First model version
Browse files- config.json +87 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2695 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/dbfs/FileStore/Iskaj/Models/VP_CGN_60EP/checkpoint-38000",
|
| 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": 47,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"torch_dtype": "float32",
|
| 84 |
+
"transformers_version": "4.12.5",
|
| 85 |
+
"use_weighted_layer_sum": false,
|
| 86 |
+
"vocab_size": 48
|
| 87 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c098377db53e759e391a74f884efe428250c482397438a5a69fa3337b129332
|
| 3 |
+
size 2490452561
|
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 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:833e07f087b1a574a62e02a64f293db56aadde68440a0087f9b759bbf46cc396
|
| 3 |
+
size 1262120497
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70b16bde4011814786bd6f501403ee42632609af2fdbaf703ee39a56e6065162
|
| 3 |
+
size 14567
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fd6fdeddacb9008f0cd0e46da625757884161d02561883f28090288544e3ec9
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e85a2a26a9ba15c82881a498c8a07d7dd64753989044b286071c6c466d8c7159
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2695 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8958114385604858,
|
| 3 |
+
"best_model_checkpoint": "/dbfs/FileStore/Iskaj/Models/VP_CGN_40EP_To_60EP/checkpoint-1600",
|
| 4 |
+
"epoch": 19.623891497130934,
|
| 5 |
+
"global_step": 18800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.05,
|
| 12 |
+
"learning_rate": 9.973903966597077e-05,
|
| 13 |
+
"loss": 0.0836,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.1,
|
| 18 |
+
"learning_rate": 9.947807933194155e-05,
|
| 19 |
+
"loss": 0.0835,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.16,
|
| 24 |
+
"learning_rate": 9.921711899791232e-05,
|
| 25 |
+
"loss": 0.0815,
|
| 26 |
+
"step": 150
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.21,
|
| 30 |
+
"learning_rate": 9.895615866388309e-05,
|
| 31 |
+
"loss": 0.0777,
|
| 32 |
+
"step": 200
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.26,
|
| 36 |
+
"learning_rate": 9.869519832985387e-05,
|
| 37 |
+
"loss": 0.0852,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.31,
|
| 42 |
+
"learning_rate": 9.843423799582464e-05,
|
| 43 |
+
"loss": 0.0747,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.37,
|
| 48 |
+
"learning_rate": 9.817327766179542e-05,
|
| 49 |
+
"loss": 0.079,
|
| 50 |
+
"step": 350
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.42,
|
| 54 |
+
"learning_rate": 9.791231732776618e-05,
|
| 55 |
+
"loss": 0.0776,
|
| 56 |
+
"step": 400
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.42,
|
| 60 |
+
"eval_loss": 0.9396540522575378,
|
| 61 |
+
"eval_runtime": 2420.5437,
|
| 62 |
+
"eval_samples_per_second": 9.084,
|
| 63 |
+
"eval_steps_per_second": 1.136,
|
| 64 |
+
"eval_wer": 0.20953813057031215,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.47,
|
| 69 |
+
"learning_rate": 9.765135699373696e-05,
|
| 70 |
+
"loss": 0.0793,
|
| 71 |
+
"step": 450
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.52,
|
| 75 |
+
"learning_rate": 9.739039665970773e-05,
|
| 76 |
+
"loss": 0.079,
|
| 77 |
+
"step": 500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.57,
|
| 81 |
+
"learning_rate": 9.712943632567851e-05,
|
| 82 |
+
"loss": 0.0723,
|
| 83 |
+
"step": 550
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.63,
|
| 87 |
+
"learning_rate": 9.686847599164928e-05,
|
| 88 |
+
"loss": 0.0749,
|
| 89 |
+
"step": 600
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.68,
|
| 93 |
+
"learning_rate": 9.660751565762006e-05,
|
| 94 |
+
"loss": 0.0694,
|
| 95 |
+
"step": 650
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.73,
|
| 99 |
+
"learning_rate": 9.634655532359081e-05,
|
| 100 |
+
"loss": 0.0765,
|
| 101 |
+
"step": 700
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.78,
|
| 105 |
+
"learning_rate": 9.608559498956159e-05,
|
| 106 |
+
"loss": 0.0655,
|
| 107 |
+
"step": 750
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.83,
|
| 111 |
+
"learning_rate": 9.582463465553236e-05,
|
| 112 |
+
"loss": 0.0663,
|
| 113 |
+
"step": 800
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.83,
|
| 117 |
+
"eval_loss": 0.9835959672927856,
|
| 118 |
+
"eval_runtime": 2404.5796,
|
| 119 |
+
"eval_samples_per_second": 9.144,
|
| 120 |
+
"eval_steps_per_second": 1.143,
|
| 121 |
+
"eval_wer": 0.2107740264301708,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.89,
|
| 126 |
+
"learning_rate": 9.556367432150314e-05,
|
| 127 |
+
"loss": 0.0752,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.94,
|
| 132 |
+
"learning_rate": 9.530271398747391e-05,
|
| 133 |
+
"loss": 0.0814,
|
| 134 |
+
"step": 900
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.99,
|
| 138 |
+
"learning_rate": 9.504175365344468e-05,
|
| 139 |
+
"loss": 0.0747,
|
| 140 |
+
"step": 950
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.04,
|
| 144 |
+
"learning_rate": 9.478079331941545e-05,
|
| 145 |
+
"loss": 0.0686,
|
| 146 |
+
"step": 1000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.1,
|
| 150 |
+
"learning_rate": 9.451983298538623e-05,
|
| 151 |
+
"loss": 0.0761,
|
| 152 |
+
"step": 1050
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.15,
|
| 156 |
+
"learning_rate": 9.4258872651357e-05,
|
| 157 |
+
"loss": 0.0695,
|
| 158 |
+
"step": 1100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.2,
|
| 162 |
+
"learning_rate": 9.399791231732778e-05,
|
| 163 |
+
"loss": 0.0659,
|
| 164 |
+
"step": 1150
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.25,
|
| 168 |
+
"learning_rate": 9.373695198329853e-05,
|
| 169 |
+
"loss": 0.0765,
|
| 170 |
+
"step": 1200
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.25,
|
| 174 |
+
"eval_loss": 0.9291716814041138,
|
| 175 |
+
"eval_runtime": 2388.9228,
|
| 176 |
+
"eval_samples_per_second": 9.204,
|
| 177 |
+
"eval_steps_per_second": 1.151,
|
| 178 |
+
"eval_wer": 0.21025595995960839,
|
| 179 |
+
"step": 1200
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.3,
|
| 183 |
+
"learning_rate": 9.347599164926931e-05,
|
| 184 |
+
"loss": 0.0735,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.36,
|
| 189 |
+
"learning_rate": 9.321503131524008e-05,
|
| 190 |
+
"loss": 0.0734,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.41,
|
| 195 |
+
"learning_rate": 9.295407098121086e-05,
|
| 196 |
+
"loss": 0.0674,
|
| 197 |
+
"step": 1350
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.46,
|
| 201 |
+
"learning_rate": 9.269311064718163e-05,
|
| 202 |
+
"loss": 0.0669,
|
| 203 |
+
"step": 1400
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.51,
|
| 207 |
+
"learning_rate": 9.243215031315241e-05,
|
| 208 |
+
"loss": 0.0689,
|
| 209 |
+
"step": 1450
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.57,
|
| 213 |
+
"learning_rate": 9.217118997912317e-05,
|
| 214 |
+
"loss": 0.0702,
|
| 215 |
+
"step": 1500
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.62,
|
| 219 |
+
"learning_rate": 9.191022964509395e-05,
|
| 220 |
+
"loss": 0.0765,
|
| 221 |
+
"step": 1550
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.67,
|
| 225 |
+
"learning_rate": 9.164926931106472e-05,
|
| 226 |
+
"loss": 0.0757,
|
| 227 |
+
"step": 1600
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.67,
|
| 231 |
+
"eval_loss": 0.8958114385604858,
|
| 232 |
+
"eval_runtime": 2362.4608,
|
| 233 |
+
"eval_samples_per_second": 9.307,
|
| 234 |
+
"eval_steps_per_second": 1.164,
|
| 235 |
+
"eval_wer": 0.20942398033103568,
|
| 236 |
+
"step": 1600
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.72,
|
| 240 |
+
"learning_rate": 9.13883089770355e-05,
|
| 241 |
+
"loss": 0.0752,
|
| 242 |
+
"step": 1650
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.77,
|
| 246 |
+
"learning_rate": 9.112734864300627e-05,
|
| 247 |
+
"loss": 0.0725,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.83,
|
| 252 |
+
"learning_rate": 9.086638830897703e-05,
|
| 253 |
+
"loss": 0.0669,
|
| 254 |
+
"step": 1750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.88,
|
| 258 |
+
"learning_rate": 9.060542797494782e-05,
|
| 259 |
+
"loss": 0.0699,
|
| 260 |
+
"step": 1800
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.93,
|
| 264 |
+
"learning_rate": 9.034446764091858e-05,
|
| 265 |
+
"loss": 0.0723,
|
| 266 |
+
"step": 1850
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.98,
|
| 270 |
+
"learning_rate": 9.008350730688935e-05,
|
| 271 |
+
"loss": 0.0712,
|
| 272 |
+
"step": 1900
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.04,
|
| 276 |
+
"learning_rate": 8.982254697286013e-05,
|
| 277 |
+
"loss": 0.0728,
|
| 278 |
+
"step": 1950
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.09,
|
| 282 |
+
"learning_rate": 8.95615866388309e-05,
|
| 283 |
+
"loss": 0.0655,
|
| 284 |
+
"step": 2000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.09,
|
| 288 |
+
"eval_loss": 0.9281134605407715,
|
| 289 |
+
"eval_runtime": 2378.3708,
|
| 290 |
+
"eval_samples_per_second": 9.245,
|
| 291 |
+
"eval_steps_per_second": 1.156,
|
| 292 |
+
"eval_wer": 0.20970277033849935,
|
| 293 |
+
"step": 2000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.14,
|
| 297 |
+
"learning_rate": 8.930062630480167e-05,
|
| 298 |
+
"loss": 0.068,
|
| 299 |
+
"step": 2050
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.19,
|
| 303 |
+
"learning_rate": 8.903966597077245e-05,
|
| 304 |
+
"loss": 0.0645,
|
| 305 |
+
"step": 2100
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.24,
|
| 309 |
+
"learning_rate": 8.877870563674322e-05,
|
| 310 |
+
"loss": 0.0665,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.3,
|
| 315 |
+
"learning_rate": 8.8517745302714e-05,
|
| 316 |
+
"loss": 0.0648,
|
| 317 |
+
"step": 2200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.35,
|
| 321 |
+
"learning_rate": 8.825678496868477e-05,
|
| 322 |
+
"loss": 0.0629,
|
| 323 |
+
"step": 2250
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.4,
|
| 327 |
+
"learning_rate": 8.799582463465554e-05,
|
| 328 |
+
"loss": 0.0643,
|
| 329 |
+
"step": 2300
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.45,
|
| 333 |
+
"learning_rate": 8.77348643006263e-05,
|
| 334 |
+
"loss": 0.0601,
|
| 335 |
+
"step": 2350
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.5,
|
| 339 |
+
"learning_rate": 8.747390396659709e-05,
|
| 340 |
+
"loss": 0.0686,
|
| 341 |
+
"step": 2400
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 2.5,
|
| 345 |
+
"eval_loss": 0.9298621416091919,
|
| 346 |
+
"eval_runtime": 2386.5283,
|
| 347 |
+
"eval_samples_per_second": 9.213,
|
| 348 |
+
"eval_steps_per_second": 1.152,
|
| 349 |
+
"eval_wer": 0.2079092944637134,
|
| 350 |
+
"step": 2400
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.56,
|
| 354 |
+
"learning_rate": 8.721294363256785e-05,
|
| 355 |
+
"loss": 0.0637,
|
| 356 |
+
"step": 2450
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.61,
|
| 360 |
+
"learning_rate": 8.695198329853864e-05,
|
| 361 |
+
"loss": 0.0675,
|
| 362 |
+
"step": 2500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 2.66,
|
| 366 |
+
"learning_rate": 8.669102296450939e-05,
|
| 367 |
+
"loss": 0.0719,
|
| 368 |
+
"step": 2550
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.71,
|
| 372 |
+
"learning_rate": 8.643006263048017e-05,
|
| 373 |
+
"loss": 0.0709,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.77,
|
| 378 |
+
"learning_rate": 8.616910229645094e-05,
|
| 379 |
+
"loss": 0.0702,
|
| 380 |
+
"step": 2650
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.82,
|
| 384 |
+
"learning_rate": 8.590814196242172e-05,
|
| 385 |
+
"loss": 0.0645,
|
| 386 |
+
"step": 2700
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 2.87,
|
| 390 |
+
"learning_rate": 8.564718162839249e-05,
|
| 391 |
+
"loss": 0.0707,
|
| 392 |
+
"step": 2750
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.92,
|
| 396 |
+
"learning_rate": 8.538622129436326e-05,
|
| 397 |
+
"loss": 0.0644,
|
| 398 |
+
"step": 2800
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.92,
|
| 402 |
+
"eval_loss": 1.0021286010742188,
|
| 403 |
+
"eval_runtime": 2368.1778,
|
| 404 |
+
"eval_samples_per_second": 9.285,
|
| 405 |
+
"eval_steps_per_second": 1.161,
|
| 406 |
+
"eval_wer": 0.20971374632304518,
|
| 407 |
+
"step": 2800
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.97,
|
| 411 |
+
"learning_rate": 8.512526096033403e-05,
|
| 412 |
+
"loss": 0.0657,
|
| 413 |
+
"step": 2850
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.03,
|
| 417 |
+
"learning_rate": 8.486430062630481e-05,
|
| 418 |
+
"loss": 0.0721,
|
| 419 |
+
"step": 2900
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.08,
|
| 423 |
+
"learning_rate": 8.460334029227558e-05,
|
| 424 |
+
"loss": 0.067,
|
| 425 |
+
"step": 2950
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 3.13,
|
| 429 |
+
"learning_rate": 8.434237995824636e-05,
|
| 430 |
+
"loss": 0.0724,
|
| 431 |
+
"step": 3000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 3.18,
|
| 435 |
+
"learning_rate": 8.408141962421713e-05,
|
| 436 |
+
"loss": 0.0666,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 3.24,
|
| 441 |
+
"learning_rate": 8.38204592901879e-05,
|
| 442 |
+
"loss": 0.0618,
|
| 443 |
+
"step": 3100
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.29,
|
| 447 |
+
"learning_rate": 8.355949895615866e-05,
|
| 448 |
+
"loss": 0.0649,
|
| 449 |
+
"step": 3150
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.34,
|
| 453 |
+
"learning_rate": 8.329853862212944e-05,
|
| 454 |
+
"loss": 0.065,
|
| 455 |
+
"step": 3200
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 3.34,
|
| 459 |
+
"eval_loss": 0.9249215722084045,
|
| 460 |
+
"eval_runtime": 2321.5241,
|
| 461 |
+
"eval_samples_per_second": 9.471,
|
| 462 |
+
"eval_steps_per_second": 1.184,
|
| 463 |
+
"eval_wer": 0.2105676779207095,
|
| 464 |
+
"step": 3200
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.39,
|
| 468 |
+
"learning_rate": 8.303757828810021e-05,
|
| 469 |
+
"loss": 0.0586,
|
| 470 |
+
"step": 3250
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.44,
|
| 474 |
+
"learning_rate": 8.277661795407099e-05,
|
| 475 |
+
"loss": 0.0683,
|
| 476 |
+
"step": 3300
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.5,
|
| 480 |
+
"learning_rate": 8.251565762004175e-05,
|
| 481 |
+
"loss": 0.0645,
|
| 482 |
+
"step": 3350
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 3.55,
|
| 486 |
+
"learning_rate": 8.225469728601253e-05,
|
| 487 |
+
"loss": 0.0655,
|
| 488 |
+
"step": 3400
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.6,
|
| 492 |
+
"learning_rate": 8.19937369519833e-05,
|
| 493 |
+
"loss": 0.0634,
|
| 494 |
+
"step": 3450
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 3.65,
|
| 498 |
+
"learning_rate": 8.173277661795408e-05,
|
| 499 |
+
"loss": 0.0726,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 3.71,
|
| 504 |
+
"learning_rate": 8.147181628392485e-05,
|
| 505 |
+
"loss": 0.0676,
|
| 506 |
+
"step": 3550
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 3.76,
|
| 510 |
+
"learning_rate": 8.121085594989561e-05,
|
| 511 |
+
"loss": 0.066,
|
| 512 |
+
"step": 3600
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.76,
|
| 516 |
+
"eval_loss": 0.97150719165802,
|
| 517 |
+
"eval_runtime": 2307.8466,
|
| 518 |
+
"eval_samples_per_second": 9.527,
|
| 519 |
+
"eval_steps_per_second": 1.191,
|
| 520 |
+
"eval_wer": 0.20944812749703648,
|
| 521 |
+
"step": 3600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 3.81,
|
| 525 |
+
"learning_rate": 8.09498956158664e-05,
|
| 526 |
+
"loss": 0.0621,
|
| 527 |
+
"step": 3650
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 3.86,
|
| 531 |
+
"learning_rate": 8.068893528183716e-05,
|
| 532 |
+
"loss": 0.061,
|
| 533 |
+
"step": 3700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.91,
|
| 537 |
+
"learning_rate": 8.042797494780793e-05,
|
| 538 |
+
"loss": 0.0667,
|
| 539 |
+
"step": 3750
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 3.97,
|
| 543 |
+
"learning_rate": 8.016701461377871e-05,
|
| 544 |
+
"loss": 0.0635,
|
| 545 |
+
"step": 3800
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 4.02,
|
| 549 |
+
"learning_rate": 7.990605427974948e-05,
|
| 550 |
+
"loss": 0.0597,
|
| 551 |
+
"step": 3850
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 4.07,
|
| 555 |
+
"learning_rate": 7.964509394572025e-05,
|
| 556 |
+
"loss": 0.0583,
|
| 557 |
+
"step": 3900
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 4.12,
|
| 561 |
+
"learning_rate": 7.938413361169103e-05,
|
| 562 |
+
"loss": 0.0626,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 4.18,
|
| 567 |
+
"learning_rate": 7.91231732776618e-05,
|
| 568 |
+
"loss": 0.0629,
|
| 569 |
+
"step": 4000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 4.18,
|
| 573 |
+
"eval_loss": 1.020053505897522,
|
| 574 |
+
"eval_runtime": 2289.6794,
|
| 575 |
+
"eval_samples_per_second": 9.603,
|
| 576 |
+
"eval_steps_per_second": 1.201,
|
| 577 |
+
"eval_wer": 0.2082912587259077,
|
| 578 |
+
"step": 4000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 4.23,
|
| 582 |
+
"learning_rate": 7.886221294363258e-05,
|
| 583 |
+
"loss": 0.0615,
|
| 584 |
+
"step": 4050
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 4.28,
|
| 588 |
+
"learning_rate": 7.860125260960335e-05,
|
| 589 |
+
"loss": 0.0576,
|
| 590 |
+
"step": 4100
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 4.33,
|
| 594 |
+
"learning_rate": 7.834029227557412e-05,
|
| 595 |
+
"loss": 0.0633,
|
| 596 |
+
"step": 4150
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 4.38,
|
| 600 |
+
"learning_rate": 7.807933194154488e-05,
|
| 601 |
+
"loss": 0.0596,
|
| 602 |
+
"step": 4200
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 4.44,
|
| 606 |
+
"learning_rate": 7.781837160751567e-05,
|
| 607 |
+
"loss": 0.0637,
|
| 608 |
+
"step": 4250
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 4.49,
|
| 612 |
+
"learning_rate": 7.755741127348643e-05,
|
| 613 |
+
"loss": 0.0627,
|
| 614 |
+
"step": 4300
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 4.54,
|
| 618 |
+
"learning_rate": 7.729645093945722e-05,
|
| 619 |
+
"loss": 0.0581,
|
| 620 |
+
"step": 4350
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 4.59,
|
| 624 |
+
"learning_rate": 7.703549060542797e-05,
|
| 625 |
+
"loss": 0.056,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 4.59,
|
| 630 |
+
"eval_loss": 1.0113539695739746,
|
| 631 |
+
"eval_runtime": 2285.7115,
|
| 632 |
+
"eval_samples_per_second": 9.62,
|
| 633 |
+
"eval_steps_per_second": 1.203,
|
| 634 |
+
"eval_wer": 0.20872371251701277,
|
| 635 |
+
"step": 4400
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 4.64,
|
| 639 |
+
"learning_rate": 7.677974947807933e-05,
|
| 640 |
+
"loss": 0.0606,
|
| 641 |
+
"step": 4450
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 4.7,
|
| 645 |
+
"learning_rate": 7.65187891440501e-05,
|
| 646 |
+
"loss": 0.0649,
|
| 647 |
+
"step": 4500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 4.75,
|
| 651 |
+
"learning_rate": 7.625782881002088e-05,
|
| 652 |
+
"loss": 0.0601,
|
| 653 |
+
"step": 4550
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 4.8,
|
| 657 |
+
"learning_rate": 7.599686847599165e-05,
|
| 658 |
+
"loss": 0.0643,
|
| 659 |
+
"step": 4600
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 4.85,
|
| 663 |
+
"learning_rate": 7.573590814196243e-05,
|
| 664 |
+
"loss": 0.0628,
|
| 665 |
+
"step": 4650
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 4.91,
|
| 669 |
+
"learning_rate": 7.547494780793319e-05,
|
| 670 |
+
"loss": 0.0559,
|
| 671 |
+
"step": 4700
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 4.96,
|
| 675 |
+
"learning_rate": 7.521398747390397e-05,
|
| 676 |
+
"loss": 0.0652,
|
| 677 |
+
"step": 4750
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.01,
|
| 681 |
+
"learning_rate": 7.495302713987474e-05,
|
| 682 |
+
"loss": 0.0642,
|
| 683 |
+
"step": 4800
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.01,
|
| 687 |
+
"eval_loss": 0.962189793586731,
|
| 688 |
+
"eval_runtime": 2298.4209,
|
| 689 |
+
"eval_samples_per_second": 9.567,
|
| 690 |
+
"eval_steps_per_second": 1.196,
|
| 691 |
+
"eval_wer": 0.20767001800061466,
|
| 692 |
+
"step": 4800
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 5.06,
|
| 696 |
+
"learning_rate": 7.469206680584552e-05,
|
| 697 |
+
"loss": 0.0624,
|
| 698 |
+
"step": 4850
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 5.11,
|
| 702 |
+
"learning_rate": 7.443110647181629e-05,
|
| 703 |
+
"loss": 0.0524,
|
| 704 |
+
"step": 4900
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 5.17,
|
| 708 |
+
"learning_rate": 7.417014613778706e-05,
|
| 709 |
+
"loss": 0.0578,
|
| 710 |
+
"step": 4950
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 5.22,
|
| 714 |
+
"learning_rate": 7.390918580375782e-05,
|
| 715 |
+
"loss": 0.0649,
|
| 716 |
+
"step": 5000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 5.27,
|
| 720 |
+
"learning_rate": 7.36482254697286e-05,
|
| 721 |
+
"loss": 0.0628,
|
| 722 |
+
"step": 5050
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 5.32,
|
| 726 |
+
"learning_rate": 7.338726513569937e-05,
|
| 727 |
+
"loss": 0.0619,
|
| 728 |
+
"step": 5100
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 5.38,
|
| 732 |
+
"learning_rate": 7.312630480167015e-05,
|
| 733 |
+
"loss": 0.0549,
|
| 734 |
+
"step": 5150
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 5.43,
|
| 738 |
+
"learning_rate": 7.286534446764092e-05,
|
| 739 |
+
"loss": 0.0621,
|
| 740 |
+
"step": 5200
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 5.43,
|
| 744 |
+
"eval_loss": 1.0184942483901978,
|
| 745 |
+
"eval_runtime": 2291.4008,
|
| 746 |
+
"eval_samples_per_second": 9.596,
|
| 747 |
+
"eval_steps_per_second": 1.2,
|
| 748 |
+
"eval_wer": 0.20862492865610044,
|
| 749 |
+
"step": 5200
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 5.48,
|
| 753 |
+
"learning_rate": 7.260438413361169e-05,
|
| 754 |
+
"loss": 0.0536,
|
| 755 |
+
"step": 5250
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 5.53,
|
| 759 |
+
"learning_rate": 7.234342379958247e-05,
|
| 760 |
+
"loss": 0.0567,
|
| 761 |
+
"step": 5300
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 5.58,
|
| 765 |
+
"learning_rate": 7.208246346555324e-05,
|
| 766 |
+
"loss": 0.0595,
|
| 767 |
+
"step": 5350
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 5.64,
|
| 771 |
+
"learning_rate": 7.182150313152402e-05,
|
| 772 |
+
"loss": 0.0582,
|
| 773 |
+
"step": 5400
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 5.69,
|
| 777 |
+
"learning_rate": 7.156054279749479e-05,
|
| 778 |
+
"loss": 0.0578,
|
| 779 |
+
"step": 5450
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 5.74,
|
| 783 |
+
"learning_rate": 7.129958246346556e-05,
|
| 784 |
+
"loss": 0.0623,
|
| 785 |
+
"step": 5500
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 5.79,
|
| 789 |
+
"learning_rate": 7.103862212943633e-05,
|
| 790 |
+
"loss": 0.0547,
|
| 791 |
+
"step": 5550
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 5.85,
|
| 795 |
+
"learning_rate": 7.077766179540711e-05,
|
| 796 |
+
"loss": 0.0579,
|
| 797 |
+
"step": 5600
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 5.85,
|
| 801 |
+
"eval_loss": 1.0210742950439453,
|
| 802 |
+
"eval_runtime": 2297.4069,
|
| 803 |
+
"eval_samples_per_second": 9.571,
|
| 804 |
+
"eval_steps_per_second": 1.197,
|
| 805 |
+
"eval_wer": 0.2089103042542916,
|
| 806 |
+
"step": 5600
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 5.9,
|
| 810 |
+
"learning_rate": 7.051670146137788e-05,
|
| 811 |
+
"loss": 0.06,
|
| 812 |
+
"step": 5650
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 5.95,
|
| 816 |
+
"learning_rate": 7.025574112734866e-05,
|
| 817 |
+
"loss": 0.0559,
|
| 818 |
+
"step": 5700
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.0,
|
| 822 |
+
"learning_rate": 6.999478079331941e-05,
|
| 823 |
+
"loss": 0.0618,
|
| 824 |
+
"step": 5750
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 6.05,
|
| 828 |
+
"learning_rate": 6.973382045929019e-05,
|
| 829 |
+
"loss": 0.057,
|
| 830 |
+
"step": 5800
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 6.11,
|
| 834 |
+
"learning_rate": 6.947286012526096e-05,
|
| 835 |
+
"loss": 0.0611,
|
| 836 |
+
"step": 5850
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 6.16,
|
| 840 |
+
"learning_rate": 6.921189979123174e-05,
|
| 841 |
+
"loss": 0.0585,
|
| 842 |
+
"step": 5900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 6.21,
|
| 846 |
+
"learning_rate": 6.895093945720251e-05,
|
| 847 |
+
"loss": 0.051,
|
| 848 |
+
"step": 5950
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 6.26,
|
| 852 |
+
"learning_rate": 6.868997912317328e-05,
|
| 853 |
+
"loss": 0.055,
|
| 854 |
+
"step": 6000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 6.26,
|
| 858 |
+
"eval_loss": 1.0093300342559814,
|
| 859 |
+
"eval_runtime": 2290.7926,
|
| 860 |
+
"eval_samples_per_second": 9.598,
|
| 861 |
+
"eval_steps_per_second": 1.2,
|
| 862 |
+
"eval_wer": 0.20797734556789743,
|
| 863 |
+
"step": 6000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 6.32,
|
| 867 |
+
"learning_rate": 6.842901878914405e-05,
|
| 868 |
+
"loss": 0.0519,
|
| 869 |
+
"step": 6050
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.37,
|
| 873 |
+
"learning_rate": 6.816805845511483e-05,
|
| 874 |
+
"loss": 0.056,
|
| 875 |
+
"step": 6100
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 6.42,
|
| 879 |
+
"learning_rate": 6.79070981210856e-05,
|
| 880 |
+
"loss": 0.0584,
|
| 881 |
+
"step": 6150
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 6.47,
|
| 885 |
+
"learning_rate": 6.764613778705638e-05,
|
| 886 |
+
"loss": 0.061,
|
| 887 |
+
"step": 6200
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 6.52,
|
| 891 |
+
"learning_rate": 6.738517745302715e-05,
|
| 892 |
+
"loss": 0.0547,
|
| 893 |
+
"step": 6250
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 6.58,
|
| 897 |
+
"learning_rate": 6.712421711899791e-05,
|
| 898 |
+
"loss": 0.0593,
|
| 899 |
+
"step": 6300
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 6.63,
|
| 903 |
+
"learning_rate": 6.686325678496868e-05,
|
| 904 |
+
"loss": 0.0592,
|
| 905 |
+
"step": 6350
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 6.68,
|
| 909 |
+
"learning_rate": 6.660229645093946e-05,
|
| 910 |
+
"loss": 0.0575,
|
| 911 |
+
"step": 6400
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 6.68,
|
| 915 |
+
"eval_loss": 0.9511479139328003,
|
| 916 |
+
"eval_runtime": 2297.8339,
|
| 917 |
+
"eval_samples_per_second": 9.569,
|
| 918 |
+
"eval_steps_per_second": 1.196,
|
| 919 |
+
"eval_wer": 0.20716951310532555,
|
| 920 |
+
"step": 6400
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 6.73,
|
| 924 |
+
"learning_rate": 6.634133611691023e-05,
|
| 925 |
+
"loss": 0.0616,
|
| 926 |
+
"step": 6450
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 6.78,
|
| 930 |
+
"learning_rate": 6.608037578288101e-05,
|
| 931 |
+
"loss": 0.0569,
|
| 932 |
+
"step": 6500
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 6.84,
|
| 936 |
+
"learning_rate": 6.581941544885177e-05,
|
| 937 |
+
"loss": 0.0559,
|
| 938 |
+
"step": 6550
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 6.89,
|
| 942 |
+
"learning_rate": 6.555845511482255e-05,
|
| 943 |
+
"loss": 0.058,
|
| 944 |
+
"step": 6600
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 6.94,
|
| 948 |
+
"learning_rate": 6.529749478079332e-05,
|
| 949 |
+
"loss": 0.0584,
|
| 950 |
+
"step": 6650
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 6.99,
|
| 954 |
+
"learning_rate": 6.50365344467641e-05,
|
| 955 |
+
"loss": 0.0589,
|
| 956 |
+
"step": 6700
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 7.05,
|
| 960 |
+
"learning_rate": 6.477557411273487e-05,
|
| 961 |
+
"loss": 0.0521,
|
| 962 |
+
"step": 6750
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 7.1,
|
| 966 |
+
"learning_rate": 6.451461377870563e-05,
|
| 967 |
+
"loss": 0.0537,
|
| 968 |
+
"step": 6800
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 7.1,
|
| 972 |
+
"eval_loss": 1.0872694253921509,
|
| 973 |
+
"eval_runtime": 2286.4716,
|
| 974 |
+
"eval_samples_per_second": 9.617,
|
| 975 |
+
"eval_steps_per_second": 1.202,
|
| 976 |
+
"eval_wer": 0.20757123413970233,
|
| 977 |
+
"step": 6800
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 7.15,
|
| 981 |
+
"learning_rate": 6.42536534446764e-05,
|
| 982 |
+
"loss": 0.0544,
|
| 983 |
+
"step": 6850
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 7.2,
|
| 987 |
+
"learning_rate": 6.399269311064718e-05,
|
| 988 |
+
"loss": 0.0545,
|
| 989 |
+
"step": 6900
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 7.25,
|
| 993 |
+
"learning_rate": 6.373173277661795e-05,
|
| 994 |
+
"loss": 0.0562,
|
| 995 |
+
"step": 6950
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 7.31,
|
| 999 |
+
"learning_rate": 6.347077244258873e-05,
|
| 1000 |
+
"loss": 0.0564,
|
| 1001 |
+
"step": 7000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 7.36,
|
| 1005 |
+
"learning_rate": 6.32098121085595e-05,
|
| 1006 |
+
"loss": 0.0528,
|
| 1007 |
+
"step": 7050
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 7.41,
|
| 1011 |
+
"learning_rate": 6.294885177453027e-05,
|
| 1012 |
+
"loss": 0.0588,
|
| 1013 |
+
"step": 7100
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 7.46,
|
| 1017 |
+
"learning_rate": 6.268789144050105e-05,
|
| 1018 |
+
"loss": 0.051,
|
| 1019 |
+
"step": 7150
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 7.52,
|
| 1023 |
+
"learning_rate": 6.242693110647182e-05,
|
| 1024 |
+
"loss": 0.0549,
|
| 1025 |
+
"step": 7200
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 7.52,
|
| 1029 |
+
"eval_loss": 1.0469272136688232,
|
| 1030 |
+
"eval_runtime": 2282.6191,
|
| 1031 |
+
"eval_samples_per_second": 9.633,
|
| 1032 |
+
"eval_steps_per_second": 1.204,
|
| 1033 |
+
"eval_wer": 0.20716951310532555,
|
| 1034 |
+
"step": 7200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 7.57,
|
| 1038 |
+
"learning_rate": 6.216597077244259e-05,
|
| 1039 |
+
"loss": 0.0547,
|
| 1040 |
+
"step": 7250
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 7.62,
|
| 1044 |
+
"learning_rate": 6.190501043841337e-05,
|
| 1045 |
+
"loss": 0.0577,
|
| 1046 |
+
"step": 7300
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 7.67,
|
| 1050 |
+
"learning_rate": 6.164405010438414e-05,
|
| 1051 |
+
"loss": 0.0553,
|
| 1052 |
+
"step": 7350
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 7.72,
|
| 1056 |
+
"learning_rate": 6.13830897703549e-05,
|
| 1057 |
+
"loss": 0.058,
|
| 1058 |
+
"step": 7400
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 7.78,
|
| 1062 |
+
"learning_rate": 6.112212943632569e-05,
|
| 1063 |
+
"loss": 0.056,
|
| 1064 |
+
"step": 7450
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 7.83,
|
| 1068 |
+
"learning_rate": 6.0861169102296455e-05,
|
| 1069 |
+
"loss": 0.0512,
|
| 1070 |
+
"step": 7500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 7.88,
|
| 1074 |
+
"learning_rate": 6.060020876826723e-05,
|
| 1075 |
+
"loss": 0.0587,
|
| 1076 |
+
"step": 7550
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 7.93,
|
| 1080 |
+
"learning_rate": 6.033924843423799e-05,
|
| 1081 |
+
"loss": 0.0559,
|
| 1082 |
+
"step": 7600
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 7.93,
|
| 1086 |
+
"eval_loss": 0.9638180732727051,
|
| 1087 |
+
"eval_runtime": 2284.7108,
|
| 1088 |
+
"eval_samples_per_second": 9.624,
|
| 1089 |
+
"eval_steps_per_second": 1.203,
|
| 1090 |
+
"eval_wer": 0.20783026737498353,
|
| 1091 |
+
"step": 7600
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 7.98,
|
| 1095 |
+
"learning_rate": 6.0078288100208766e-05,
|
| 1096 |
+
"loss": 0.0597,
|
| 1097 |
+
"step": 7650
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 8.04,
|
| 1101 |
+
"learning_rate": 5.981732776617954e-05,
|
| 1102 |
+
"loss": 0.0531,
|
| 1103 |
+
"step": 7700
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 8.09,
|
| 1107 |
+
"learning_rate": 5.9556367432150315e-05,
|
| 1108 |
+
"loss": 0.0528,
|
| 1109 |
+
"step": 7750
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 8.14,
|
| 1113 |
+
"learning_rate": 5.929540709812109e-05,
|
| 1114 |
+
"loss": 0.0551,
|
| 1115 |
+
"step": 7800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 8.19,
|
| 1119 |
+
"learning_rate": 5.9034446764091865e-05,
|
| 1120 |
+
"loss": 0.0507,
|
| 1121 |
+
"step": 7850
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 8.25,
|
| 1125 |
+
"learning_rate": 5.877348643006263e-05,
|
| 1126 |
+
"loss": 0.0518,
|
| 1127 |
+
"step": 7900
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 8.3,
|
| 1131 |
+
"learning_rate": 5.85125260960334e-05,
|
| 1132 |
+
"loss": 0.0534,
|
| 1133 |
+
"step": 7950
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 8.35,
|
| 1137 |
+
"learning_rate": 5.8251565762004176e-05,
|
| 1138 |
+
"loss": 0.056,
|
| 1139 |
+
"step": 8000
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 8.35,
|
| 1143 |
+
"eval_loss": 1.0094764232635498,
|
| 1144 |
+
"eval_runtime": 2282.77,
|
| 1145 |
+
"eval_samples_per_second": 9.632,
|
| 1146 |
+
"eval_steps_per_second": 1.204,
|
| 1147 |
+
"eval_wer": 0.20796417438644246,
|
| 1148 |
+
"step": 8000
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 8.4,
|
| 1152 |
+
"learning_rate": 5.799060542797495e-05,
|
| 1153 |
+
"loss": 0.0519,
|
| 1154 |
+
"step": 8050
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 8.45,
|
| 1158 |
+
"learning_rate": 5.7729645093945725e-05,
|
| 1159 |
+
"loss": 0.0521,
|
| 1160 |
+
"step": 8100
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 8.51,
|
| 1164 |
+
"learning_rate": 5.746868475991649e-05,
|
| 1165 |
+
"loss": 0.0527,
|
| 1166 |
+
"step": 8150
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 8.56,
|
| 1170 |
+
"learning_rate": 5.720772442588727e-05,
|
| 1171 |
+
"loss": 0.0553,
|
| 1172 |
+
"step": 8200
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 8.61,
|
| 1176 |
+
"learning_rate": 5.694676409185804e-05,
|
| 1177 |
+
"loss": 0.0559,
|
| 1178 |
+
"step": 8250
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 8.66,
|
| 1182 |
+
"learning_rate": 5.668580375782882e-05,
|
| 1183 |
+
"loss": 0.0571,
|
| 1184 |
+
"step": 8300
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 8.72,
|
| 1188 |
+
"learning_rate": 5.642484342379959e-05,
|
| 1189 |
+
"loss": 0.0517,
|
| 1190 |
+
"step": 8350
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 8.77,
|
| 1194 |
+
"learning_rate": 5.6163883089770354e-05,
|
| 1195 |
+
"loss": 0.0537,
|
| 1196 |
+
"step": 8400
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 8.77,
|
| 1200 |
+
"eval_loss": 1.101858139038086,
|
| 1201 |
+
"eval_runtime": 2292.137,
|
| 1202 |
+
"eval_samples_per_second": 9.593,
|
| 1203 |
+
"eval_steps_per_second": 1.199,
|
| 1204 |
+
"eval_wer": 0.20929007331957677,
|
| 1205 |
+
"step": 8400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 8.82,
|
| 1209 |
+
"learning_rate": 5.590292275574113e-05,
|
| 1210 |
+
"loss": 0.0544,
|
| 1211 |
+
"step": 8450
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 8.87,
|
| 1215 |
+
"learning_rate": 5.5641962421711903e-05,
|
| 1216 |
+
"loss": 0.0519,
|
| 1217 |
+
"step": 8500
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 8.92,
|
| 1221 |
+
"learning_rate": 5.538100208768268e-05,
|
| 1222 |
+
"loss": 0.0542,
|
| 1223 |
+
"step": 8550
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 8.98,
|
| 1227 |
+
"learning_rate": 5.512004175365345e-05,
|
| 1228 |
+
"loss": 0.0494,
|
| 1229 |
+
"step": 8600
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 9.03,
|
| 1233 |
+
"learning_rate": 5.4859081419624214e-05,
|
| 1234 |
+
"loss": 0.0481,
|
| 1235 |
+
"step": 8650
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 9.08,
|
| 1239 |
+
"learning_rate": 5.459812108559499e-05,
|
| 1240 |
+
"loss": 0.0554,
|
| 1241 |
+
"step": 8700
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 9.13,
|
| 1245 |
+
"learning_rate": 5.4342379958246346e-05,
|
| 1246 |
+
"loss": 0.0515,
|
| 1247 |
+
"step": 8750
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 9.19,
|
| 1251 |
+
"learning_rate": 5.408141962421712e-05,
|
| 1252 |
+
"loss": 0.0497,
|
| 1253 |
+
"step": 8800
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 9.19,
|
| 1257 |
+
"eval_loss": 1.110959529876709,
|
| 1258 |
+
"eval_runtime": 2293.7481,
|
| 1259 |
+
"eval_samples_per_second": 9.586,
|
| 1260 |
+
"eval_steps_per_second": 1.198,
|
| 1261 |
+
"eval_wer": 0.20725073539096456,
|
| 1262 |
+
"step": 8800
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 9.24,
|
| 1266 |
+
"learning_rate": 5.3820459290187896e-05,
|
| 1267 |
+
"loss": 0.0542,
|
| 1268 |
+
"step": 8850
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 9.29,
|
| 1272 |
+
"learning_rate": 5.355949895615867e-05,
|
| 1273 |
+
"loss": 0.0564,
|
| 1274 |
+
"step": 8900
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 9.34,
|
| 1278 |
+
"learning_rate": 5.329853862212943e-05,
|
| 1279 |
+
"loss": 0.0532,
|
| 1280 |
+
"step": 8950
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 9.39,
|
| 1284 |
+
"learning_rate": 5.303757828810021e-05,
|
| 1285 |
+
"loss": 0.0464,
|
| 1286 |
+
"step": 9000
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 9.45,
|
| 1290 |
+
"learning_rate": 5.277661795407098e-05,
|
| 1291 |
+
"loss": 0.0509,
|
| 1292 |
+
"step": 9050
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 9.5,
|
| 1296 |
+
"learning_rate": 5.2515657620041756e-05,
|
| 1297 |
+
"loss": 0.0507,
|
| 1298 |
+
"step": 9100
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 9.55,
|
| 1302 |
+
"learning_rate": 5.225469728601253e-05,
|
| 1303 |
+
"loss": 0.0541,
|
| 1304 |
+
"step": 9150
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 9.6,
|
| 1308 |
+
"learning_rate": 5.19937369519833e-05,
|
| 1309 |
+
"loss": 0.0509,
|
| 1310 |
+
"step": 9200
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 9.6,
|
| 1314 |
+
"eval_loss": 0.994656503200531,
|
| 1315 |
+
"eval_runtime": 2296.169,
|
| 1316 |
+
"eval_samples_per_second": 9.576,
|
| 1317 |
+
"eval_steps_per_second": 1.197,
|
| 1318 |
+
"eval_wer": 0.20686877112877025,
|
| 1319 |
+
"step": 9200
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 9.66,
|
| 1323 |
+
"learning_rate": 5.1732776617954074e-05,
|
| 1324 |
+
"loss": 0.0525,
|
| 1325 |
+
"step": 9250
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 9.71,
|
| 1329 |
+
"learning_rate": 5.147181628392484e-05,
|
| 1330 |
+
"loss": 0.0517,
|
| 1331 |
+
"step": 9300
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 9.76,
|
| 1335 |
+
"learning_rate": 5.121085594989562e-05,
|
| 1336 |
+
"loss": 0.0476,
|
| 1337 |
+
"step": 9350
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 9.81,
|
| 1341 |
+
"learning_rate": 5.094989561586639e-05,
|
| 1342 |
+
"loss": 0.0517,
|
| 1343 |
+
"step": 9400
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 9.86,
|
| 1347 |
+
"learning_rate": 5.068893528183716e-05,
|
| 1348 |
+
"loss": 0.0453,
|
| 1349 |
+
"step": 9450
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 9.92,
|
| 1353 |
+
"learning_rate": 5.0427974947807934e-05,
|
| 1354 |
+
"loss": 0.048,
|
| 1355 |
+
"step": 9500
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 9.97,
|
| 1359 |
+
"learning_rate": 5.016701461377871e-05,
|
| 1360 |
+
"loss": 0.0475,
|
| 1361 |
+
"step": 9550
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 10.02,
|
| 1365 |
+
"learning_rate": 4.9906054279749484e-05,
|
| 1366 |
+
"loss": 0.05,
|
| 1367 |
+
"step": 9600
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 10.02,
|
| 1371 |
+
"eval_loss": 1.0523179769515991,
|
| 1372 |
+
"eval_runtime": 2296.5421,
|
| 1373 |
+
"eval_samples_per_second": 9.574,
|
| 1374 |
+
"eval_steps_per_second": 1.197,
|
| 1375 |
+
"eval_wer": 0.20557360495236424,
|
| 1376 |
+
"step": 9600
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 10.07,
|
| 1380 |
+
"learning_rate": 4.964509394572025e-05,
|
| 1381 |
+
"loss": 0.0452,
|
| 1382 |
+
"step": 9650
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 10.13,
|
| 1386 |
+
"learning_rate": 4.938413361169103e-05,
|
| 1387 |
+
"loss": 0.0477,
|
| 1388 |
+
"step": 9700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 10.18,
|
| 1392 |
+
"learning_rate": 4.91231732776618e-05,
|
| 1393 |
+
"loss": 0.05,
|
| 1394 |
+
"step": 9750
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 10.23,
|
| 1398 |
+
"learning_rate": 4.886221294363257e-05,
|
| 1399 |
+
"loss": 0.0426,
|
| 1400 |
+
"step": 9800
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 10.28,
|
| 1404 |
+
"learning_rate": 4.8601252609603344e-05,
|
| 1405 |
+
"loss": 0.0462,
|
| 1406 |
+
"step": 9850
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 10.33,
|
| 1410 |
+
"learning_rate": 4.834029227557412e-05,
|
| 1411 |
+
"loss": 0.0517,
|
| 1412 |
+
"step": 9900
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 10.39,
|
| 1416 |
+
"learning_rate": 4.807933194154489e-05,
|
| 1417 |
+
"loss": 0.0483,
|
| 1418 |
+
"step": 9950
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 10.44,
|
| 1422 |
+
"learning_rate": 4.781837160751566e-05,
|
| 1423 |
+
"loss": 0.0441,
|
| 1424 |
+
"step": 10000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 10.44,
|
| 1428 |
+
"eval_loss": 1.0726261138916016,
|
| 1429 |
+
"eval_runtime": 2296.7153,
|
| 1430 |
+
"eval_samples_per_second": 9.574,
|
| 1431 |
+
"eval_steps_per_second": 1.197,
|
| 1432 |
+
"eval_wer": 0.20558897133072837,
|
| 1433 |
+
"step": 10000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 10.49,
|
| 1437 |
+
"learning_rate": 4.755741127348643e-05,
|
| 1438 |
+
"loss": 0.05,
|
| 1439 |
+
"step": 10050
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 10.54,
|
| 1443 |
+
"learning_rate": 4.7296450939457205e-05,
|
| 1444 |
+
"loss": 0.0494,
|
| 1445 |
+
"step": 10100
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 10.59,
|
| 1449 |
+
"learning_rate": 4.703549060542798e-05,
|
| 1450 |
+
"loss": 0.0485,
|
| 1451 |
+
"step": 10150
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 10.65,
|
| 1455 |
+
"learning_rate": 4.677453027139875e-05,
|
| 1456 |
+
"loss": 0.049,
|
| 1457 |
+
"step": 10200
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 10.7,
|
| 1461 |
+
"learning_rate": 4.651356993736952e-05,
|
| 1462 |
+
"loss": 0.0466,
|
| 1463 |
+
"step": 10250
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 10.75,
|
| 1467 |
+
"learning_rate": 4.62526096033403e-05,
|
| 1468 |
+
"loss": 0.0472,
|
| 1469 |
+
"step": 10300
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 10.8,
|
| 1473 |
+
"learning_rate": 4.5991649269311065e-05,
|
| 1474 |
+
"loss": 0.0544,
|
| 1475 |
+
"step": 10350
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 10.86,
|
| 1479 |
+
"learning_rate": 4.573068893528184e-05,
|
| 1480 |
+
"loss": 0.0456,
|
| 1481 |
+
"step": 10400
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 10.86,
|
| 1485 |
+
"eval_loss": 1.0436075925827026,
|
| 1486 |
+
"eval_runtime": 2285.586,
|
| 1487 |
+
"eval_samples_per_second": 9.62,
|
| 1488 |
+
"eval_steps_per_second": 1.203,
|
| 1489 |
+
"eval_wer": 0.20674584010185715,
|
| 1490 |
+
"step": 10400
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 10.91,
|
| 1494 |
+
"learning_rate": 4.546972860125261e-05,
|
| 1495 |
+
"loss": 0.0494,
|
| 1496 |
+
"step": 10450
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 10.96,
|
| 1500 |
+
"learning_rate": 4.520876826722338e-05,
|
| 1501 |
+
"loss": 0.0496,
|
| 1502 |
+
"step": 10500
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 11.01,
|
| 1506 |
+
"learning_rate": 4.494780793319416e-05,
|
| 1507 |
+
"loss": 0.0548,
|
| 1508 |
+
"step": 10550
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 11.06,
|
| 1512 |
+
"learning_rate": 4.4686847599164926e-05,
|
| 1513 |
+
"loss": 0.045,
|
| 1514 |
+
"step": 10600
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 11.12,
|
| 1518 |
+
"learning_rate": 4.44258872651357e-05,
|
| 1519 |
+
"loss": 0.0454,
|
| 1520 |
+
"step": 10650
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 11.17,
|
| 1524 |
+
"learning_rate": 4.417014613778706e-05,
|
| 1525 |
+
"loss": 0.047,
|
| 1526 |
+
"step": 10700
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 11.22,
|
| 1530 |
+
"learning_rate": 4.390918580375783e-05,
|
| 1531 |
+
"loss": 0.0457,
|
| 1532 |
+
"step": 10750
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 11.27,
|
| 1536 |
+
"learning_rate": 4.364822546972861e-05,
|
| 1537 |
+
"loss": 0.0469,
|
| 1538 |
+
"step": 10800
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 11.27,
|
| 1542 |
+
"eval_loss": 1.102227807044983,
|
| 1543 |
+
"eval_runtime": 2288.4521,
|
| 1544 |
+
"eval_samples_per_second": 9.608,
|
| 1545 |
+
"eval_steps_per_second": 1.201,
|
| 1546 |
+
"eval_wer": 0.20656144356148748,
|
| 1547 |
+
"step": 10800
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 11.33,
|
| 1551 |
+
"learning_rate": 4.3387265135699376e-05,
|
| 1552 |
+
"loss": 0.0487,
|
| 1553 |
+
"step": 10850
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 11.38,
|
| 1557 |
+
"learning_rate": 4.312630480167015e-05,
|
| 1558 |
+
"loss": 0.0479,
|
| 1559 |
+
"step": 10900
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 11.43,
|
| 1563 |
+
"learning_rate": 4.2865344467640925e-05,
|
| 1564 |
+
"loss": 0.0476,
|
| 1565 |
+
"step": 10950
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 11.48,
|
| 1569 |
+
"learning_rate": 4.260438413361169e-05,
|
| 1570 |
+
"loss": 0.0497,
|
| 1571 |
+
"step": 11000
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 11.53,
|
| 1575 |
+
"learning_rate": 4.234342379958247e-05,
|
| 1576 |
+
"loss": 0.0487,
|
| 1577 |
+
"step": 11050
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 11.59,
|
| 1581 |
+
"learning_rate": 4.208246346555324e-05,
|
| 1582 |
+
"loss": 0.0478,
|
| 1583 |
+
"step": 11100
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 11.64,
|
| 1587 |
+
"learning_rate": 4.182150313152401e-05,
|
| 1588 |
+
"loss": 0.0515,
|
| 1589 |
+
"step": 11150
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 11.69,
|
| 1593 |
+
"learning_rate": 4.1560542797494786e-05,
|
| 1594 |
+
"loss": 0.0549,
|
| 1595 |
+
"step": 11200
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 11.69,
|
| 1599 |
+
"eval_loss": 1.093299150466919,
|
| 1600 |
+
"eval_runtime": 2284.3046,
|
| 1601 |
+
"eval_samples_per_second": 9.626,
|
| 1602 |
+
"eval_steps_per_second": 1.203,
|
| 1603 |
+
"eval_wer": 0.20669096017912808,
|
| 1604 |
+
"step": 11200
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 11.74,
|
| 1608 |
+
"learning_rate": 4.1299582463465554e-05,
|
| 1609 |
+
"loss": 0.0465,
|
| 1610 |
+
"step": 11250
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 11.79,
|
| 1614 |
+
"learning_rate": 4.103862212943633e-05,
|
| 1615 |
+
"loss": 0.0426,
|
| 1616 |
+
"step": 11300
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 11.85,
|
| 1620 |
+
"learning_rate": 4.07776617954071e-05,
|
| 1621 |
+
"loss": 0.047,
|
| 1622 |
+
"step": 11350
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 11.9,
|
| 1626 |
+
"learning_rate": 4.051670146137787e-05,
|
| 1627 |
+
"loss": 0.0492,
|
| 1628 |
+
"step": 11400
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 11.95,
|
| 1632 |
+
"learning_rate": 4.0255741127348646e-05,
|
| 1633 |
+
"loss": 0.0506,
|
| 1634 |
+
"step": 11450
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 12.0,
|
| 1638 |
+
"learning_rate": 3.999478079331942e-05,
|
| 1639 |
+
"loss": 0.0467,
|
| 1640 |
+
"step": 11500
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 12.06,
|
| 1644 |
+
"learning_rate": 3.973382045929019e-05,
|
| 1645 |
+
"loss": 0.0501,
|
| 1646 |
+
"step": 11550
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 12.11,
|
| 1650 |
+
"learning_rate": 3.9472860125260964e-05,
|
| 1651 |
+
"loss": 0.0452,
|
| 1652 |
+
"step": 11600
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 12.11,
|
| 1656 |
+
"eval_loss": 1.0941613912582397,
|
| 1657 |
+
"eval_runtime": 2283.0889,
|
| 1658 |
+
"eval_samples_per_second": 9.631,
|
| 1659 |
+
"eval_steps_per_second": 1.204,
|
| 1660 |
+
"eval_wer": 0.20626070158493218,
|
| 1661 |
+
"step": 11600
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 12.16,
|
| 1665 |
+
"learning_rate": 3.921189979123173e-05,
|
| 1666 |
+
"loss": 0.051,
|
| 1667 |
+
"step": 11650
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 12.21,
|
| 1671 |
+
"learning_rate": 3.8950939457202506e-05,
|
| 1672 |
+
"loss": 0.0453,
|
| 1673 |
+
"step": 11700
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 12.26,
|
| 1677 |
+
"learning_rate": 3.868997912317328e-05,
|
| 1678 |
+
"loss": 0.0458,
|
| 1679 |
+
"step": 11750
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 12.32,
|
| 1683 |
+
"learning_rate": 3.842901878914405e-05,
|
| 1684 |
+
"loss": 0.0462,
|
| 1685 |
+
"step": 11800
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 12.37,
|
| 1689 |
+
"learning_rate": 3.8168058455114824e-05,
|
| 1690 |
+
"loss": 0.0466,
|
| 1691 |
+
"step": 11850
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 12.42,
|
| 1695 |
+
"learning_rate": 3.79070981210856e-05,
|
| 1696 |
+
"loss": 0.0486,
|
| 1697 |
+
"step": 11900
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 12.47,
|
| 1701 |
+
"learning_rate": 3.764613778705637e-05,
|
| 1702 |
+
"loss": 0.0481,
|
| 1703 |
+
"step": 11950
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 12.53,
|
| 1707 |
+
"learning_rate": 3.738517745302714e-05,
|
| 1708 |
+
"loss": 0.0456,
|
| 1709 |
+
"step": 12000
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 12.53,
|
| 1713 |
+
"eval_loss": 1.0609982013702393,
|
| 1714 |
+
"eval_runtime": 2297.0722,
|
| 1715 |
+
"eval_samples_per_second": 9.572,
|
| 1716 |
+
"eval_steps_per_second": 1.197,
|
| 1717 |
+
"eval_wer": 0.20719366027132632,
|
| 1718 |
+
"step": 12000
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 12.58,
|
| 1722 |
+
"learning_rate": 3.712421711899791e-05,
|
| 1723 |
+
"loss": 0.0476,
|
| 1724 |
+
"step": 12050
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 12.63,
|
| 1728 |
+
"learning_rate": 3.6863256784968685e-05,
|
| 1729 |
+
"loss": 0.0479,
|
| 1730 |
+
"step": 12100
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 12.68,
|
| 1734 |
+
"learning_rate": 3.660229645093946e-05,
|
| 1735 |
+
"loss": 0.0481,
|
| 1736 |
+
"step": 12150
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 12.73,
|
| 1740 |
+
"learning_rate": 3.634133611691023e-05,
|
| 1741 |
+
"loss": 0.0486,
|
| 1742 |
+
"step": 12200
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 12.79,
|
| 1746 |
+
"learning_rate": 3.6080375782881e-05,
|
| 1747 |
+
"loss": 0.0457,
|
| 1748 |
+
"step": 12250
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 12.84,
|
| 1752 |
+
"learning_rate": 3.5824634655532366e-05,
|
| 1753 |
+
"loss": 0.1915,
|
| 1754 |
+
"step": 12300
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 12.89,
|
| 1758 |
+
"learning_rate": 3.5563674321503134e-05,
|
| 1759 |
+
"loss": 0.0439,
|
| 1760 |
+
"step": 12350
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 12.94,
|
| 1764 |
+
"learning_rate": 3.530271398747391e-05,
|
| 1765 |
+
"loss": 0.0445,
|
| 1766 |
+
"step": 12400
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 12.94,
|
| 1770 |
+
"eval_loss": 1.0950515270233154,
|
| 1771 |
+
"eval_runtime": 2287.2126,
|
| 1772 |
+
"eval_samples_per_second": 9.613,
|
| 1773 |
+
"eval_steps_per_second": 1.202,
|
| 1774 |
+
"eval_wer": 0.20619484567765728,
|
| 1775 |
+
"step": 12400
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 13.0,
|
| 1779 |
+
"learning_rate": 3.504175365344468e-05,
|
| 1780 |
+
"loss": 0.0426,
|
| 1781 |
+
"step": 12450
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 13.05,
|
| 1785 |
+
"learning_rate": 3.478079331941545e-05,
|
| 1786 |
+
"loss": 0.0435,
|
| 1787 |
+
"step": 12500
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 13.1,
|
| 1791 |
+
"learning_rate": 3.451983298538623e-05,
|
| 1792 |
+
"loss": 0.0453,
|
| 1793 |
+
"step": 12550
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 13.15,
|
| 1797 |
+
"learning_rate": 3.4258872651356995e-05,
|
| 1798 |
+
"loss": 0.0419,
|
| 1799 |
+
"step": 12600
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 13.2,
|
| 1803 |
+
"learning_rate": 3.399791231732777e-05,
|
| 1804 |
+
"loss": 0.0476,
|
| 1805 |
+
"step": 12650
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 13.26,
|
| 1809 |
+
"learning_rate": 3.3736951983298544e-05,
|
| 1810 |
+
"loss": 0.0457,
|
| 1811 |
+
"step": 12700
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 13.31,
|
| 1815 |
+
"learning_rate": 3.347599164926931e-05,
|
| 1816 |
+
"loss": 0.0485,
|
| 1817 |
+
"step": 12750
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 13.36,
|
| 1821 |
+
"learning_rate": 3.321503131524009e-05,
|
| 1822 |
+
"loss": 0.0525,
|
| 1823 |
+
"step": 12800
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 13.36,
|
| 1827 |
+
"eval_loss": 1.1166725158691406,
|
| 1828 |
+
"eval_runtime": 2280.6976,
|
| 1829 |
+
"eval_samples_per_second": 9.641,
|
| 1830 |
+
"eval_steps_per_second": 1.205,
|
| 1831 |
+
"eval_wer": 0.20630460552311541,
|
| 1832 |
+
"step": 12800
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 13.41,
|
| 1836 |
+
"learning_rate": 3.2954070981210855e-05,
|
| 1837 |
+
"loss": 0.0454,
|
| 1838 |
+
"step": 12850
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 13.47,
|
| 1842 |
+
"learning_rate": 3.269311064718163e-05,
|
| 1843 |
+
"loss": 0.0441,
|
| 1844 |
+
"step": 12900
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 13.52,
|
| 1848 |
+
"learning_rate": 3.2432150313152405e-05,
|
| 1849 |
+
"loss": 0.0458,
|
| 1850 |
+
"step": 12950
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 13.57,
|
| 1854 |
+
"learning_rate": 3.217118997912317e-05,
|
| 1855 |
+
"loss": 0.0435,
|
| 1856 |
+
"step": 13000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 13.62,
|
| 1860 |
+
"learning_rate": 3.191022964509395e-05,
|
| 1861 |
+
"loss": 0.0404,
|
| 1862 |
+
"step": 13050
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 13.67,
|
| 1866 |
+
"learning_rate": 3.164926931106472e-05,
|
| 1867 |
+
"loss": 0.0495,
|
| 1868 |
+
"step": 13100
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 13.73,
|
| 1872 |
+
"learning_rate": 3.138830897703549e-05,
|
| 1873 |
+
"loss": 0.0529,
|
| 1874 |
+
"step": 13150
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 13.78,
|
| 1878 |
+
"learning_rate": 3.1127348643006265e-05,
|
| 1879 |
+
"loss": 0.0445,
|
| 1880 |
+
"step": 13200
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 13.78,
|
| 1884 |
+
"eval_loss": 1.1110990047454834,
|
| 1885 |
+
"eval_runtime": 2281.2238,
|
| 1886 |
+
"eval_samples_per_second": 9.639,
|
| 1887 |
+
"eval_steps_per_second": 1.205,
|
| 1888 |
+
"eval_wer": 0.20564385125345744,
|
| 1889 |
+
"step": 13200
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 13.83,
|
| 1893 |
+
"learning_rate": 3.086638830897703e-05,
|
| 1894 |
+
"loss": 0.0412,
|
| 1895 |
+
"step": 13250
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 13.88,
|
| 1899 |
+
"learning_rate": 3.060542797494781e-05,
|
| 1900 |
+
"loss": 0.045,
|
| 1901 |
+
"step": 13300
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 13.93,
|
| 1905 |
+
"learning_rate": 3.0344467640918583e-05,
|
| 1906 |
+
"loss": 0.0432,
|
| 1907 |
+
"step": 13350
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 13.99,
|
| 1911 |
+
"learning_rate": 3.008350730688935e-05,
|
| 1912 |
+
"loss": 0.0439,
|
| 1913 |
+
"step": 13400
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 14.04,
|
| 1917 |
+
"learning_rate": 2.9822546972860126e-05,
|
| 1918 |
+
"loss": 0.0458,
|
| 1919 |
+
"step": 13450
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 14.09,
|
| 1923 |
+
"learning_rate": 2.95615866388309e-05,
|
| 1924 |
+
"loss": 0.0492,
|
| 1925 |
+
"step": 13500
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 14.14,
|
| 1929 |
+
"learning_rate": 2.9300626304801672e-05,
|
| 1930 |
+
"loss": 0.0425,
|
| 1931 |
+
"step": 13550
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 14.2,
|
| 1935 |
+
"learning_rate": 2.9039665970772447e-05,
|
| 1936 |
+
"loss": 0.0435,
|
| 1937 |
+
"step": 13600
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 14.2,
|
| 1941 |
+
"eval_loss": 1.131235957145691,
|
| 1942 |
+
"eval_runtime": 2281.1607,
|
| 1943 |
+
"eval_samples_per_second": 9.639,
|
| 1944 |
+
"eval_steps_per_second": 1.205,
|
| 1945 |
+
"eval_wer": 0.2058567853536462,
|
| 1946 |
+
"step": 13600
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 14.25,
|
| 1950 |
+
"learning_rate": 2.8778705636743215e-05,
|
| 1951 |
+
"loss": 0.0429,
|
| 1952 |
+
"step": 13650
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 14.3,
|
| 1956 |
+
"learning_rate": 2.851774530271399e-05,
|
| 1957 |
+
"loss": 0.0428,
|
| 1958 |
+
"step": 13700
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 14.35,
|
| 1962 |
+
"learning_rate": 2.8256784968684764e-05,
|
| 1963 |
+
"loss": 0.0458,
|
| 1964 |
+
"step": 13750
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 14.4,
|
| 1968 |
+
"learning_rate": 2.7995824634655532e-05,
|
| 1969 |
+
"loss": 0.0411,
|
| 1970 |
+
"step": 13800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 14.46,
|
| 1974 |
+
"learning_rate": 2.7734864300626307e-05,
|
| 1975 |
+
"loss": 0.0444,
|
| 1976 |
+
"step": 13850
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 14.51,
|
| 1980 |
+
"learning_rate": 2.7473903966597082e-05,
|
| 1981 |
+
"loss": 0.0457,
|
| 1982 |
+
"step": 13900
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 14.56,
|
| 1986 |
+
"learning_rate": 2.721294363256785e-05,
|
| 1987 |
+
"loss": 0.042,
|
| 1988 |
+
"step": 13950
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 14.61,
|
| 1992 |
+
"learning_rate": 2.6951983298538625e-05,
|
| 1993 |
+
"loss": 0.0448,
|
| 1994 |
+
"step": 14000
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 14.61,
|
| 1998 |
+
"eval_loss": 1.1048628091812134,
|
| 1999 |
+
"eval_runtime": 2270.1056,
|
| 2000 |
+
"eval_samples_per_second": 9.686,
|
| 2001 |
+
"eval_steps_per_second": 1.211,
|
| 2002 |
+
"eval_wer": 0.2052684725819906,
|
| 2003 |
+
"step": 14000
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 14.67,
|
| 2007 |
+
"learning_rate": 2.6691022964509393e-05,
|
| 2008 |
+
"loss": 0.0431,
|
| 2009 |
+
"step": 14050
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 14.72,
|
| 2013 |
+
"learning_rate": 2.6430062630480167e-05,
|
| 2014 |
+
"loss": 0.0416,
|
| 2015 |
+
"step": 14100
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 14.77,
|
| 2019 |
+
"learning_rate": 2.6169102296450942e-05,
|
| 2020 |
+
"loss": 0.0451,
|
| 2021 |
+
"step": 14150
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 14.82,
|
| 2025 |
+
"learning_rate": 2.590814196242171e-05,
|
| 2026 |
+
"loss": 0.0432,
|
| 2027 |
+
"step": 14200
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 14.87,
|
| 2031 |
+
"learning_rate": 2.5647181628392485e-05,
|
| 2032 |
+
"loss": 0.0457,
|
| 2033 |
+
"step": 14250
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 14.93,
|
| 2037 |
+
"learning_rate": 2.538622129436326e-05,
|
| 2038 |
+
"loss": 0.0428,
|
| 2039 |
+
"step": 14300
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 14.98,
|
| 2043 |
+
"learning_rate": 2.5125260960334028e-05,
|
| 2044 |
+
"loss": 0.0437,
|
| 2045 |
+
"step": 14350
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 15.03,
|
| 2049 |
+
"learning_rate": 2.4864300626304803e-05,
|
| 2050 |
+
"loss": 0.0439,
|
| 2051 |
+
"step": 14400
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 15.03,
|
| 2055 |
+
"eval_loss": 1.11318838596344,
|
| 2056 |
+
"eval_runtime": 2284.6428,
|
| 2057 |
+
"eval_samples_per_second": 9.624,
|
| 2058 |
+
"eval_steps_per_second": 1.203,
|
| 2059 |
+
"eval_wer": 0.2054462835316328,
|
| 2060 |
+
"step": 14400
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 15.08,
|
| 2064 |
+
"learning_rate": 2.4603340292275574e-05,
|
| 2065 |
+
"loss": 0.0444,
|
| 2066 |
+
"step": 14450
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 15.14,
|
| 2070 |
+
"learning_rate": 2.434237995824635e-05,
|
| 2071 |
+
"loss": 0.0484,
|
| 2072 |
+
"step": 14500
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 15.19,
|
| 2076 |
+
"learning_rate": 2.408141962421712e-05,
|
| 2077 |
+
"loss": 0.0457,
|
| 2078 |
+
"step": 14550
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 15.24,
|
| 2082 |
+
"learning_rate": 2.3820459290187895e-05,
|
| 2083 |
+
"loss": 0.0398,
|
| 2084 |
+
"step": 14600
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 15.29,
|
| 2088 |
+
"learning_rate": 2.3559498956158667e-05,
|
| 2089 |
+
"loss": 0.0409,
|
| 2090 |
+
"step": 14650
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 15.34,
|
| 2094 |
+
"learning_rate": 2.3298538622129438e-05,
|
| 2095 |
+
"loss": 0.0464,
|
| 2096 |
+
"step": 14700
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 15.4,
|
| 2100 |
+
"learning_rate": 2.303757828810021e-05,
|
| 2101 |
+
"loss": 0.0395,
|
| 2102 |
+
"step": 14750
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 15.45,
|
| 2106 |
+
"learning_rate": 2.2776617954070984e-05,
|
| 2107 |
+
"loss": 0.0436,
|
| 2108 |
+
"step": 14800
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 15.45,
|
| 2112 |
+
"eval_loss": 1.1486200094223022,
|
| 2113 |
+
"eval_runtime": 2280.6212,
|
| 2114 |
+
"eval_samples_per_second": 9.641,
|
| 2115 |
+
"eval_steps_per_second": 1.205,
|
| 2116 |
+
"eval_wer": 0.20591386047328444,
|
| 2117 |
+
"step": 14800
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 15.5,
|
| 2121 |
+
"learning_rate": 2.2515657620041756e-05,
|
| 2122 |
+
"loss": 0.0413,
|
| 2123 |
+
"step": 14850
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 15.55,
|
| 2127 |
+
"learning_rate": 2.2254697286012527e-05,
|
| 2128 |
+
"loss": 0.0392,
|
| 2129 |
+
"step": 14900
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 15.61,
|
| 2133 |
+
"learning_rate": 2.19937369519833e-05,
|
| 2134 |
+
"loss": 0.0406,
|
| 2135 |
+
"step": 14950
|
| 2136 |
+
},
|
| 2137 |
+
{
|
| 2138 |
+
"epoch": 15.66,
|
| 2139 |
+
"learning_rate": 2.1732776617954073e-05,
|
| 2140 |
+
"loss": 0.0419,
|
| 2141 |
+
"step": 15000
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 15.71,
|
| 2145 |
+
"learning_rate": 2.1471816283924845e-05,
|
| 2146 |
+
"loss": 0.0451,
|
| 2147 |
+
"step": 15050
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 15.76,
|
| 2151 |
+
"learning_rate": 2.1210855949895616e-05,
|
| 2152 |
+
"loss": 0.0454,
|
| 2153 |
+
"step": 15100
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 15.81,
|
| 2157 |
+
"learning_rate": 2.0949895615866387e-05,
|
| 2158 |
+
"loss": 0.042,
|
| 2159 |
+
"step": 15150
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 15.87,
|
| 2163 |
+
"learning_rate": 2.0688935281837162e-05,
|
| 2164 |
+
"loss": 0.043,
|
| 2165 |
+
"step": 15200
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 15.87,
|
| 2169 |
+
"eval_loss": 1.1113590002059937,
|
| 2170 |
+
"eval_runtime": 2271.0619,
|
| 2171 |
+
"eval_samples_per_second": 9.682,
|
| 2172 |
+
"eval_steps_per_second": 1.21,
|
| 2173 |
+
"eval_wer": 0.20565482723800324,
|
| 2174 |
+
"step": 15200
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 15.92,
|
| 2178 |
+
"learning_rate": 2.0427974947807934e-05,
|
| 2179 |
+
"loss": 0.0384,
|
| 2180 |
+
"step": 15250
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 15.97,
|
| 2184 |
+
"learning_rate": 2.0167014613778705e-05,
|
| 2185 |
+
"loss": 0.0461,
|
| 2186 |
+
"step": 15300
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 16.02,
|
| 2190 |
+
"learning_rate": 1.9906054279749476e-05,
|
| 2191 |
+
"loss": 0.0414,
|
| 2192 |
+
"step": 15350
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 16.08,
|
| 2196 |
+
"learning_rate": 1.964509394572025e-05,
|
| 2197 |
+
"loss": 0.0398,
|
| 2198 |
+
"step": 15400
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 16.13,
|
| 2202 |
+
"learning_rate": 1.9384133611691023e-05,
|
| 2203 |
+
"loss": 0.0391,
|
| 2204 |
+
"step": 15450
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 16.18,
|
| 2208 |
+
"learning_rate": 1.9123173277661798e-05,
|
| 2209 |
+
"loss": 0.0398,
|
| 2210 |
+
"step": 15500
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 16.23,
|
| 2214 |
+
"learning_rate": 1.886221294363257e-05,
|
| 2215 |
+
"loss": 0.0454,
|
| 2216 |
+
"step": 15550
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 16.28,
|
| 2220 |
+
"learning_rate": 1.8601252609603344e-05,
|
| 2221 |
+
"loss": 0.046,
|
| 2222 |
+
"step": 15600
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 16.28,
|
| 2226 |
+
"eval_loss": 1.1185530424118042,
|
| 2227 |
+
"eval_runtime": 2277.3852,
|
| 2228 |
+
"eval_samples_per_second": 9.655,
|
| 2229 |
+
"eval_steps_per_second": 1.207,
|
| 2230 |
+
"eval_wer": 0.2055275058172718,
|
| 2231 |
+
"step": 15600
|
| 2232 |
+
},
|
| 2233 |
+
{
|
| 2234 |
+
"epoch": 16.34,
|
| 2235 |
+
"learning_rate": 1.8340292275574115e-05,
|
| 2236 |
+
"loss": 0.0391,
|
| 2237 |
+
"step": 15650
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 16.39,
|
| 2241 |
+
"learning_rate": 1.8079331941544887e-05,
|
| 2242 |
+
"loss": 0.0443,
|
| 2243 |
+
"step": 15700
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 16.44,
|
| 2247 |
+
"learning_rate": 1.7818371607515658e-05,
|
| 2248 |
+
"loss": 0.0424,
|
| 2249 |
+
"step": 15750
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 16.49,
|
| 2253 |
+
"learning_rate": 1.7557411273486433e-05,
|
| 2254 |
+
"loss": 0.0455,
|
| 2255 |
+
"step": 15800
|
| 2256 |
+
},
|
| 2257 |
+
{
|
| 2258 |
+
"epoch": 16.54,
|
| 2259 |
+
"learning_rate": 1.7296450939457204e-05,
|
| 2260 |
+
"loss": 0.0415,
|
| 2261 |
+
"step": 15850
|
| 2262 |
+
},
|
| 2263 |
+
{
|
| 2264 |
+
"epoch": 16.6,
|
| 2265 |
+
"learning_rate": 1.7035490605427976e-05,
|
| 2266 |
+
"loss": 0.0427,
|
| 2267 |
+
"step": 15900
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"epoch": 16.65,
|
| 2271 |
+
"learning_rate": 1.6774530271398747e-05,
|
| 2272 |
+
"loss": 0.0428,
|
| 2273 |
+
"step": 15950
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 16.7,
|
| 2277 |
+
"learning_rate": 1.6513569937369522e-05,
|
| 2278 |
+
"loss": 0.0432,
|
| 2279 |
+
"step": 16000
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 16.7,
|
| 2283 |
+
"eval_loss": 1.1420904397964478,
|
| 2284 |
+
"eval_runtime": 2286.2257,
|
| 2285 |
+
"eval_samples_per_second": 9.618,
|
| 2286 |
+
"eval_steps_per_second": 1.202,
|
| 2287 |
+
"eval_wer": 0.20530579092944637,
|
| 2288 |
+
"step": 16000
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 16.75,
|
| 2292 |
+
"learning_rate": 1.6252609603340293e-05,
|
| 2293 |
+
"loss": 0.0434,
|
| 2294 |
+
"step": 16050
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 16.81,
|
| 2298 |
+
"learning_rate": 1.5991649269311065e-05,
|
| 2299 |
+
"loss": 0.0422,
|
| 2300 |
+
"step": 16100
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 16.86,
|
| 2304 |
+
"learning_rate": 1.5730688935281836e-05,
|
| 2305 |
+
"loss": 0.0435,
|
| 2306 |
+
"step": 16150
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 16.91,
|
| 2310 |
+
"learning_rate": 1.546972860125261e-05,
|
| 2311 |
+
"loss": 0.0423,
|
| 2312 |
+
"step": 16200
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 16.96,
|
| 2316 |
+
"learning_rate": 1.5208768267223384e-05,
|
| 2317 |
+
"loss": 0.0453,
|
| 2318 |
+
"step": 16250
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 17.01,
|
| 2322 |
+
"learning_rate": 1.4947807933194155e-05,
|
| 2323 |
+
"loss": 0.0401,
|
| 2324 |
+
"step": 16300
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"epoch": 17.07,
|
| 2328 |
+
"learning_rate": 1.4686847599164927e-05,
|
| 2329 |
+
"loss": 0.0435,
|
| 2330 |
+
"step": 16350
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 17.12,
|
| 2334 |
+
"learning_rate": 1.4425887265135702e-05,
|
| 2335 |
+
"loss": 0.0403,
|
| 2336 |
+
"step": 16400
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 17.12,
|
| 2340 |
+
"eval_loss": 1.1468006372451782,
|
| 2341 |
+
"eval_runtime": 2284.2746,
|
| 2342 |
+
"eval_samples_per_second": 9.626,
|
| 2343 |
+
"eval_steps_per_second": 1.203,
|
| 2344 |
+
"eval_wer": 0.20491065548579707,
|
| 2345 |
+
"step": 16400
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 17.17,
|
| 2349 |
+
"learning_rate": 1.4164926931106473e-05,
|
| 2350 |
+
"loss": 0.0397,
|
| 2351 |
+
"step": 16450
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"epoch": 17.22,
|
| 2355 |
+
"learning_rate": 1.3903966597077244e-05,
|
| 2356 |
+
"loss": 0.0375,
|
| 2357 |
+
"step": 16500
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 17.28,
|
| 2361 |
+
"learning_rate": 1.3643006263048016e-05,
|
| 2362 |
+
"loss": 0.0424,
|
| 2363 |
+
"step": 16550
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 17.33,
|
| 2367 |
+
"learning_rate": 1.338204592901879e-05,
|
| 2368 |
+
"loss": 0.0438,
|
| 2369 |
+
"step": 16600
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 17.38,
|
| 2373 |
+
"learning_rate": 1.3121085594989562e-05,
|
| 2374 |
+
"loss": 0.0385,
|
| 2375 |
+
"step": 16650
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 17.43,
|
| 2379 |
+
"learning_rate": 1.2860125260960335e-05,
|
| 2380 |
+
"loss": 0.0402,
|
| 2381 |
+
"step": 16700
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 17.48,
|
| 2385 |
+
"learning_rate": 1.2599164926931107e-05,
|
| 2386 |
+
"loss": 0.0457,
|
| 2387 |
+
"step": 16750
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 17.54,
|
| 2391 |
+
"learning_rate": 1.233820459290188e-05,
|
| 2392 |
+
"loss": 0.0426,
|
| 2393 |
+
"step": 16800
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 17.54,
|
| 2397 |
+
"eval_loss": 1.149783730506897,
|
| 2398 |
+
"eval_runtime": 2292.0315,
|
| 2399 |
+
"eval_samples_per_second": 9.593,
|
| 2400 |
+
"eval_steps_per_second": 1.199,
|
| 2401 |
+
"eval_wer": 0.2040084295561312,
|
| 2402 |
+
"step": 16800
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 17.59,
|
| 2406 |
+
"learning_rate": 1.2077244258872653e-05,
|
| 2407 |
+
"loss": 0.0387,
|
| 2408 |
+
"step": 16850
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 17.64,
|
| 2412 |
+
"learning_rate": 1.1816283924843424e-05,
|
| 2413 |
+
"loss": 0.0399,
|
| 2414 |
+
"step": 16900
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 17.69,
|
| 2418 |
+
"learning_rate": 1.1555323590814197e-05,
|
| 2419 |
+
"loss": 0.0412,
|
| 2420 |
+
"step": 16950
|
| 2421 |
+
},
|
| 2422 |
+
{
|
| 2423 |
+
"epoch": 17.74,
|
| 2424 |
+
"learning_rate": 1.1294363256784969e-05,
|
| 2425 |
+
"loss": 0.041,
|
| 2426 |
+
"step": 17000
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 17.8,
|
| 2430 |
+
"learning_rate": 1.1033402922755742e-05,
|
| 2431 |
+
"loss": 0.0401,
|
| 2432 |
+
"step": 17050
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 17.85,
|
| 2436 |
+
"learning_rate": 1.0772442588726513e-05,
|
| 2437 |
+
"loss": 0.0422,
|
| 2438 |
+
"step": 17100
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 17.9,
|
| 2442 |
+
"learning_rate": 1.0511482254697288e-05,
|
| 2443 |
+
"loss": 0.0384,
|
| 2444 |
+
"step": 17150
|
| 2445 |
+
},
|
| 2446 |
+
{
|
| 2447 |
+
"epoch": 17.95,
|
| 2448 |
+
"learning_rate": 1.025052192066806e-05,
|
| 2449 |
+
"loss": 0.0367,
|
| 2450 |
+
"step": 17200
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"epoch": 17.95,
|
| 2454 |
+
"eval_loss": 1.1410751342773438,
|
| 2455 |
+
"eval_runtime": 2289.1143,
|
| 2456 |
+
"eval_samples_per_second": 9.605,
|
| 2457 |
+
"eval_steps_per_second": 1.201,
|
| 2458 |
+
"eval_wer": 0.20410501822013435,
|
| 2459 |
+
"step": 17200
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 18.01,
|
| 2463 |
+
"learning_rate": 9.989561586638832e-06,
|
| 2464 |
+
"loss": 0.0424,
|
| 2465 |
+
"step": 17250
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 18.06,
|
| 2469 |
+
"learning_rate": 9.728601252609604e-06,
|
| 2470 |
+
"loss": 0.0362,
|
| 2471 |
+
"step": 17300
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 18.11,
|
| 2475 |
+
"learning_rate": 9.467640918580377e-06,
|
| 2476 |
+
"loss": 0.0418,
|
| 2477 |
+
"step": 17350
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 18.16,
|
| 2481 |
+
"learning_rate": 9.206680584551148e-06,
|
| 2482 |
+
"loss": 0.0423,
|
| 2483 |
+
"step": 17400
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 18.21,
|
| 2487 |
+
"learning_rate": 8.945720250521922e-06,
|
| 2488 |
+
"loss": 0.0403,
|
| 2489 |
+
"step": 17450
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 18.27,
|
| 2493 |
+
"learning_rate": 8.684759916492693e-06,
|
| 2494 |
+
"loss": 0.0429,
|
| 2495 |
+
"step": 17500
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 18.32,
|
| 2499 |
+
"learning_rate": 8.423799582463466e-06,
|
| 2500 |
+
"loss": 0.0377,
|
| 2501 |
+
"step": 17550
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 18.37,
|
| 2505 |
+
"learning_rate": 8.162839248434239e-06,
|
| 2506 |
+
"loss": 0.0398,
|
| 2507 |
+
"step": 17600
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 18.37,
|
| 2511 |
+
"eval_loss": 1.1493802070617676,
|
| 2512 |
+
"eval_runtime": 2285.3013,
|
| 2513 |
+
"eval_samples_per_second": 9.621,
|
| 2514 |
+
"eval_steps_per_second": 1.203,
|
| 2515 |
+
"eval_wer": 0.20455942398033103,
|
| 2516 |
+
"step": 17600
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 18.42,
|
| 2520 |
+
"learning_rate": 7.901878914405012e-06,
|
| 2521 |
+
"loss": 0.041,
|
| 2522 |
+
"step": 17650
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 18.48,
|
| 2526 |
+
"learning_rate": 7.640918580375784e-06,
|
| 2527 |
+
"loss": 0.0453,
|
| 2528 |
+
"step": 17700
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 18.53,
|
| 2532 |
+
"learning_rate": 7.379958246346557e-06,
|
| 2533 |
+
"loss": 0.0404,
|
| 2534 |
+
"step": 17750
|
| 2535 |
+
},
|
| 2536 |
+
{
|
| 2537 |
+
"epoch": 18.58,
|
| 2538 |
+
"learning_rate": 7.118997912317328e-06,
|
| 2539 |
+
"loss": 0.0456,
|
| 2540 |
+
"step": 17800
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 18.63,
|
| 2544 |
+
"learning_rate": 6.858037578288101e-06,
|
| 2545 |
+
"loss": 0.0411,
|
| 2546 |
+
"step": 17850
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 18.68,
|
| 2550 |
+
"learning_rate": 6.597077244258873e-06,
|
| 2551 |
+
"loss": 0.0441,
|
| 2552 |
+
"step": 17900
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 18.74,
|
| 2556 |
+
"learning_rate": 6.336116910229645e-06,
|
| 2557 |
+
"loss": 0.0429,
|
| 2558 |
+
"step": 17950
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 18.79,
|
| 2562 |
+
"learning_rate": 6.075156576200418e-06,
|
| 2563 |
+
"loss": 0.037,
|
| 2564 |
+
"step": 18000
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 18.79,
|
| 2568 |
+
"eval_loss": 1.1569510698318481,
|
| 2569 |
+
"eval_runtime": 2284.6736,
|
| 2570 |
+
"eval_samples_per_second": 9.624,
|
| 2571 |
+
"eval_steps_per_second": 1.203,
|
| 2572 |
+
"eval_wer": 0.20445844492250956,
|
| 2573 |
+
"step": 18000
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 18.84,
|
| 2577 |
+
"learning_rate": 5.81419624217119e-06,
|
| 2578 |
+
"loss": 0.0359,
|
| 2579 |
+
"step": 18050
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 18.89,
|
| 2583 |
+
"learning_rate": 5.553235908141963e-06,
|
| 2584 |
+
"loss": 0.0395,
|
| 2585 |
+
"step": 18100
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 18.95,
|
| 2589 |
+
"learning_rate": 5.29749478079332e-06,
|
| 2590 |
+
"loss": 0.0445,
|
| 2591 |
+
"step": 18150
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 19.0,
|
| 2595 |
+
"learning_rate": 5.036534446764092e-06,
|
| 2596 |
+
"loss": 0.0422,
|
| 2597 |
+
"step": 18200
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 19.05,
|
| 2601 |
+
"learning_rate": 4.775574112734864e-06,
|
| 2602 |
+
"loss": 0.0343,
|
| 2603 |
+
"step": 18250
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 19.1,
|
| 2607 |
+
"learning_rate": 4.514613778705637e-06,
|
| 2608 |
+
"loss": 0.0409,
|
| 2609 |
+
"step": 18300
|
| 2610 |
+
},
|
| 2611 |
+
{
|
| 2612 |
+
"epoch": 19.15,
|
| 2613 |
+
"learning_rate": 4.25365344467641e-06,
|
| 2614 |
+
"loss": 0.041,
|
| 2615 |
+
"step": 18350
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 19.21,
|
| 2619 |
+
"learning_rate": 3.992693110647182e-06,
|
| 2620 |
+
"loss": 0.0383,
|
| 2621 |
+
"step": 18400
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 19.21,
|
| 2625 |
+
"eval_loss": 1.1532527208328247,
|
| 2626 |
+
"eval_runtime": 2289.8513,
|
| 2627 |
+
"eval_samples_per_second": 9.602,
|
| 2628 |
+
"eval_steps_per_second": 1.201,
|
| 2629 |
+
"eval_wer": 0.2042784387759582,
|
| 2630 |
+
"step": 18400
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 19.26,
|
| 2634 |
+
"learning_rate": 3.7317327766179542e-06,
|
| 2635 |
+
"loss": 0.0427,
|
| 2636 |
+
"step": 18450
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 19.31,
|
| 2640 |
+
"learning_rate": 3.4707724425887265e-06,
|
| 2641 |
+
"loss": 0.0391,
|
| 2642 |
+
"step": 18500
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 19.36,
|
| 2646 |
+
"learning_rate": 3.209812108559499e-06,
|
| 2647 |
+
"loss": 0.0351,
|
| 2648 |
+
"step": 18550
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 19.42,
|
| 2652 |
+
"learning_rate": 2.9488517745302714e-06,
|
| 2653 |
+
"loss": 0.037,
|
| 2654 |
+
"step": 18600
|
| 2655 |
+
},
|
| 2656 |
+
{
|
| 2657 |
+
"epoch": 19.47,
|
| 2658 |
+
"learning_rate": 2.687891440501044e-06,
|
| 2659 |
+
"loss": 0.0423,
|
| 2660 |
+
"step": 18650
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 19.52,
|
| 2664 |
+
"learning_rate": 2.4269311064718164e-06,
|
| 2665 |
+
"loss": 0.0434,
|
| 2666 |
+
"step": 18700
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 19.57,
|
| 2670 |
+
"learning_rate": 2.165970772442589e-06,
|
| 2671 |
+
"loss": 0.0435,
|
| 2672 |
+
"step": 18750
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 19.62,
|
| 2676 |
+
"learning_rate": 1.9050104384133613e-06,
|
| 2677 |
+
"loss": 0.0374,
|
| 2678 |
+
"step": 18800
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 19.62,
|
| 2682 |
+
"eval_loss": 1.1447900533676147,
|
| 2683 |
+
"eval_runtime": 2296.5158,
|
| 2684 |
+
"eval_samples_per_second": 9.575,
|
| 2685 |
+
"eval_steps_per_second": 1.197,
|
| 2686 |
+
"eval_wer": 0.20419941168722835,
|
| 2687 |
+
"step": 18800
|
| 2688 |
+
}
|
| 2689 |
+
],
|
| 2690 |
+
"max_steps": 19160,
|
| 2691 |
+
"num_train_epochs": 20,
|
| 2692 |
+
"total_flos": 1.280273400848764e+20,
|
| 2693 |
+
"trial_name": null,
|
| 2694 |
+
"trial_params": null
|
| 2695 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d322a9f9aa3f29096569575470be8c1b4321194ac12b5ea5d4f26ce0c241192
|
| 3 |
+
size 2863
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"e": 0, "n": 1, "a": 2, "i": 3, "t": 4, "o": 5, "d": 6, "r": 7, " ": 8, "l": 9, "s": 10, "h": 11, "g": 12, "m": 13, "k": 14, "v": 15, "j": 16, "w": 17, "z": 18, "u": 19, "b": 20, "c": 21, "p": 22, "f": 23, "y": 24, "\u00e9": 25, "'": 26, "x": 27, "\u00eb": 28, "q": 29, "-": 30, "\u00ea": 31, "\u00e0": 32, "\u00e4": 33, "\u00e8": 34, "\u00ef": 35, "\u00e2": 36, "\u00fb": 37, "\u00f6": 38, "\u00f4": 39, "\u00fc": 40, "\u00ee": 41, "\u00e7": 42, "\u00e6": 43, "\u00f9": 44, "\u0153": 45, "[UNK]": 46, "[PAD]": 47}
|