134 |
+
"learning_rate": 0.00014454545454545452,
|
| 1135 |
+
"loss": 0.0488,
|
| 1136 |
+
"step": 1640
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 18.33,
|
| 1140 |
+
"learning_rate": 0.00014318181818181818,
|
| 1141 |
+
"loss": 0.0739,
|
| 1142 |
+
"step": 1650
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 18.44,
|
| 1146 |
+
"learning_rate": 0.0001418181818181818,
|
| 1147 |
+
"loss": 0.0711,
|
| 1148 |
+
"step": 1660
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 18.56,
|
| 1152 |
+
"learning_rate": 0.00014045454545454544,
|
| 1153 |
+
"loss": 0.0893,
|
| 1154 |
+
"step": 1670
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 18.67,
|
| 1158 |
+
"learning_rate": 0.00013909090909090908,
|
| 1159 |
+
"loss": 0.0419,
|
| 1160 |
+
"step": 1680
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 18.78,
|
| 1164 |
+
"learning_rate": 0.0001377272727272727,
|
| 1165 |
+
"loss": 0.0625,
|
| 1166 |
+
"step": 1690
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 18.89,
|
| 1170 |
+
"learning_rate": 0.00013636363636363634,
|
| 1171 |
+
"loss": 0.0497,
|
| 1172 |
+
"step": 1700
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 18.89,
|
| 1176 |
+
"eval_loss": 0.030962202697992325,
|
| 1177 |
+
"eval_runtime": 150.4238,
|
| 1178 |
+
"eval_samples_per_second": 4.773,
|
| 1179 |
+
"eval_steps_per_second": 0.598,
|
| 1180 |
+
"eval_wer": 0.09443739424703891,
|
| 1181 |
+
"step": 1700
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 19.0,
|
| 1185 |
+
"learning_rate": 0.000135,
|
| 1186 |
+
"loss": 0.0405,
|
| 1187 |
+
"step": 1710
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 19.11,
|
| 1191 |
+
"learning_rate": 0.0001336363636363636,
|
| 1192 |
+
"loss": 0.0697,
|
| 1193 |
+
"step": 1720
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 19.22,
|
| 1197 |
+
"learning_rate": 0.00013227272727272727,
|
| 1198 |
+
"loss": 0.0434,
|
| 1199 |
+
"step": 1730
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 19.33,
|
| 1203 |
+
"learning_rate": 0.0001309090909090909,
|
| 1204 |
+
"loss": 0.0696,
|
| 1205 |
+
"step": 1740
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 19.44,
|
| 1209 |
+
"learning_rate": 0.00012954545454545453,
|
| 1210 |
+
"loss": 0.0538,
|
| 1211 |
+
"step": 1750
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 19.56,
|
| 1215 |
+
"learning_rate": 0.00012818181818181817,
|
| 1216 |
+
"loss": 0.0951,
|
| 1217 |
+
"step": 1760
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 19.67,
|
| 1221 |
+
"learning_rate": 0.0001268181818181818,
|
| 1222 |
+
"loss": 0.0545,
|
| 1223 |
+
"step": 1770
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 19.78,
|
| 1227 |
+
"learning_rate": 0.00012545454545454543,
|
| 1228 |
+
"loss": 0.0622,
|
| 1229 |
+
"step": 1780
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 19.89,
|
| 1233 |
+
"learning_rate": 0.0001240909090909091,
|
| 1234 |
+
"loss": 0.0884,
|
| 1235 |
+
"step": 1790
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 20.0,
|
| 1239 |
+
"learning_rate": 0.00012272727272727272,
|
| 1240 |
+
"loss": 0.0669,
|
| 1241 |
+
"step": 1800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 20.0,
|
| 1245 |
+
"eval_loss": 0.026656057685613632,
|
| 1246 |
+
"eval_runtime": 150.2929,
|
| 1247 |
+
"eval_samples_per_second": 4.777,
|
| 1248 |
+
"eval_steps_per_second": 0.599,
|
| 1249 |
+
"eval_wer": 0.0942258883248731,
|
| 1250 |
+
"step": 1800
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 20.11,
|
| 1254 |
+
"learning_rate": 0.00012136363636363636,
|
| 1255 |
+
"loss": 0.066,
|
| 1256 |
+
"step": 1810
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 20.22,
|
| 1260 |
+
"learning_rate": 0.00011999999999999999,
|
| 1261 |
+
"loss": 0.0337,
|
| 1262 |
+
"step": 1820
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 20.33,
|
| 1266 |
+
"learning_rate": 0.00011863636363636362,
|
| 1267 |
+
"loss": 0.0766,
|
| 1268 |
+
"step": 1830
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 20.44,
|
| 1272 |
+
"learning_rate": 0.00011727272727272727,
|
| 1273 |
+
"loss": 0.0625,
|
| 1274 |
+
"step": 1840
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 20.56,
|
| 1278 |
+
"learning_rate": 0.0001159090909090909,
|
| 1279 |
+
"loss": 0.0539,
|
| 1280 |
+
"step": 1850
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 20.67,
|
| 1284 |
+
"learning_rate": 0.00011454545454545453,
|
| 1285 |
+
"loss": 0.0605,
|
| 1286 |
+
"step": 1860
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 20.78,
|
| 1290 |
+
"learning_rate": 0.00011318181818181817,
|
| 1291 |
+
"loss": 0.0465,
|
| 1292 |
+
"step": 1870
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 20.89,
|
| 1296 |
+
"learning_rate": 0.00011181818181818181,
|
| 1297 |
+
"loss": 0.0642,
|
| 1298 |
+
"step": 1880
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 21.0,
|
| 1302 |
+
"learning_rate": 0.00011045454545454545,
|
| 1303 |
+
"loss": 0.047,
|
| 1304 |
+
"step": 1890
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 21.11,
|
| 1308 |
+
"learning_rate": 0.00010909090909090908,
|
| 1309 |
+
"loss": 0.0532,
|
| 1310 |
+
"step": 1900
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 21.11,
|
| 1314 |
+
"eval_loss": 0.02508886530995369,
|
| 1315 |
+
"eval_runtime": 150.8935,
|
| 1316 |
+
"eval_samples_per_second": 4.758,
|
| 1317 |
+
"eval_steps_per_second": 0.596,
|
| 1318 |
+
"eval_wer": 0.09105329949238579,
|
| 1319 |
+
"step": 1900
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 21.22,
|
| 1323 |
+
"learning_rate": 0.00010772727272727271,
|
| 1324 |
+
"loss": 0.0245,
|
| 1325 |
+
"step": 1910
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 21.33,
|
| 1329 |
+
"learning_rate": 0.00010636363636363636,
|
| 1330 |
+
"loss": 0.0614,
|
| 1331 |
+
"step": 1920
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 21.44,
|
| 1335 |
+
"learning_rate": 0.00010499999999999999,
|
| 1336 |
+
"loss": 0.0333,
|
| 1337 |
+
"step": 1930
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 21.56,
|
| 1341 |
+
"learning_rate": 0.00010363636363636362,
|
| 1342 |
+
"loss": 0.0463,
|
| 1343 |
+
"step": 1940
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 21.67,
|
| 1347 |
+
"learning_rate": 0.00010227272727272726,
|
| 1348 |
+
"loss": 0.0509,
|
| 1349 |
+
"step": 1950
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 21.78,
|
| 1353 |
+
"learning_rate": 0.0001009090909090909,
|
| 1354 |
+
"loss": 0.0522,
|
| 1355 |
+
"step": 1960
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 21.89,
|
| 1359 |
+
"learning_rate": 9.954545454545455e-05,
|
| 1360 |
+
"loss": 0.032,
|
| 1361 |
+
"step": 1970
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 22.0,
|
| 1365 |
+
"learning_rate": 9.818181818181817e-05,
|
| 1366 |
+
"loss": 0.0491,
|
| 1367 |
+
"step": 1980
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 22.11,
|
| 1371 |
+
"learning_rate": 9.681818181818181e-05,
|
| 1372 |
+
"loss": 0.0603,
|
| 1373 |
+
"step": 1990
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 22.22,
|
| 1377 |
+
"learning_rate": 9.545454545454545e-05,
|
| 1378 |
+
"loss": 0.0269,
|
| 1379 |
+
"step": 2000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 22.22,
|
| 1383 |
+
"eval_loss": 0.02213282324373722,
|
| 1384 |
+
"eval_runtime": 150.9545,
|
| 1385 |
+
"eval_samples_per_second": 4.756,
|
| 1386 |
+
"eval_steps_per_second": 0.596,
|
| 1387 |
+
"eval_wer": 0.09020727580372251,
|
| 1388 |
+
"step": 2000
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 22.33,
|
| 1392 |
+
"learning_rate": 9.40909090909091e-05,
|
| 1393 |
+
"loss": 0.0544,
|
| 1394 |
+
"step": 2010
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 22.44,
|
| 1398 |
+
"learning_rate": 9.272727272727271e-05,
|
| 1399 |
+
"loss": 0.0331,
|
| 1400 |
+
"step": 2020
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 22.56,
|
| 1404 |
+
"learning_rate": 9.136363636363636e-05,
|
| 1405 |
+
"loss": 0.0502,
|
| 1406 |
+
"step": 2030
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 22.67,
|
| 1410 |
+
"learning_rate": 8.999999999999999e-05,
|
| 1411 |
+
"loss": 0.0493,
|
| 1412 |
+
"step": 2040
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 22.78,
|
| 1416 |
+
"learning_rate": 8.863636363636364e-05,
|
| 1417 |
+
"loss": 0.0579,
|
| 1418 |
+
"step": 2050
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 22.89,
|
| 1422 |
+
"learning_rate": 8.727272727272726e-05,
|
| 1423 |
+
"loss": 0.0392,
|
| 1424 |
+
"step": 2060
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 23.0,
|
| 1428 |
+
"learning_rate": 8.59090909090909e-05,
|
| 1429 |
+
"loss": 0.0423,
|
| 1430 |
+
"step": 2070
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 23.11,
|
| 1434 |
+
"learning_rate": 8.454545454545454e-05,
|
| 1435 |
+
"loss": 0.0696,
|
| 1436 |
+
"step": 2080
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 23.22,
|
| 1440 |
+
"learning_rate": 8.318181818181818e-05,
|
| 1441 |
+
"loss": 0.0304,
|
| 1442 |
+
"step": 2090
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 23.33,
|
| 1446 |
+
"learning_rate": 8.18181818181818e-05,
|
| 1447 |
+
"loss": 0.0686,
|
| 1448 |
+
"step": 2100
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 23.33,
|
| 1452 |
+
"eval_loss": 0.020356198772788048,
|
| 1453 |
+
"eval_runtime": 151.0382,
|
| 1454 |
+
"eval_samples_per_second": 4.754,
|
| 1455 |
+
"eval_steps_per_second": 0.596,
|
| 1456 |
+
"eval_wer": 0.08745769881556684,
|
| 1457 |
+
"step": 2100
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 23.44,
|
| 1461 |
+
"learning_rate": 8.045454545454545e-05,
|
| 1462 |
+
"loss": 0.0523,
|
| 1463 |
+
"step": 2110
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 23.56,
|
| 1467 |
+
"learning_rate": 7.909090909090908e-05,
|
| 1468 |
+
"loss": 0.0583,
|
| 1469 |
+
"step": 2120
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 23.67,
|
| 1473 |
+
"learning_rate": 7.772727272727273e-05,
|
| 1474 |
+
"loss": 0.0495,
|
| 1475 |
+
"step": 2130
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 23.78,
|
| 1479 |
+
"learning_rate": 7.636363636363635e-05,
|
| 1480 |
+
"loss": 0.0476,
|
| 1481 |
+
"step": 2140
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 23.89,
|
| 1485 |
+
"learning_rate": 7.5e-05,
|
| 1486 |
+
"loss": 0.0321,
|
| 1487 |
+
"step": 2150
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 24.0,
|
| 1491 |
+
"learning_rate": 7.363636363636363e-05,
|
| 1492 |
+
"loss": 0.0393,
|
| 1493 |
+
"step": 2160
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 24.11,
|
| 1497 |
+
"learning_rate": 7.227272727272726e-05,
|
| 1498 |
+
"loss": 0.0425,
|
| 1499 |
+
"step": 2170
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 24.22,
|
| 1503 |
+
"learning_rate": 7.09090909090909e-05,
|
| 1504 |
+
"loss": 0.064,
|
| 1505 |
+
"step": 2180
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 24.33,
|
| 1509 |
+
"learning_rate": 6.954545454545454e-05,
|
| 1510 |
+
"loss": 0.048,
|
| 1511 |
+
"step": 2190
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 24.44,
|
| 1515 |
+
"learning_rate": 6.818181818181817e-05,
|
| 1516 |
+
"loss": 0.0242,
|
| 1517 |
+
"step": 2200
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 24.44,
|
| 1521 |
+
"eval_loss": 0.018909653648734093,
|
| 1522 |
+
"eval_runtime": 151.2527,
|
| 1523 |
+
"eval_samples_per_second": 4.747,
|
| 1524 |
+
"eval_steps_per_second": 0.595,
|
| 1525 |
+
"eval_wer": 0.08724619289340102,
|
| 1526 |
+
"step": 2200
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 24.56,
|
| 1530 |
+
"learning_rate": 6.68181818181818e-05,
|
| 1531 |
+
"loss": 0.0481,
|
| 1532 |
+
"step": 2210
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 24.67,
|
| 1536 |
+
"learning_rate": 6.545454545454545e-05,
|
| 1537 |
+
"loss": 0.0442,
|
| 1538 |
+
"step": 2220
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 24.78,
|
| 1542 |
+
"learning_rate": 6.409090909090908e-05,
|
| 1543 |
+
"loss": 0.05,
|
| 1544 |
+
"step": 2230
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 24.89,
|
| 1548 |
+
"learning_rate": 6.272727272727272e-05,
|
| 1549 |
+
"loss": 0.0474,
|
| 1550 |
+
"step": 2240
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 25.0,
|
| 1554 |
+
"learning_rate": 6.136363636363636e-05,
|
| 1555 |
+
"loss": 0.0391,
|
| 1556 |
+
"step": 2250
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 25.11,
|
| 1560 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 1561 |
+
"loss": 0.068,
|
| 1562 |
+
"step": 2260
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 25.22,
|
| 1566 |
+
"learning_rate": 5.8636363636363634e-05,
|
| 1567 |
+
"loss": 0.0189,
|
| 1568 |
+
"step": 2270
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 25.33,
|
| 1572 |
+
"learning_rate": 5.727272727272727e-05,
|
| 1573 |
+
"loss": 0.0396,
|
| 1574 |
+
"step": 2280
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 25.44,
|
| 1578 |
+
"learning_rate": 5.590909090909091e-05,
|
| 1579 |
+
"loss": 0.0317,
|
| 1580 |
+
"step": 2290
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 25.56,
|
| 1584 |
+
"learning_rate": 5.454545454545454e-05,
|
| 1585 |
+
"loss": 0.0454,
|
| 1586 |
+
"step": 2300
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 25.56,
|
| 1590 |
+
"eval_loss": 0.01684204488992691,
|
| 1591 |
+
"eval_runtime": 150.5397,
|
| 1592 |
+
"eval_samples_per_second": 4.77,
|
| 1593 |
+
"eval_steps_per_second": 0.598,
|
| 1594 |
+
"eval_wer": 0.08513113367174281,
|
| 1595 |
+
"step": 2300
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 25.67,
|
| 1599 |
+
"learning_rate": 5.318181818181818e-05,
|
| 1600 |
+
"loss": 0.0317,
|
| 1601 |
+
"step": 2310
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 25.78,
|
| 1605 |
+
"learning_rate": 5.181818181818181e-05,
|
| 1606 |
+
"loss": 0.0315,
|
| 1607 |
+
"step": 2320
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 25.89,
|
| 1611 |
+
"learning_rate": 5.045454545454545e-05,
|
| 1612 |
+
"loss": 0.0302,
|
| 1613 |
+
"step": 2330
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 26.0,
|
| 1617 |
+
"learning_rate": 4.9090909090909084e-05,
|
| 1618 |
+
"loss": 0.0359,
|
| 1619 |
+
"step": 2340
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 26.11,
|
| 1623 |
+
"learning_rate": 4.7727272727272724e-05,
|
| 1624 |
+
"loss": 0.0433,
|
| 1625 |
+
"step": 2350
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 26.22,
|
| 1629 |
+
"learning_rate": 4.6363636363636356e-05,
|
| 1630 |
+
"loss": 0.0222,
|
| 1631 |
+
"step": 2360
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 26.33,
|
| 1635 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 1636 |
+
"loss": 0.0474,
|
| 1637 |
+
"step": 2370
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 26.44,
|
| 1641 |
+
"learning_rate": 4.363636363636363e-05,
|
| 1642 |
+
"loss": 0.026,
|
| 1643 |
+
"step": 2380
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 26.56,
|
| 1647 |
+
"learning_rate": 4.227272727272727e-05,
|
| 1648 |
+
"loss": 0.0391,
|
| 1649 |
+
"step": 2390
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 26.67,
|
| 1653 |
+
"learning_rate": 4.09090909090909e-05,
|
| 1654 |
+
"loss": 0.0424,
|
| 1655 |
+
"step": 2400
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 26.67,
|
| 1659 |
+
"eval_loss": 0.016244513913989067,
|
| 1660 |
+
"eval_runtime": 151.2464,
|
| 1661 |
+
"eval_samples_per_second": 4.747,
|
| 1662 |
+
"eval_steps_per_second": 0.595,
|
| 1663 |
+
"eval_wer": 0.08555414551607445,
|
| 1664 |
+
"step": 2400
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 26.78,
|
| 1668 |
+
"learning_rate": 3.954545454545454e-05,
|
| 1669 |
+
"loss": 0.035,
|
| 1670 |
+
"step": 2410
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 26.89,
|
| 1674 |
+
"learning_rate": 3.8181818181818174e-05,
|
| 1675 |
+
"loss": 0.0218,
|
| 1676 |
+
"step": 2420
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 27.0,
|
| 1680 |
+
"learning_rate": 3.681818181818181e-05,
|
| 1681 |
+
"loss": 0.0363,
|
| 1682 |
+
"step": 2430
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 27.11,
|
| 1686 |
+
"learning_rate": 3.545454545454545e-05,
|
| 1687 |
+
"loss": 0.0285,
|
| 1688 |
+
"step": 2440
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 27.22,
|
| 1692 |
+
"learning_rate": 3.4090909090909085e-05,
|
| 1693 |
+
"loss": 0.0257,
|
| 1694 |
+
"step": 2450
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 27.33,
|
| 1698 |
+
"learning_rate": 3.2727272727272725e-05,
|
| 1699 |
+
"loss": 0.0466,
|
| 1700 |
+
"step": 2460
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 27.44,
|
| 1704 |
+
"learning_rate": 3.136363636363636e-05,
|
| 1705 |
+
"loss": 0.0194,
|
| 1706 |
+
"step": 2470
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 27.56,
|
| 1710 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 1711 |
+
"loss": 0.0579,
|
| 1712 |
+
"step": 2480
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 27.67,
|
| 1716 |
+
"learning_rate": 2.8636363636363634e-05,
|
| 1717 |
+
"loss": 0.0414,
|
| 1718 |
+
"step": 2490
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 27.78,
|
| 1722 |
+
"learning_rate": 2.727272727272727e-05,
|
| 1723 |
+
"loss": 0.0641,
|
| 1724 |
+
"step": 2500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 27.78,
|
| 1728 |
+
"eval_loss": 0.01549264881759882,
|
| 1729 |
+
"eval_runtime": 151.2416,
|
| 1730 |
+
"eval_samples_per_second": 4.747,
|
| 1731 |
+
"eval_steps_per_second": 0.595,
|
| 1732 |
+
"eval_wer": 0.0850253807106599,
|
| 1733 |
+
"step": 2500
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 27.89,
|
| 1737 |
+
"learning_rate": 2.5909090909090906e-05,
|
| 1738 |
+
"loss": 0.0307,
|
| 1739 |
+
"step": 2510
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 28.0,
|
| 1743 |
+
"learning_rate": 2.4545454545454542e-05,
|
| 1744 |
+
"loss": 0.0327,
|
| 1745 |
+
"step": 2520
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 28.11,
|
| 1749 |
+
"learning_rate": 2.3181818181818178e-05,
|
| 1750 |
+
"loss": 0.0472,
|
| 1751 |
+
"step": 2530
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 28.22,
|
| 1755 |
+
"learning_rate": 2.1818181818181814e-05,
|
| 1756 |
+
"loss": 0.0312,
|
| 1757 |
+
"step": 2540
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 28.33,
|
| 1761 |
+
"learning_rate": 2.045454545454545e-05,
|
| 1762 |
+
"loss": 0.0543,
|
| 1763 |
+
"step": 2550
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 28.44,
|
| 1767 |
+
"learning_rate": 1.9090909090909087e-05,
|
| 1768 |
+
"loss": 0.0294,
|
| 1769 |
+
"step": 2560
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 28.56,
|
| 1773 |
+
"learning_rate": 1.7727272727272726e-05,
|
| 1774 |
+
"loss": 0.044,
|
| 1775 |
+
"step": 2570
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 28.67,
|
| 1779 |
+
"learning_rate": 1.6363636363636363e-05,
|
| 1780 |
+
"loss": 0.0359,
|
| 1781 |
+
"step": 2580
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 28.78,
|
| 1785 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 1786 |
+
"loss": 0.0294,
|
| 1787 |
+
"step": 2590
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 28.89,
|
| 1791 |
+
"learning_rate": 1.3636363636363635e-05,
|
| 1792 |
+
"loss": 0.0188,
|
| 1793 |
+
"step": 2600
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 28.89,
|
| 1797 |
+
"eval_loss": 0.014889966696500778,
|
| 1798 |
+
"eval_runtime": 151.5825,
|
| 1799 |
+
"eval_samples_per_second": 4.737,
|
| 1800 |
+
"eval_steps_per_second": 0.594,
|
| 1801 |
+
"eval_wer": 0.08491962774957698,
|
| 1802 |
+
"step": 2600
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 29.0,
|
| 1806 |
+
"learning_rate": 1.2272727272727271e-05,
|
| 1807 |
+
"loss": 0.0307,
|
| 1808 |
+
"step": 2610
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 29.11,
|
| 1812 |
+
"learning_rate": 1.0909090909090907e-05,
|
| 1813 |
+
"loss": 0.0453,
|
| 1814 |
+
"step": 2620
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 29.22,
|
| 1818 |
+
"learning_rate": 9.545454545454543e-06,
|
| 1819 |
+
"loss": 0.0252,
|
| 1820 |
+
"step": 2630
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 29.33,
|
| 1824 |
+
"learning_rate": 8.181818181818181e-06,
|
| 1825 |
+
"loss": 0.0431,
|
| 1826 |
+
"step": 2640
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 29.44,
|
| 1830 |
+
"learning_rate": 6.8181818181818174e-06,
|
| 1831 |
+
"loss": 0.0322,
|
| 1832 |
+
"step": 2650
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 29.56,
|
| 1836 |
+
"learning_rate": 5.454545454545454e-06,
|
| 1837 |
+
"loss": 0.0282,
|
| 1838 |
+
"step": 2660
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 29.67,
|
| 1842 |
+
"learning_rate": 4.090909090909091e-06,
|
| 1843 |
+
"loss": 0.0194,
|
| 1844 |
+
"step": 2670
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 29.78,
|
| 1848 |
+
"learning_rate": 2.727272727272727e-06,
|
| 1849 |
+
"loss": 0.0388,
|
| 1850 |
+
"step": 2680
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 29.89,
|
| 1854 |
+
"learning_rate": 1.3636363636363634e-06,
|
| 1855 |
+
"loss": 0.0264,
|
| 1856 |
+
"step": 2690
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 30.0,
|
| 1860 |
+
"learning_rate": 0.0,
|
| 1861 |
+
"loss": 0.0319,
|
| 1862 |
+
"step": 2700
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 30.0,
|
| 1866 |
+
"eval_loss": 0.01460767537355423,
|
| 1867 |
+
"eval_runtime": 151.6395,
|
| 1868 |
+
"eval_samples_per_second": 4.735,
|
| 1869 |
+
"eval_steps_per_second": 0.594,
|
| 1870 |
+
"eval_wer": 0.0850253807106599,
|
| 1871 |
+
"step": 2700
|
| 1872 |
+
}
|
| 1873 |
+
],
|
| 1874 |
+
"max_steps": 2700,
|
| 1875 |
+
"num_train_epochs": 30,
|
| 1876 |
+
"total_flos": 1.0112899150967486e+19,
|
| 1877 |
+
"trial_name": null,
|
| 1878 |
+
"trial_params": null
|
| 1879 |
+
}
|
new_iloko_model/checkpoint-2700/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df4ceaeb36eed95f90c10e832d109616b001844d347dc46998b11093d52357e2
|
| 3 |
+
size 2799
|
new_iloko_model/config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"codevector_dim": 768,
|
| 12 |
+
"contrastive_logits_temperature": 0.1,
|
| 13 |
+
"conv_bias": true,
|
| 14 |
+
"conv_dim": [
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512
|
| 22 |
+
],
|
| 23 |
+
"conv_kernel": [
|
| 24 |
+
10,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
3,
|
| 28 |
+
3,
|
| 29 |
+
2,
|
| 30 |
+
2
|
| 31 |
+
],
|
| 32 |
+
"conv_stride": [
|
| 33 |
+
5,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2
|
| 40 |
+
],
|
| 41 |
+
"ctc_loss_reduction": "mean",
|
| 42 |
+
"ctc_zero_infinity": false,
|
| 43 |
+
"diversity_loss_weight": 0.1,
|
| 44 |
+
"do_stable_layer_norm": true,
|
| 45 |
+
"eos_token_id": 2,
|
| 46 |
+
"feat_extract_activation": "gelu",
|
| 47 |
+
"feat_extract_dropout": 0.0,
|
| 48 |
+
"feat_extract_norm": "layer",
|
| 49 |
+
"feat_proj_dropout": 0.0,
|
| 50 |
+
"feat_quantizer_dropout": 0.0,
|
| 51 |
+
"final_dropout": 0.0,
|
| 52 |
+
"gradient_checkpointing": false,
|
| 53 |
+
"hidden_act": "gelu",
|
| 54 |
+
"hidden_dropout": 0.1,
|
| 55 |
+
"hidden_size": 1024,
|
| 56 |
+
"initializer_range": 0.02,
|
| 57 |
+
"intermediate_size": 4096,
|
| 58 |
+
"layer_norm_eps": 1e-05,
|
| 59 |
+
"layerdrop": 0.1,
|
| 60 |
+
"mask_channel_length": 10,
|
| 61 |
+
"mask_channel_min_space": 1,
|
| 62 |
+
"mask_channel_other": 0.0,
|
| 63 |
+
"mask_channel_prob": 0.0,
|
| 64 |
+
"mask_channel_selection": "static",
|
| 65 |
+
"mask_feature_length": 10,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_space": 1,
|
| 69 |
+
"mask_time_other": 0.0,
|
| 70 |
+
"mask_time_prob": 0.05,
|
| 71 |
+
"mask_time_selection": "static",
|
| 72 |
+
"model_type": "wav2vec2",
|
| 73 |
+
"num_attention_heads": 16,
|
| 74 |
+
"num_codevector_groups": 2,
|
| 75 |
+
"num_codevectors_per_group": 320,
|
| 76 |
+
"num_conv_pos_embedding_groups": 16,
|
| 77 |
+
"num_conv_pos_embeddings": 128,
|
| 78 |
+
"num_feat_extract_layers": 7,
|
| 79 |
+
"num_hidden_layers": 24,
|
| 80 |
+
"num_negatives": 100,
|
| 81 |
+
"pad_token_id": 25,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"torch_dtype": "float32",
|
| 84 |
+
"transformers_version": "4.11.3",
|
| 85 |
+
"use_weighted_layer_sum": false,
|
| 86 |
+
"vocab_size": 26
|
| 87 |
+
}
|
new_iloko_model/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
new_iloko_model/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70e90cd9ed6cdbe5b4cbc33655ab7debe797fbe31c485236430ae9fceed04647
|
| 3 |
+
size 1262030257
|
new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634895404.4292953/events.out.tfevents.1634895404.e109e789bd2a.91.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e42b1b11657c7894e0e066c414dfe34da2975a0fb7172a27ca3a61acc352dcca
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634896681.5475419/events.out.tfevents.1634896681.e109e789bd2a.91.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6cabfe24da4181eaf0d1e4b82961c25dc9d431c0ff3ee9ead5ac297710a575b
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634896893.7200775/events.out.tfevents.1634896893.e109e789bd2a.91.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a8102a85cbfe71178cf61d7e2be24d385794b8604dd1b2c7b3695be02ddbdb5
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/1634897154.445655/events.out.tfevents.1634897154.e109e789bd2a.91.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7ee9c413ea2c49f7a2b94f6c5a918a4fca974bb8c9d99a34063b09e28331a35
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_09-31-54_e109e789bd2a/events.out.tfevents.1634895404.e109e789bd2a.91.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06ab73d2c4fb80f9028693001a1eb8b0df22863c5528b3b153012c1259ce4e92
|
| 3 |
+
size 17682
|
new_iloko_model/runs/Oct22_10-50-27_8cab4545d814/1634899887.187254/events.out.tfevents.1634899887.8cab4545d814.91.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5928db5166c8a7ed7c926aa0d7036492f37afe6132ed91569b56a5d21284bebf
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_10-50-27_8cab4545d814/events.out.tfevents.1634899887.8cab4545d814.91.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58826c30677ba378837aa3b7637b89f8aac9ccd518ab06053b050ac00708e1cf
|
| 3 |
+
size 21716
|
new_iloko_model/runs/Oct22_13-59-07_9090242a067f/1634911877.7595568/events.out.tfevents.1634911877.9090242a067f.90.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86fcb2312d34dbe1874efddcfa1bfe496ab0879bdcf10fce9f63b7cb9bd2f7aa
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_13-59-07_9090242a067f/events.out.tfevents.1634911877.9090242a067f.90.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22e300362b446735bc59b24a80474a16c8f0ad96a929ecec7f5acbf772f8fc03
|
| 3 |
+
size 40
|
new_iloko_model/runs/Oct22_14-44-05_33dae6ed61d6/1634914033.677216/events.out.tfevents.1634914033.33dae6ed61d6.92.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6309a07f4d536c365731e85a12d473badf7d9adb2bf6ea30b15b3ac56b1bf2ed
|
| 3 |
+
size 4537
|
new_iloko_model/runs/Oct22_14-44-05_33dae6ed61d6/events.out.tfevents.1634914033.33dae6ed61d6.92.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c2233a061d2210cbdb7c94f8d49e3724a78b053e5d7f330b4086fb6e2c72660
|
| 3 |
+
size 38750
|
new_iloko_model/special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
new_iloko_model/tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
new_iloko_model/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df4ceaeb36eed95f90c10e832d109616b001844d347dc46998b11093d52357e2
|
| 3 |
+
size 2799
|
new_iloko_model/vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"'": 0, "e": 1, "u": 2, "m": 3, "w": 4, "o": 5, "h": 6, "\n": 7, "l": 8, "a": 9, "‘": 10, "n": 11, "b": 12, "i": 13, "d": 14, "k": 15, "g": 16, "t": 17, "y": 19, "r": 20, "’": 21, "p": 22, "s": 23, "|": 18, "[UNK]": 24, "[PAD]": 25}
|
optimizer-004.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78aafee95b6216ac171e53021eafed64e9842d37a5c8e431a5cbf498db55d6cb
|
| 3 |
+
size 2490272145
|
optimizer-005.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81be8385323fcdf9399055951167eebe18ad8bd1db52cf6b9e5524c9bc1634ff
|
| 3 |
+
size 2490272145
|