Robert Henderson
commited on
Commit
•
670fa97
1
Parent(s):
e8824ec
checkpoint 7600 for uspanteko asr
Browse files- checkpoint-7600/config.json +115 -0
- checkpoint-7600/optimizer.pt +3 -0
- checkpoint-7600/preprocessor_config.json +10 -0
- checkpoint-7600/pytorch_model.bin +3 -0
- checkpoint-7600/rng_state.pth +3 -0
- checkpoint-7600/scaler.pt +3 -0
- checkpoint-7600/scheduler.pt +3 -0
- checkpoint-7600/trainer_state.json +1612 -0
- checkpoint-7600/training_args.bin +3 -0
checkpoint-7600/config.json
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 768,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": true,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.0,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"hidden_act": "gelu",
|
| 56 |
+
"hidden_dropout": 0.1,
|
| 57 |
+
"hidden_size": 1024,
|
| 58 |
+
"initializer_range": 0.02,
|
| 59 |
+
"intermediate_size": 4096,
|
| 60 |
+
"layer_norm_eps": 1e-05,
|
| 61 |
+
"layerdrop": 0.1,
|
| 62 |
+
"mask_channel_length": 10,
|
| 63 |
+
"mask_channel_min_space": 1,
|
| 64 |
+
"mask_channel_other": 0.0,
|
| 65 |
+
"mask_channel_prob": 0.0,
|
| 66 |
+
"mask_channel_selection": "static",
|
| 67 |
+
"mask_feature_length": 10,
|
| 68 |
+
"mask_feature_min_masks": 0,
|
| 69 |
+
"mask_feature_prob": 0.0,
|
| 70 |
+
"mask_time_length": 10,
|
| 71 |
+
"mask_time_min_masks": 2,
|
| 72 |
+
"mask_time_min_space": 1,
|
| 73 |
+
"mask_time_other": 0.0,
|
| 74 |
+
"mask_time_prob": 0.05,
|
| 75 |
+
"mask_time_selection": "static",
|
| 76 |
+
"model_type": "wav2vec2",
|
| 77 |
+
"num_adapter_layers": 3,
|
| 78 |
+
"num_attention_heads": 16,
|
| 79 |
+
"num_codevector_groups": 2,
|
| 80 |
+
"num_codevectors_per_group": 320,
|
| 81 |
+
"num_conv_pos_embedding_groups": 16,
|
| 82 |
+
"num_conv_pos_embeddings": 128,
|
| 83 |
+
"num_feat_extract_layers": 7,
|
| 84 |
+
"num_hidden_layers": 24,
|
| 85 |
+
"num_negatives": 100,
|
| 86 |
+
"output_hidden_size": 1024,
|
| 87 |
+
"pad_token_id": 45,
|
| 88 |
+
"proj_codevector_dim": 768,
|
| 89 |
+
"tdnn_dilation": [
|
| 90 |
+
1,
|
| 91 |
+
2,
|
| 92 |
+
3,
|
| 93 |
+
1,
|
| 94 |
+
1
|
| 95 |
+
],
|
| 96 |
+
"tdnn_dim": [
|
| 97 |
+
512,
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
1500
|
| 102 |
+
],
|
| 103 |
+
"tdnn_kernel": [
|
| 104 |
+
5,
|
| 105 |
+
3,
|
| 106 |
+
3,
|
| 107 |
+
1,
|
| 108 |
+
1
|
| 109 |
+
],
|
| 110 |
+
"torch_dtype": "float32",
|
| 111 |
+
"transformers_version": "4.28.0",
|
| 112 |
+
"use_weighted_layer_sum": false,
|
| 113 |
+
"vocab_size": 46,
|
| 114 |
+
"xvector_output_dim": 512
|
| 115 |
+
}
|
checkpoint-7600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e2d6561bebacc7cb7093fc89d8a34f6f6deaf724c6ddae98c6de8951900690d9
|
| 3 |
+
size 2490438522
|
checkpoint-7600/preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
checkpoint-7600/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b10e9c751b72e6ee89af20ea4db2c6c8f79ada3515adf76c21f8142e79635d7
|
| 3 |
+
size 1262090858
|
checkpoint-7600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb97940f746ca804e9b1d7f13308fdd351ce11f795ec40decec943b33f4b812c
|
| 3 |
+
size 14308
|
checkpoint-7600/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e6d93faf7357af9dca7dd3d931417a9f92c153d3d2ced8fddf444b3ac48d43c
|
| 3 |
+
size 988
|
checkpoint-7600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfd7753fe9ffa8644105e0a5dd1fbac1365ed034e3d08304c24355946574ce51
|
| 3 |
+
size 1064
|
checkpoint-7600/trainer_state.json
ADDED
|
@@ -0,0 +1,1612 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 71.36150234741784,
|
| 5 |
+
"global_step": 7600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.47,
|
| 12 |
+
"learning_rate": 2.8199999999999998e-05,
|
| 13 |
+
"loss": 9.9485,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.94,
|
| 18 |
+
"learning_rate": 5.82e-05,
|
| 19 |
+
"loss": 5.4005,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.94,
|
| 24 |
+
"eval_loss": 3.5134339332580566,
|
| 25 |
+
"eval_runtime": 29.4124,
|
| 26 |
+
"eval_samples_per_second": 7.242,
|
| 27 |
+
"eval_steps_per_second": 0.918,
|
| 28 |
+
"eval_wer": 0.9822416974169742,
|
| 29 |
+
"step": 100
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.41,
|
| 33 |
+
"learning_rate": 8.819999999999999e-05,
|
| 34 |
+
"loss": 3.2998,
|
| 35 |
+
"step": 150
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.88,
|
| 39 |
+
"learning_rate": 0.0001182,
|
| 40 |
+
"loss": 3.0848,
|
| 41 |
+
"step": 200
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.88,
|
| 45 |
+
"eval_loss": 3.0378262996673584,
|
| 46 |
+
"eval_runtime": 29.389,
|
| 47 |
+
"eval_samples_per_second": 7.248,
|
| 48 |
+
"eval_steps_per_second": 0.919,
|
| 49 |
+
"eval_wer": 0.9822416974169742,
|
| 50 |
+
"step": 200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.35,
|
| 54 |
+
"learning_rate": 0.0001482,
|
| 55 |
+
"loss": 3.0465,
|
| 56 |
+
"step": 250
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.82,
|
| 60 |
+
"learning_rate": 0.00017819999999999997,
|
| 61 |
+
"loss": 3.0385,
|
| 62 |
+
"step": 300
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.82,
|
| 66 |
+
"eval_loss": 3.0117852687835693,
|
| 67 |
+
"eval_runtime": 29.166,
|
| 68 |
+
"eval_samples_per_second": 7.303,
|
| 69 |
+
"eval_steps_per_second": 0.926,
|
| 70 |
+
"eval_wer": 0.9822416974169742,
|
| 71 |
+
"step": 300
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.29,
|
| 75 |
+
"learning_rate": 0.00020819999999999996,
|
| 76 |
+
"loss": 3.0281,
|
| 77 |
+
"step": 350
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.76,
|
| 81 |
+
"learning_rate": 0.0002382,
|
| 82 |
+
"loss": 2.9897,
|
| 83 |
+
"step": 400
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 3.76,
|
| 87 |
+
"eval_loss": 2.9294049739837646,
|
| 88 |
+
"eval_runtime": 29.3884,
|
| 89 |
+
"eval_samples_per_second": 7.248,
|
| 90 |
+
"eval_steps_per_second": 0.919,
|
| 91 |
+
"eval_wer": 0.9822416974169742,
|
| 92 |
+
"step": 400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.23,
|
| 96 |
+
"learning_rate": 0.00026819999999999996,
|
| 97 |
+
"loss": 2.5043,
|
| 98 |
+
"step": 450
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.69,
|
| 102 |
+
"learning_rate": 0.0002982,
|
| 103 |
+
"loss": 1.4181,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.69,
|
| 108 |
+
"eval_loss": 0.9793046116828918,
|
| 109 |
+
"eval_runtime": 29.05,
|
| 110 |
+
"eval_samples_per_second": 7.332,
|
| 111 |
+
"eval_steps_per_second": 0.929,
|
| 112 |
+
"eval_wer": 0.9068265682656826,
|
| 113 |
+
"step": 500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.16,
|
| 117 |
+
"learning_rate": 0.00029852664576802507,
|
| 118 |
+
"loss": 1.044,
|
| 119 |
+
"step": 550
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 5.63,
|
| 123 |
+
"learning_rate": 0.0002969592476489028,
|
| 124 |
+
"loss": 0.8274,
|
| 125 |
+
"step": 600
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.63,
|
| 129 |
+
"eval_loss": 0.6433392763137817,
|
| 130 |
+
"eval_runtime": 29.4518,
|
| 131 |
+
"eval_samples_per_second": 7.232,
|
| 132 |
+
"eval_steps_per_second": 0.917,
|
| 133 |
+
"eval_wer": 0.6335332103321033,
|
| 134 |
+
"step": 600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.1,
|
| 138 |
+
"learning_rate": 0.00029539184952978055,
|
| 139 |
+
"loss": 0.7475,
|
| 140 |
+
"step": 650
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.57,
|
| 144 |
+
"learning_rate": 0.0002938244514106583,
|
| 145 |
+
"loss": 0.6117,
|
| 146 |
+
"step": 700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 6.57,
|
| 150 |
+
"eval_loss": 0.5542563796043396,
|
| 151 |
+
"eval_runtime": 29.4264,
|
| 152 |
+
"eval_samples_per_second": 7.238,
|
| 153 |
+
"eval_steps_per_second": 0.918,
|
| 154 |
+
"eval_wer": 0.566420664206642,
|
| 155 |
+
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 7.04,
|
| 159 |
+
"learning_rate": 0.00029225705329153604,
|
| 160 |
+
"loss": 0.5792,
|
| 161 |
+
"step": 750
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 7.51,
|
| 165 |
+
"learning_rate": 0.0002906896551724138,
|
| 166 |
+
"loss": 0.4713,
|
| 167 |
+
"step": 800
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 7.51,
|
| 171 |
+
"eval_loss": 0.5372164845466614,
|
| 172 |
+
"eval_runtime": 29.4964,
|
| 173 |
+
"eval_samples_per_second": 7.221,
|
| 174 |
+
"eval_steps_per_second": 0.915,
|
| 175 |
+
"eval_wer": 0.5166051660516605,
|
| 176 |
+
"step": 800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 7.98,
|
| 180 |
+
"learning_rate": 0.0002891222570532915,
|
| 181 |
+
"loss": 0.4702,
|
| 182 |
+
"step": 850
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 8.45,
|
| 186 |
+
"learning_rate": 0.00028755485893416926,
|
| 187 |
+
"loss": 0.4045,
|
| 188 |
+
"step": 900
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 8.45,
|
| 192 |
+
"eval_loss": 0.49404770135879517,
|
| 193 |
+
"eval_runtime": 29.3183,
|
| 194 |
+
"eval_samples_per_second": 7.265,
|
| 195 |
+
"eval_steps_per_second": 0.921,
|
| 196 |
+
"eval_wer": 0.4534132841328413,
|
| 197 |
+
"step": 900
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 8.92,
|
| 201 |
+
"learning_rate": 0.000285987460815047,
|
| 202 |
+
"loss": 0.3913,
|
| 203 |
+
"step": 950
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 9.39,
|
| 207 |
+
"learning_rate": 0.00028442006269592475,
|
| 208 |
+
"loss": 0.3514,
|
| 209 |
+
"step": 1000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.39,
|
| 213 |
+
"eval_loss": 0.4763822853565216,
|
| 214 |
+
"eval_runtime": 29.42,
|
| 215 |
+
"eval_samples_per_second": 7.24,
|
| 216 |
+
"eval_steps_per_second": 0.918,
|
| 217 |
+
"eval_wer": 0.4432656826568266,
|
| 218 |
+
"step": 1000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.86,
|
| 222 |
+
"learning_rate": 0.0002828526645768025,
|
| 223 |
+
"loss": 0.3558,
|
| 224 |
+
"step": 1050
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 10.33,
|
| 228 |
+
"learning_rate": 0.00028128526645768024,
|
| 229 |
+
"loss": 0.333,
|
| 230 |
+
"step": 1100
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 10.33,
|
| 234 |
+
"eval_loss": 0.4512215554714203,
|
| 235 |
+
"eval_runtime": 29.3921,
|
| 236 |
+
"eval_samples_per_second": 7.247,
|
| 237 |
+
"eval_steps_per_second": 0.919,
|
| 238 |
+
"eval_wer": 0.4232011070110701,
|
| 239 |
+
"step": 1100
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 10.8,
|
| 243 |
+
"learning_rate": 0.000279717868338558,
|
| 244 |
+
"loss": 0.3235,
|
| 245 |
+
"step": 1150
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 11.27,
|
| 249 |
+
"learning_rate": 0.0002781504702194357,
|
| 250 |
+
"loss": 0.2939,
|
| 251 |
+
"step": 1200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 11.27,
|
| 255 |
+
"eval_loss": 0.4824497699737549,
|
| 256 |
+
"eval_runtime": 29.3229,
|
| 257 |
+
"eval_samples_per_second": 7.264,
|
| 258 |
+
"eval_steps_per_second": 0.921,
|
| 259 |
+
"eval_wer": 0.42273985239852396,
|
| 260 |
+
"step": 1200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 11.74,
|
| 264 |
+
"learning_rate": 0.00027658307210031346,
|
| 265 |
+
"loss": 0.2949,
|
| 266 |
+
"step": 1250
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 12.21,
|
| 270 |
+
"learning_rate": 0.0002750156739811912,
|
| 271 |
+
"loss": 0.2593,
|
| 272 |
+
"step": 1300
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 12.21,
|
| 276 |
+
"eval_loss": 0.4765360653400421,
|
| 277 |
+
"eval_runtime": 29.1743,
|
| 278 |
+
"eval_samples_per_second": 7.301,
|
| 279 |
+
"eval_steps_per_second": 0.925,
|
| 280 |
+
"eval_wer": 0.40751845018450183,
|
| 281 |
+
"step": 1300
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 12.68,
|
| 285 |
+
"learning_rate": 0.00027344827586206895,
|
| 286 |
+
"loss": 0.2675,
|
| 287 |
+
"step": 1350
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 13.15,
|
| 291 |
+
"learning_rate": 0.0002718808777429467,
|
| 292 |
+
"loss": 0.2433,
|
| 293 |
+
"step": 1400
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 13.15,
|
| 297 |
+
"eval_loss": 0.4478711187839508,
|
| 298 |
+
"eval_runtime": 29.2927,
|
| 299 |
+
"eval_samples_per_second": 7.271,
|
| 300 |
+
"eval_steps_per_second": 0.922,
|
| 301 |
+
"eval_wer": 0.3863007380073801,
|
| 302 |
+
"step": 1400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 13.62,
|
| 306 |
+
"learning_rate": 0.00027031347962382443,
|
| 307 |
+
"loss": 0.2233,
|
| 308 |
+
"step": 1450
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 14.08,
|
| 312 |
+
"learning_rate": 0.0002687460815047022,
|
| 313 |
+
"loss": 0.23,
|
| 314 |
+
"step": 1500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 14.08,
|
| 318 |
+
"eval_loss": 0.45760419964790344,
|
| 319 |
+
"eval_runtime": 29.3313,
|
| 320 |
+
"eval_samples_per_second": 7.262,
|
| 321 |
+
"eval_steps_per_second": 0.921,
|
| 322 |
+
"eval_wer": 0.39990774907749077,
|
| 323 |
+
"step": 1500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 14.55,
|
| 327 |
+
"learning_rate": 0.0002671786833855799,
|
| 328 |
+
"loss": 0.2054,
|
| 329 |
+
"step": 1550
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 15.02,
|
| 333 |
+
"learning_rate": 0.00026561128526645766,
|
| 334 |
+
"loss": 0.2077,
|
| 335 |
+
"step": 1600
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 15.02,
|
| 339 |
+
"eval_loss": 0.4995218515396118,
|
| 340 |
+
"eval_runtime": 29.3519,
|
| 341 |
+
"eval_samples_per_second": 7.257,
|
| 342 |
+
"eval_steps_per_second": 0.92,
|
| 343 |
+
"eval_wer": 0.39275830258302585,
|
| 344 |
+
"step": 1600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 15.49,
|
| 348 |
+
"learning_rate": 0.0002640438871473354,
|
| 349 |
+
"loss": 0.223,
|
| 350 |
+
"step": 1650
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 15.96,
|
| 354 |
+
"learning_rate": 0.00026247648902821315,
|
| 355 |
+
"loss": 0.1775,
|
| 356 |
+
"step": 1700
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 15.96,
|
| 360 |
+
"eval_loss": 0.4842701554298401,
|
| 361 |
+
"eval_runtime": 29.4414,
|
| 362 |
+
"eval_samples_per_second": 7.235,
|
| 363 |
+
"eval_steps_per_second": 0.917,
|
| 364 |
+
"eval_wer": 0.39321955719557194,
|
| 365 |
+
"step": 1700
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 16.43,
|
| 369 |
+
"learning_rate": 0.0002609090909090909,
|
| 370 |
+
"loss": 0.1818,
|
| 371 |
+
"step": 1750
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 16.9,
|
| 375 |
+
"learning_rate": 0.00025934169278996863,
|
| 376 |
+
"loss": 0.1893,
|
| 377 |
+
"step": 1800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 16.9,
|
| 381 |
+
"eval_loss": 0.47495996952056885,
|
| 382 |
+
"eval_runtime": 29.1664,
|
| 383 |
+
"eval_samples_per_second": 7.303,
|
| 384 |
+
"eval_steps_per_second": 0.926,
|
| 385 |
+
"eval_wer": 0.3768450184501845,
|
| 386 |
+
"step": 1800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 17.37,
|
| 390 |
+
"learning_rate": 0.0002577742946708464,
|
| 391 |
+
"loss": 0.1805,
|
| 392 |
+
"step": 1850
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 17.84,
|
| 396 |
+
"learning_rate": 0.0002562068965517241,
|
| 397 |
+
"loss": 0.1708,
|
| 398 |
+
"step": 1900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 17.84,
|
| 402 |
+
"eval_loss": 0.49353307485580444,
|
| 403 |
+
"eval_runtime": 28.8557,
|
| 404 |
+
"eval_samples_per_second": 7.382,
|
| 405 |
+
"eval_steps_per_second": 0.936,
|
| 406 |
+
"eval_wer": 0.3798431734317343,
|
| 407 |
+
"step": 1900
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 18.31,
|
| 411 |
+
"learning_rate": 0.00025463949843260186,
|
| 412 |
+
"loss": 0.1646,
|
| 413 |
+
"step": 1950
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 18.78,
|
| 417 |
+
"learning_rate": 0.0002530721003134796,
|
| 418 |
+
"loss": 0.1502,
|
| 419 |
+
"step": 2000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 18.78,
|
| 423 |
+
"eval_loss": 0.5329748392105103,
|
| 424 |
+
"eval_runtime": 28.897,
|
| 425 |
+
"eval_samples_per_second": 7.371,
|
| 426 |
+
"eval_steps_per_second": 0.934,
|
| 427 |
+
"eval_wer": 0.3886070110701107,
|
| 428 |
+
"step": 2000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 19.25,
|
| 432 |
+
"learning_rate": 0.00025150470219435734,
|
| 433 |
+
"loss": 0.1508,
|
| 434 |
+
"step": 2050
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 19.72,
|
| 438 |
+
"learning_rate": 0.0002499373040752351,
|
| 439 |
+
"loss": 0.1498,
|
| 440 |
+
"step": 2100
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 19.72,
|
| 444 |
+
"eval_loss": 0.5085766315460205,
|
| 445 |
+
"eval_runtime": 28.7784,
|
| 446 |
+
"eval_samples_per_second": 7.401,
|
| 447 |
+
"eval_steps_per_second": 0.938,
|
| 448 |
+
"eval_wer": 0.37961254612546125,
|
| 449 |
+
"step": 2100
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 20.19,
|
| 453 |
+
"learning_rate": 0.00024836990595611283,
|
| 454 |
+
"loss": 0.1524,
|
| 455 |
+
"step": 2150
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 20.66,
|
| 459 |
+
"learning_rate": 0.00024680250783699057,
|
| 460 |
+
"loss": 0.1517,
|
| 461 |
+
"step": 2200
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 20.66,
|
| 465 |
+
"eval_loss": 0.5103166699409485,
|
| 466 |
+
"eval_runtime": 28.7711,
|
| 467 |
+
"eval_samples_per_second": 7.403,
|
| 468 |
+
"eval_steps_per_second": 0.938,
|
| 469 |
+
"eval_wer": 0.3828413284132841,
|
| 470 |
+
"step": 2200
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 21.13,
|
| 474 |
+
"learning_rate": 0.0002452351097178683,
|
| 475 |
+
"loss": 0.1637,
|
| 476 |
+
"step": 2250
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 21.6,
|
| 480 |
+
"learning_rate": 0.00024366771159874606,
|
| 481 |
+
"loss": 0.1376,
|
| 482 |
+
"step": 2300
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 21.6,
|
| 486 |
+
"eval_loss": 0.5393425226211548,
|
| 487 |
+
"eval_runtime": 28.8325,
|
| 488 |
+
"eval_samples_per_second": 7.388,
|
| 489 |
+
"eval_steps_per_second": 0.936,
|
| 490 |
+
"eval_wer": 0.375,
|
| 491 |
+
"step": 2300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 22.07,
|
| 495 |
+
"learning_rate": 0.00024210031347962383,
|
| 496 |
+
"loss": 0.1332,
|
| 497 |
+
"step": 2350
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 22.54,
|
| 501 |
+
"learning_rate": 0.00024053291536050154,
|
| 502 |
+
"loss": 0.12,
|
| 503 |
+
"step": 2400
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 22.54,
|
| 507 |
+
"eval_loss": 0.550984263420105,
|
| 508 |
+
"eval_runtime": 28.8353,
|
| 509 |
+
"eval_samples_per_second": 7.387,
|
| 510 |
+
"eval_steps_per_second": 0.936,
|
| 511 |
+
"eval_wer": 0.37915129151291516,
|
| 512 |
+
"step": 2400
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 23.0,
|
| 516 |
+
"learning_rate": 0.0002389655172413793,
|
| 517 |
+
"loss": 0.1428,
|
| 518 |
+
"step": 2450
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 23.47,
|
| 522 |
+
"learning_rate": 0.00023739811912225703,
|
| 523 |
+
"loss": 0.1293,
|
| 524 |
+
"step": 2500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 23.47,
|
| 528 |
+
"eval_loss": 0.5512908101081848,
|
| 529 |
+
"eval_runtime": 28.7565,
|
| 530 |
+
"eval_samples_per_second": 7.407,
|
| 531 |
+
"eval_steps_per_second": 0.939,
|
| 532 |
+
"eval_wer": 0.3731549815498155,
|
| 533 |
+
"step": 2500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 23.94,
|
| 537 |
+
"learning_rate": 0.00023583072100313477,
|
| 538 |
+
"loss": 0.1329,
|
| 539 |
+
"step": 2550
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 24.41,
|
| 543 |
+
"learning_rate": 0.0002342633228840125,
|
| 544 |
+
"loss": 0.1228,
|
| 545 |
+
"step": 2600
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 24.41,
|
| 549 |
+
"eval_loss": 0.5537130236625671,
|
| 550 |
+
"eval_runtime": 28.9207,
|
| 551 |
+
"eval_samples_per_second": 7.365,
|
| 552 |
+
"eval_steps_per_second": 0.934,
|
| 553 |
+
"eval_wer": 0.37615313653136534,
|
| 554 |
+
"step": 2600
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 24.88,
|
| 558 |
+
"learning_rate": 0.00023269592476489026,
|
| 559 |
+
"loss": 0.1232,
|
| 560 |
+
"step": 2650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 25.35,
|
| 564 |
+
"learning_rate": 0.00023112852664576802,
|
| 565 |
+
"loss": 0.1116,
|
| 566 |
+
"step": 2700
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 25.35,
|
| 570 |
+
"eval_loss": 0.557085394859314,
|
| 571 |
+
"eval_runtime": 29.0557,
|
| 572 |
+
"eval_samples_per_second": 7.331,
|
| 573 |
+
"eval_steps_per_second": 0.929,
|
| 574 |
+
"eval_wer": 0.37107933579335795,
|
| 575 |
+
"step": 2700
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 25.82,
|
| 579 |
+
"learning_rate": 0.00022956112852664574,
|
| 580 |
+
"loss": 0.1226,
|
| 581 |
+
"step": 2750
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 26.29,
|
| 585 |
+
"learning_rate": 0.0002279937304075235,
|
| 586 |
+
"loss": 0.1135,
|
| 587 |
+
"step": 2800
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 26.29,
|
| 591 |
+
"eval_loss": 0.5842786431312561,
|
| 592 |
+
"eval_runtime": 28.9955,
|
| 593 |
+
"eval_samples_per_second": 7.346,
|
| 594 |
+
"eval_steps_per_second": 0.931,
|
| 595 |
+
"eval_wer": 0.37546125461254615,
|
| 596 |
+
"step": 2800
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 26.76,
|
| 600 |
+
"learning_rate": 0.00022642633228840123,
|
| 601 |
+
"loss": 0.1275,
|
| 602 |
+
"step": 2850
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 27.23,
|
| 606 |
+
"learning_rate": 0.00022485893416927897,
|
| 607 |
+
"loss": 0.1107,
|
| 608 |
+
"step": 2900
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 27.23,
|
| 612 |
+
"eval_loss": 0.5824669599533081,
|
| 613 |
+
"eval_runtime": 28.8912,
|
| 614 |
+
"eval_samples_per_second": 7.372,
|
| 615 |
+
"eval_steps_per_second": 0.935,
|
| 616 |
+
"eval_wer": 0.37661439114391143,
|
| 617 |
+
"step": 2900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 27.7,
|
| 621 |
+
"learning_rate": 0.0002232915360501567,
|
| 622 |
+
"loss": 0.1201,
|
| 623 |
+
"step": 2950
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 28.17,
|
| 627 |
+
"learning_rate": 0.00022172413793103445,
|
| 628 |
+
"loss": 0.1086,
|
| 629 |
+
"step": 3000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 28.17,
|
| 633 |
+
"eval_loss": 0.5459932684898376,
|
| 634 |
+
"eval_runtime": 29.3059,
|
| 635 |
+
"eval_samples_per_second": 7.268,
|
| 636 |
+
"eval_steps_per_second": 0.921,
|
| 637 |
+
"eval_wer": 0.37038745387453875,
|
| 638 |
+
"step": 3000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 28.64,
|
| 642 |
+
"learning_rate": 0.00022015673981191222,
|
| 643 |
+
"loss": 0.0977,
|
| 644 |
+
"step": 3050
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 29.11,
|
| 648 |
+
"learning_rate": 0.00021858934169278994,
|
| 649 |
+
"loss": 0.1055,
|
| 650 |
+
"step": 3100
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 29.11,
|
| 654 |
+
"eval_loss": 0.5809206366539001,
|
| 655 |
+
"eval_runtime": 29.2262,
|
| 656 |
+
"eval_samples_per_second": 7.288,
|
| 657 |
+
"eval_steps_per_second": 0.924,
|
| 658 |
+
"eval_wer": 0.3694649446494465,
|
| 659 |
+
"step": 3100
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 29.58,
|
| 663 |
+
"learning_rate": 0.0002170219435736677,
|
| 664 |
+
"loss": 0.1084,
|
| 665 |
+
"step": 3150
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 30.05,
|
| 669 |
+
"learning_rate": 0.00021545454545454542,
|
| 670 |
+
"loss": 0.1139,
|
| 671 |
+
"step": 3200
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 30.05,
|
| 675 |
+
"eval_loss": 0.5718210935592651,
|
| 676 |
+
"eval_runtime": 29.1425,
|
| 677 |
+
"eval_samples_per_second": 7.309,
|
| 678 |
+
"eval_steps_per_second": 0.926,
|
| 679 |
+
"eval_wer": 0.37845940959409596,
|
| 680 |
+
"step": 3200
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 30.52,
|
| 684 |
+
"learning_rate": 0.0002138871473354232,
|
| 685 |
+
"loss": 0.1026,
|
| 686 |
+
"step": 3250
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 30.99,
|
| 690 |
+
"learning_rate": 0.00021231974921630094,
|
| 691 |
+
"loss": 0.1152,
|
| 692 |
+
"step": 3300
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 30.99,
|
| 696 |
+
"eval_loss": 0.584858775138855,
|
| 697 |
+
"eval_runtime": 28.8726,
|
| 698 |
+
"eval_samples_per_second": 7.377,
|
| 699 |
+
"eval_steps_per_second": 0.935,
|
| 700 |
+
"eval_wer": 0.36300738007380073,
|
| 701 |
+
"step": 3300
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 31.46,
|
| 705 |
+
"learning_rate": 0.00021075235109717865,
|
| 706 |
+
"loss": 0.0929,
|
| 707 |
+
"step": 3350
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 31.92,
|
| 711 |
+
"learning_rate": 0.00020918495297805642,
|
| 712 |
+
"loss": 0.0959,
|
| 713 |
+
"step": 3400
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 31.92,
|
| 717 |
+
"eval_loss": 0.5689957737922668,
|
| 718 |
+
"eval_runtime": 29.1034,
|
| 719 |
+
"eval_samples_per_second": 7.319,
|
| 720 |
+
"eval_steps_per_second": 0.928,
|
| 721 |
+
"eval_wer": 0.3706180811808118,
|
| 722 |
+
"step": 3400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 32.39,
|
| 726 |
+
"learning_rate": 0.00020761755485893414,
|
| 727 |
+
"loss": 0.0958,
|
| 728 |
+
"step": 3450
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 32.86,
|
| 732 |
+
"learning_rate": 0.0002060501567398119,
|
| 733 |
+
"loss": 0.0913,
|
| 734 |
+
"step": 3500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 32.86,
|
| 738 |
+
"eval_loss": 0.5901004076004028,
|
| 739 |
+
"eval_runtime": 29.0974,
|
| 740 |
+
"eval_samples_per_second": 7.32,
|
| 741 |
+
"eval_steps_per_second": 0.928,
|
| 742 |
+
"eval_wer": 0.36185424354243545,
|
| 743 |
+
"step": 3500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 33.33,
|
| 747 |
+
"learning_rate": 0.00020448275862068962,
|
| 748 |
+
"loss": 0.0946,
|
| 749 |
+
"step": 3550
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 33.8,
|
| 753 |
+
"learning_rate": 0.0002029153605015674,
|
| 754 |
+
"loss": 0.0876,
|
| 755 |
+
"step": 3600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 33.8,
|
| 759 |
+
"eval_loss": 0.5581740140914917,
|
| 760 |
+
"eval_runtime": 29.0897,
|
| 761 |
+
"eval_samples_per_second": 7.322,
|
| 762 |
+
"eval_steps_per_second": 0.928,
|
| 763 |
+
"eval_wer": 0.3634686346863469,
|
| 764 |
+
"step": 3600
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 34.27,
|
| 768 |
+
"learning_rate": 0.00020134796238244513,
|
| 769 |
+
"loss": 0.094,
|
| 770 |
+
"step": 3650
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 34.74,
|
| 774 |
+
"learning_rate": 0.00019978056426332285,
|
| 775 |
+
"loss": 0.0999,
|
| 776 |
+
"step": 3700
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 34.74,
|
| 780 |
+
"eval_loss": 0.5737707018852234,
|
| 781 |
+
"eval_runtime": 29.1504,
|
| 782 |
+
"eval_samples_per_second": 7.307,
|
| 783 |
+
"eval_steps_per_second": 0.926,
|
| 784 |
+
"eval_wer": 0.36185424354243545,
|
| 785 |
+
"step": 3700
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 35.21,
|
| 789 |
+
"learning_rate": 0.00019821316614420062,
|
| 790 |
+
"loss": 0.0855,
|
| 791 |
+
"step": 3750
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 35.68,
|
| 795 |
+
"learning_rate": 0.00019664576802507833,
|
| 796 |
+
"loss": 0.0844,
|
| 797 |
+
"step": 3800
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 35.68,
|
| 801 |
+
"eval_loss": 0.5892208218574524,
|
| 802 |
+
"eval_runtime": 29.199,
|
| 803 |
+
"eval_samples_per_second": 7.295,
|
| 804 |
+
"eval_steps_per_second": 0.925,
|
| 805 |
+
"eval_wer": 0.3521678966789668,
|
| 806 |
+
"step": 3800
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 36.15,
|
| 810 |
+
"learning_rate": 0.0001950783699059561,
|
| 811 |
+
"loss": 0.0881,
|
| 812 |
+
"step": 3850
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 36.62,
|
| 816 |
+
"learning_rate": 0.00019351097178683382,
|
| 817 |
+
"loss": 0.0774,
|
| 818 |
+
"step": 3900
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 36.62,
|
| 822 |
+
"eval_loss": 0.6038769483566284,
|
| 823 |
+
"eval_runtime": 29.1467,
|
| 824 |
+
"eval_samples_per_second": 7.308,
|
| 825 |
+
"eval_steps_per_second": 0.926,
|
| 826 |
+
"eval_wer": 0.36485239852398527,
|
| 827 |
+
"step": 3900
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 37.09,
|
| 831 |
+
"learning_rate": 0.0001919435736677116,
|
| 832 |
+
"loss": 0.0793,
|
| 833 |
+
"step": 3950
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 37.56,
|
| 837 |
+
"learning_rate": 0.00019037617554858933,
|
| 838 |
+
"loss": 0.0944,
|
| 839 |
+
"step": 4000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 37.56,
|
| 843 |
+
"eval_loss": 0.580478310585022,
|
| 844 |
+
"eval_runtime": 29.0539,
|
| 845 |
+
"eval_samples_per_second": 7.331,
|
| 846 |
+
"eval_steps_per_second": 0.929,
|
| 847 |
+
"eval_wer": 0.36047047970479706,
|
| 848 |
+
"step": 4000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 38.03,
|
| 852 |
+
"learning_rate": 0.00018880877742946707,
|
| 853 |
+
"loss": 0.078,
|
| 854 |
+
"step": 4050
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 38.5,
|
| 858 |
+
"learning_rate": 0.00018724137931034482,
|
| 859 |
+
"loss": 0.0803,
|
| 860 |
+
"step": 4100
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 38.5,
|
| 864 |
+
"eval_loss": 0.5946702361106873,
|
| 865 |
+
"eval_runtime": 29.1085,
|
| 866 |
+
"eval_samples_per_second": 7.317,
|
| 867 |
+
"eval_steps_per_second": 0.928,
|
| 868 |
+
"eval_wer": 0.3514760147601476,
|
| 869 |
+
"step": 4100
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 38.97,
|
| 873 |
+
"learning_rate": 0.00018567398119122253,
|
| 874 |
+
"loss": 0.0779,
|
| 875 |
+
"step": 4150
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 39.44,
|
| 879 |
+
"learning_rate": 0.0001841065830721003,
|
| 880 |
+
"loss": 0.0721,
|
| 881 |
+
"step": 4200
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 39.44,
|
| 885 |
+
"eval_loss": 0.6369604468345642,
|
| 886 |
+
"eval_runtime": 29.163,
|
| 887 |
+
"eval_samples_per_second": 7.304,
|
| 888 |
+
"eval_steps_per_second": 0.926,
|
| 889 |
+
"eval_wer": 0.36531365313653136,
|
| 890 |
+
"step": 4200
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 39.91,
|
| 894 |
+
"learning_rate": 0.00018253918495297804,
|
| 895 |
+
"loss": 0.0783,
|
| 896 |
+
"step": 4250
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 40.38,
|
| 900 |
+
"learning_rate": 0.0001809717868338558,
|
| 901 |
+
"loss": 0.0854,
|
| 902 |
+
"step": 4300
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 40.38,
|
| 906 |
+
"eval_loss": 0.6401636600494385,
|
| 907 |
+
"eval_runtime": 29.0975,
|
| 908 |
+
"eval_samples_per_second": 7.32,
|
| 909 |
+
"eval_steps_per_second": 0.928,
|
| 910 |
+
"eval_wer": 0.3613929889298893,
|
| 911 |
+
"step": 4300
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 40.85,
|
| 915 |
+
"learning_rate": 0.00017940438871473353,
|
| 916 |
+
"loss": 0.0719,
|
| 917 |
+
"step": 4350
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 41.31,
|
| 921 |
+
"learning_rate": 0.00017783699059561127,
|
| 922 |
+
"loss": 0.0775,
|
| 923 |
+
"step": 4400
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 41.31,
|
| 927 |
+
"eval_loss": 0.6300048828125,
|
| 928 |
+
"eval_runtime": 29.124,
|
| 929 |
+
"eval_samples_per_second": 7.314,
|
| 930 |
+
"eval_steps_per_second": 0.927,
|
| 931 |
+
"eval_wer": 0.35724169741697415,
|
| 932 |
+
"step": 4400
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 41.78,
|
| 936 |
+
"learning_rate": 0.00017626959247648901,
|
| 937 |
+
"loss": 0.0747,
|
| 938 |
+
"step": 4450
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 42.25,
|
| 942 |
+
"learning_rate": 0.00017470219435736673,
|
| 943 |
+
"loss": 0.073,
|
| 944 |
+
"step": 4500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 42.25,
|
| 948 |
+
"eval_loss": 0.6028448343276978,
|
| 949 |
+
"eval_runtime": 29.1165,
|
| 950 |
+
"eval_samples_per_second": 7.315,
|
| 951 |
+
"eval_steps_per_second": 0.927,
|
| 952 |
+
"eval_wer": 0.3547047970479705,
|
| 953 |
+
"step": 4500
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 42.72,
|
| 957 |
+
"learning_rate": 0.0001731347962382445,
|
| 958 |
+
"loss": 0.0608,
|
| 959 |
+
"step": 4550
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 43.19,
|
| 963 |
+
"learning_rate": 0.00017156739811912227,
|
| 964 |
+
"loss": 0.0591,
|
| 965 |
+
"step": 4600
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 43.19,
|
| 969 |
+
"eval_loss": 0.6349148750305176,
|
| 970 |
+
"eval_runtime": 28.9365,
|
| 971 |
+
"eval_samples_per_second": 7.361,
|
| 972 |
+
"eval_steps_per_second": 0.933,
|
| 973 |
+
"eval_wer": 0.36531365313653136,
|
| 974 |
+
"step": 4600
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 43.66,
|
| 978 |
+
"learning_rate": 0.00016999999999999999,
|
| 979 |
+
"loss": 0.0766,
|
| 980 |
+
"step": 4650
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 44.13,
|
| 984 |
+
"learning_rate": 0.00016843260188087773,
|
| 985 |
+
"loss": 0.0641,
|
| 986 |
+
"step": 4700
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 44.13,
|
| 990 |
+
"eval_loss": 0.6087113618850708,
|
| 991 |
+
"eval_runtime": 29.0785,
|
| 992 |
+
"eval_samples_per_second": 7.325,
|
| 993 |
+
"eval_steps_per_second": 0.929,
|
| 994 |
+
"eval_wer": 0.35447416974169743,
|
| 995 |
+
"step": 4700
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 44.6,
|
| 999 |
+
"learning_rate": 0.00016686520376175547,
|
| 1000 |
+
"loss": 0.0672,
|
| 1001 |
+
"step": 4750
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 45.07,
|
| 1005 |
+
"learning_rate": 0.0001652978056426332,
|
| 1006 |
+
"loss": 0.0731,
|
| 1007 |
+
"step": 4800
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 45.07,
|
| 1011 |
+
"eval_loss": 0.6003465056419373,
|
| 1012 |
+
"eval_runtime": 29.0263,
|
| 1013 |
+
"eval_samples_per_second": 7.338,
|
| 1014 |
+
"eval_steps_per_second": 0.93,
|
| 1015 |
+
"eval_wer": 0.36300738007380073,
|
| 1016 |
+
"step": 4800
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 45.54,
|
| 1020 |
+
"learning_rate": 0.00016373040752351096,
|
| 1021 |
+
"loss": 0.0658,
|
| 1022 |
+
"step": 4850
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 46.01,
|
| 1026 |
+
"learning_rate": 0.0001621630094043887,
|
| 1027 |
+
"loss": 0.0651,
|
| 1028 |
+
"step": 4900
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 46.01,
|
| 1032 |
+
"eval_loss": 0.6361915469169617,
|
| 1033 |
+
"eval_runtime": 29.5463,
|
| 1034 |
+
"eval_samples_per_second": 7.209,
|
| 1035 |
+
"eval_steps_per_second": 0.914,
|
| 1036 |
+
"eval_wer": 0.34940036900369004,
|
| 1037 |
+
"step": 4900
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 46.48,
|
| 1041 |
+
"learning_rate": 0.00016059561128526647,
|
| 1042 |
+
"loss": 0.064,
|
| 1043 |
+
"step": 4950
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 46.95,
|
| 1047 |
+
"learning_rate": 0.00015902821316614418,
|
| 1048 |
+
"loss": 0.0679,
|
| 1049 |
+
"step": 5000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 46.95,
|
| 1053 |
+
"eval_loss": 0.6263274550437927,
|
| 1054 |
+
"eval_runtime": 29.0253,
|
| 1055 |
+
"eval_samples_per_second": 7.338,
|
| 1056 |
+
"eval_steps_per_second": 0.93,
|
| 1057 |
+
"eval_wer": 0.3551660516605166,
|
| 1058 |
+
"step": 5000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 47.42,
|
| 1062 |
+
"learning_rate": 0.00015746081504702193,
|
| 1063 |
+
"loss": 0.0589,
|
| 1064 |
+
"step": 5050
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 47.89,
|
| 1068 |
+
"learning_rate": 0.00015589341692789967,
|
| 1069 |
+
"loss": 0.0716,
|
| 1070 |
+
"step": 5100
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 47.89,
|
| 1074 |
+
"eval_loss": 0.6083901524543762,
|
| 1075 |
+
"eval_runtime": 29.0652,
|
| 1076 |
+
"eval_samples_per_second": 7.328,
|
| 1077 |
+
"eval_steps_per_second": 0.929,
|
| 1078 |
+
"eval_wer": 0.3526291512915129,
|
| 1079 |
+
"step": 5100
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 48.36,
|
| 1083 |
+
"learning_rate": 0.0001543260188087774,
|
| 1084 |
+
"loss": 0.073,
|
| 1085 |
+
"step": 5150
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 48.83,
|
| 1089 |
+
"learning_rate": 0.00015275862068965518,
|
| 1090 |
+
"loss": 0.066,
|
| 1091 |
+
"step": 5200
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 48.83,
|
| 1095 |
+
"eval_loss": 0.6158061027526855,
|
| 1096 |
+
"eval_runtime": 29.3027,
|
| 1097 |
+
"eval_samples_per_second": 7.269,
|
| 1098 |
+
"eval_steps_per_second": 0.921,
|
| 1099 |
+
"eval_wer": 0.3563191881918819,
|
| 1100 |
+
"step": 5200
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 49.3,
|
| 1104 |
+
"learning_rate": 0.0001511912225705329,
|
| 1105 |
+
"loss": 0.0581,
|
| 1106 |
+
"step": 5250
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 49.77,
|
| 1110 |
+
"learning_rate": 0.00014962382445141064,
|
| 1111 |
+
"loss": 0.0564,
|
| 1112 |
+
"step": 5300
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 49.77,
|
| 1116 |
+
"eval_loss": 0.6484010815620422,
|
| 1117 |
+
"eval_runtime": 29.5466,
|
| 1118 |
+
"eval_samples_per_second": 7.209,
|
| 1119 |
+
"eval_steps_per_second": 0.914,
|
| 1120 |
+
"eval_wer": 0.35793357933579334,
|
| 1121 |
+
"step": 5300
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 50.23,
|
| 1125 |
+
"learning_rate": 0.00014805642633228838,
|
| 1126 |
+
"loss": 0.0585,
|
| 1127 |
+
"step": 5350
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 50.7,
|
| 1131 |
+
"learning_rate": 0.00014648902821316615,
|
| 1132 |
+
"loss": 0.0679,
|
| 1133 |
+
"step": 5400
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 50.7,
|
| 1137 |
+
"eval_loss": 0.6405696868896484,
|
| 1138 |
+
"eval_runtime": 29.3581,
|
| 1139 |
+
"eval_samples_per_second": 7.255,
|
| 1140 |
+
"eval_steps_per_second": 0.92,
|
| 1141 |
+
"eval_wer": 0.3549354243542435,
|
| 1142 |
+
"step": 5400
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 51.17,
|
| 1146 |
+
"learning_rate": 0.00014492163009404387,
|
| 1147 |
+
"loss": 0.0573,
|
| 1148 |
+
"step": 5450
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 51.64,
|
| 1152 |
+
"learning_rate": 0.0001433542319749216,
|
| 1153 |
+
"loss": 0.061,
|
| 1154 |
+
"step": 5500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 51.64,
|
| 1158 |
+
"eval_loss": 0.6469581127166748,
|
| 1159 |
+
"eval_runtime": 29.5701,
|
| 1160 |
+
"eval_samples_per_second": 7.203,
|
| 1161 |
+
"eval_steps_per_second": 0.913,
|
| 1162 |
+
"eval_wer": 0.3533210332103321,
|
| 1163 |
+
"step": 5500
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 52.11,
|
| 1167 |
+
"learning_rate": 0.00014178683385579935,
|
| 1168 |
+
"loss": 0.0582,
|
| 1169 |
+
"step": 5550
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 52.58,
|
| 1173 |
+
"learning_rate": 0.0001402194357366771,
|
| 1174 |
+
"loss": 0.0566,
|
| 1175 |
+
"step": 5600
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 52.58,
|
| 1179 |
+
"eval_loss": 0.659860372543335,
|
| 1180 |
+
"eval_runtime": 29.2997,
|
| 1181 |
+
"eval_samples_per_second": 7.27,
|
| 1182 |
+
"eval_steps_per_second": 0.922,
|
| 1183 |
+
"eval_wer": 0.3514760147601476,
|
| 1184 |
+
"step": 5600
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 53.05,
|
| 1188 |
+
"learning_rate": 0.00013865203761755484,
|
| 1189 |
+
"loss": 0.0552,
|
| 1190 |
+
"step": 5650
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 53.52,
|
| 1194 |
+
"learning_rate": 0.0001370846394984326,
|
| 1195 |
+
"loss": 0.0614,
|
| 1196 |
+
"step": 5700
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 53.52,
|
| 1200 |
+
"eval_loss": 0.6916360855102539,
|
| 1201 |
+
"eval_runtime": 29.2037,
|
| 1202 |
+
"eval_samples_per_second": 7.294,
|
| 1203 |
+
"eval_steps_per_second": 0.925,
|
| 1204 |
+
"eval_wer": 0.3570110701107011,
|
| 1205 |
+
"step": 5700
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 53.99,
|
| 1209 |
+
"learning_rate": 0.00013551724137931035,
|
| 1210 |
+
"loss": 0.0588,
|
| 1211 |
+
"step": 5750
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 54.46,
|
| 1215 |
+
"learning_rate": 0.00013394984326018806,
|
| 1216 |
+
"loss": 0.0567,
|
| 1217 |
+
"step": 5800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 54.46,
|
| 1221 |
+
"eval_loss": 0.6793951988220215,
|
| 1222 |
+
"eval_runtime": 29.1822,
|
| 1223 |
+
"eval_samples_per_second": 7.299,
|
| 1224 |
+
"eval_steps_per_second": 0.925,
|
| 1225 |
+
"eval_wer": 0.3503228782287823,
|
| 1226 |
+
"step": 5800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 54.93,
|
| 1230 |
+
"learning_rate": 0.0001323824451410658,
|
| 1231 |
+
"loss": 0.0598,
|
| 1232 |
+
"step": 5850
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 55.4,
|
| 1236 |
+
"learning_rate": 0.00013081504702194355,
|
| 1237 |
+
"loss": 0.0548,
|
| 1238 |
+
"step": 5900
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 55.4,
|
| 1242 |
+
"eval_loss": 0.6582257151603699,
|
| 1243 |
+
"eval_runtime": 29.0712,
|
| 1244 |
+
"eval_samples_per_second": 7.327,
|
| 1245 |
+
"eval_steps_per_second": 0.929,
|
| 1246 |
+
"eval_wer": 0.3556273062730627,
|
| 1247 |
+
"step": 5900
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 55.87,
|
| 1251 |
+
"learning_rate": 0.0001292476489028213,
|
| 1252 |
+
"loss": 0.0587,
|
| 1253 |
+
"step": 5950
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 56.34,
|
| 1257 |
+
"learning_rate": 0.00012768025078369906,
|
| 1258 |
+
"loss": 0.052,
|
| 1259 |
+
"step": 6000
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 56.34,
|
| 1263 |
+
"eval_loss": 0.6707574725151062,
|
| 1264 |
+
"eval_runtime": 29.1671,
|
| 1265 |
+
"eval_samples_per_second": 7.303,
|
| 1266 |
+
"eval_steps_per_second": 0.926,
|
| 1267 |
+
"eval_wer": 0.34870848708487084,
|
| 1268 |
+
"step": 6000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 56.81,
|
| 1272 |
+
"learning_rate": 0.0001261128526645768,
|
| 1273 |
+
"loss": 0.0544,
|
| 1274 |
+
"step": 6050
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 57.28,
|
| 1278 |
+
"learning_rate": 0.00012454545454545455,
|
| 1279 |
+
"loss": 0.058,
|
| 1280 |
+
"step": 6100
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 57.28,
|
| 1284 |
+
"eval_loss": 0.6558502912521362,
|
| 1285 |
+
"eval_runtime": 29.1303,
|
| 1286 |
+
"eval_samples_per_second": 7.312,
|
| 1287 |
+
"eval_steps_per_second": 0.927,
|
| 1288 |
+
"eval_wer": 0.3507841328413284,
|
| 1289 |
+
"step": 6100
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 57.75,
|
| 1293 |
+
"learning_rate": 0.0001229780564263323,
|
| 1294 |
+
"loss": 0.0476,
|
| 1295 |
+
"step": 6150
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 58.22,
|
| 1299 |
+
"learning_rate": 0.00012141065830721002,
|
| 1300 |
+
"loss": 0.0558,
|
| 1301 |
+
"step": 6200
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 58.22,
|
| 1305 |
+
"eval_loss": 0.6627236008644104,
|
| 1306 |
+
"eval_runtime": 29.1431,
|
| 1307 |
+
"eval_samples_per_second": 7.309,
|
| 1308 |
+
"eval_steps_per_second": 0.926,
|
| 1309 |
+
"eval_wer": 0.35239852398523985,
|
| 1310 |
+
"step": 6200
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 58.69,
|
| 1314 |
+
"learning_rate": 0.00011984326018808776,
|
| 1315 |
+
"loss": 0.0524,
|
| 1316 |
+
"step": 6250
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 59.15,
|
| 1320 |
+
"learning_rate": 0.0001182758620689655,
|
| 1321 |
+
"loss": 0.0567,
|
| 1322 |
+
"step": 6300
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 59.15,
|
| 1326 |
+
"eval_loss": 0.6643950343132019,
|
| 1327 |
+
"eval_runtime": 29.149,
|
| 1328 |
+
"eval_samples_per_second": 7.307,
|
| 1329 |
+
"eval_steps_per_second": 0.926,
|
| 1330 |
+
"eval_wer": 0.3512453874538745,
|
| 1331 |
+
"step": 6300
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 59.62,
|
| 1335 |
+
"learning_rate": 0.00011670846394984326,
|
| 1336 |
+
"loss": 0.048,
|
| 1337 |
+
"step": 6350
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 60.09,
|
| 1341 |
+
"learning_rate": 0.000115141065830721,
|
| 1342 |
+
"loss": 0.0551,
|
| 1343 |
+
"step": 6400
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 60.09,
|
| 1347 |
+
"eval_loss": 0.6813772916793823,
|
| 1348 |
+
"eval_runtime": 29.2298,
|
| 1349 |
+
"eval_samples_per_second": 7.287,
|
| 1350 |
+
"eval_steps_per_second": 0.924,
|
| 1351 |
+
"eval_wer": 0.3484778597785978,
|
| 1352 |
+
"step": 6400
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 60.56,
|
| 1356 |
+
"learning_rate": 0.00011357366771159873,
|
| 1357 |
+
"loss": 0.0515,
|
| 1358 |
+
"step": 6450
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 61.03,
|
| 1362 |
+
"learning_rate": 0.00011200626959247647,
|
| 1363 |
+
"loss": 0.0467,
|
| 1364 |
+
"step": 6500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 61.03,
|
| 1368 |
+
"eval_loss": 0.7115447521209717,
|
| 1369 |
+
"eval_runtime": 29.1151,
|
| 1370 |
+
"eval_samples_per_second": 7.316,
|
| 1371 |
+
"eval_steps_per_second": 0.927,
|
| 1372 |
+
"eval_wer": 0.34801660516605165,
|
| 1373 |
+
"step": 6500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 61.5,
|
| 1377 |
+
"learning_rate": 0.00011043887147335422,
|
| 1378 |
+
"loss": 0.0472,
|
| 1379 |
+
"step": 6550
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 61.97,
|
| 1383 |
+
"learning_rate": 0.00010887147335423196,
|
| 1384 |
+
"loss": 0.0487,
|
| 1385 |
+
"step": 6600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 61.97,
|
| 1389 |
+
"eval_loss": 0.6461969614028931,
|
| 1390 |
+
"eval_runtime": 29.2046,
|
| 1391 |
+
"eval_samples_per_second": 7.293,
|
| 1392 |
+
"eval_steps_per_second": 0.925,
|
| 1393 |
+
"eval_wer": 0.34870848708487084,
|
| 1394 |
+
"step": 6600
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 62.44,
|
| 1398 |
+
"learning_rate": 0.00010730407523510972,
|
| 1399 |
+
"loss": 0.0497,
|
| 1400 |
+
"step": 6650
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 62.91,
|
| 1404 |
+
"learning_rate": 0.00010573667711598746,
|
| 1405 |
+
"loss": 0.0554,
|
| 1406 |
+
"step": 6700
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 62.91,
|
| 1410 |
+
"eval_loss": 0.6031398773193359,
|
| 1411 |
+
"eval_runtime": 29.1799,
|
| 1412 |
+
"eval_samples_per_second": 7.3,
|
| 1413 |
+
"eval_steps_per_second": 0.925,
|
| 1414 |
+
"eval_wer": 0.34686346863468637,
|
| 1415 |
+
"step": 6700
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 63.38,
|
| 1419 |
+
"learning_rate": 0.0001041692789968652,
|
| 1420 |
+
"loss": 0.0435,
|
| 1421 |
+
"step": 6750
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 63.85,
|
| 1425 |
+
"learning_rate": 0.00010260188087774294,
|
| 1426 |
+
"loss": 0.0513,
|
| 1427 |
+
"step": 6800
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 63.85,
|
| 1431 |
+
"eval_loss": 0.6592887043952942,
|
| 1432 |
+
"eval_runtime": 29.1723,
|
| 1433 |
+
"eval_samples_per_second": 7.301,
|
| 1434 |
+
"eval_steps_per_second": 0.926,
|
| 1435 |
+
"eval_wer": 0.34801660516605165,
|
| 1436 |
+
"step": 6800
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 64.32,
|
| 1440 |
+
"learning_rate": 0.00010103448275862067,
|
| 1441 |
+
"loss": 0.0501,
|
| 1442 |
+
"step": 6850
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 64.79,
|
| 1446 |
+
"learning_rate": 9.946708463949841e-05,
|
| 1447 |
+
"loss": 0.0484,
|
| 1448 |
+
"step": 6900
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 64.79,
|
| 1452 |
+
"eval_loss": 0.6585196852684021,
|
| 1453 |
+
"eval_runtime": 29.3855,
|
| 1454 |
+
"eval_samples_per_second": 7.248,
|
| 1455 |
+
"eval_steps_per_second": 0.919,
|
| 1456 |
+
"eval_wer": 0.34432656826568264,
|
| 1457 |
+
"step": 6900
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 65.26,
|
| 1461 |
+
"learning_rate": 9.789968652037617e-05,
|
| 1462 |
+
"loss": 0.0449,
|
| 1463 |
+
"step": 6950
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 65.73,
|
| 1467 |
+
"learning_rate": 9.633228840125391e-05,
|
| 1468 |
+
"loss": 0.043,
|
| 1469 |
+
"step": 7000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 65.73,
|
| 1473 |
+
"eval_loss": 0.6810497641563416,
|
| 1474 |
+
"eval_runtime": 29.273,
|
| 1475 |
+
"eval_samples_per_second": 7.276,
|
| 1476 |
+
"eval_steps_per_second": 0.922,
|
| 1477 |
+
"eval_wer": 0.3404059040590406,
|
| 1478 |
+
"step": 7000
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 66.2,
|
| 1482 |
+
"learning_rate": 9.476489028213166e-05,
|
| 1483 |
+
"loss": 0.0494,
|
| 1484 |
+
"step": 7050
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 66.67,
|
| 1488 |
+
"learning_rate": 9.31974921630094e-05,
|
| 1489 |
+
"loss": 0.0487,
|
| 1490 |
+
"step": 7100
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 66.67,
|
| 1494 |
+
"eval_loss": 0.6745949387550354,
|
| 1495 |
+
"eval_runtime": 29.2853,
|
| 1496 |
+
"eval_samples_per_second": 7.273,
|
| 1497 |
+
"eval_steps_per_second": 0.922,
|
| 1498 |
+
"eval_wer": 0.3507841328413284,
|
| 1499 |
+
"step": 7100
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 67.14,
|
| 1503 |
+
"learning_rate": 9.163009404388714e-05,
|
| 1504 |
+
"loss": 0.0428,
|
| 1505 |
+
"step": 7150
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 67.61,
|
| 1509 |
+
"learning_rate": 9.006269592476488e-05,
|
| 1510 |
+
"loss": 0.0421,
|
| 1511 |
+
"step": 7200
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 67.61,
|
| 1515 |
+
"eval_loss": 0.6884201169013977,
|
| 1516 |
+
"eval_runtime": 29.0927,
|
| 1517 |
+
"eval_samples_per_second": 7.321,
|
| 1518 |
+
"eval_steps_per_second": 0.928,
|
| 1519 |
+
"eval_wer": 0.3514760147601476,
|
| 1520 |
+
"step": 7200
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 68.08,
|
| 1524 |
+
"learning_rate": 8.849529780564261e-05,
|
| 1525 |
+
"loss": 0.045,
|
| 1526 |
+
"step": 7250
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 68.54,
|
| 1530 |
+
"learning_rate": 8.692789968652037e-05,
|
| 1531 |
+
"loss": 0.0475,
|
| 1532 |
+
"step": 7300
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 68.54,
|
| 1536 |
+
"eval_loss": 0.6797507405281067,
|
| 1537 |
+
"eval_runtime": 29.0724,
|
| 1538 |
+
"eval_samples_per_second": 7.327,
|
| 1539 |
+
"eval_steps_per_second": 0.929,
|
| 1540 |
+
"eval_wer": 0.34801660516605165,
|
| 1541 |
+
"step": 7300
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 69.01,
|
| 1545 |
+
"learning_rate": 8.536050156739811e-05,
|
| 1546 |
+
"loss": 0.0475,
|
| 1547 |
+
"step": 7350
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 69.48,
|
| 1551 |
+
"learning_rate": 8.379310344827585e-05,
|
| 1552 |
+
"loss": 0.0423,
|
| 1553 |
+
"step": 7400
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 69.48,
|
| 1557 |
+
"eval_loss": 0.7019987106323242,
|
| 1558 |
+
"eval_runtime": 29.1029,
|
| 1559 |
+
"eval_samples_per_second": 7.319,
|
| 1560 |
+
"eval_steps_per_second": 0.928,
|
| 1561 |
+
"eval_wer": 0.3484778597785978,
|
| 1562 |
+
"step": 7400
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 69.95,
|
| 1566 |
+
"learning_rate": 8.22257053291536e-05,
|
| 1567 |
+
"loss": 0.0418,
|
| 1568 |
+
"step": 7450
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 70.42,
|
| 1572 |
+
"learning_rate": 8.065830721003134e-05,
|
| 1573 |
+
"loss": 0.0483,
|
| 1574 |
+
"step": 7500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 70.42,
|
| 1578 |
+
"eval_loss": 0.6741328835487366,
|
| 1579 |
+
"eval_runtime": 29.3432,
|
| 1580 |
+
"eval_samples_per_second": 7.259,
|
| 1581 |
+
"eval_steps_per_second": 0.92,
|
| 1582 |
+
"eval_wer": 0.3482472324723247,
|
| 1583 |
+
"step": 7500
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 70.89,
|
| 1587 |
+
"learning_rate": 7.909090909090908e-05,
|
| 1588 |
+
"loss": 0.0437,
|
| 1589 |
+
"step": 7550
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 71.36,
|
| 1593 |
+
"learning_rate": 7.752351097178684e-05,
|
| 1594 |
+
"loss": 0.0435,
|
| 1595 |
+
"step": 7600
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 71.36,
|
| 1599 |
+
"eval_loss": 0.693108081817627,
|
| 1600 |
+
"eval_runtime": 29.4586,
|
| 1601 |
+
"eval_samples_per_second": 7.23,
|
| 1602 |
+
"eval_steps_per_second": 0.917,
|
| 1603 |
+
"eval_wer": 0.34086715867158673,
|
| 1604 |
+
"step": 7600
|
| 1605 |
+
}
|
| 1606 |
+
],
|
| 1607 |
+
"max_steps": 10070,
|
| 1608 |
+
"num_train_epochs": 95,
|
| 1609 |
+
"total_flos": 3.519219302842646e+19,
|
| 1610 |
+
"trial_name": null,
|
| 1611 |
+
"trial_params": null
|
| 1612 |
+
}
|
checkpoint-7600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9c07bf6381da61cf3041d6ee03574663f30281259ab3dbc7d89dfd005e6b99c
|
| 3 |
+
size 4024
|