Commit
•
fea7d21
1
Parent(s):
27892bb
Add model files
Browse files- config.json +76 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- trainer_state.json +2718 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"Wav2Vec2ForCTC"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"conv_bias": true,
|
11 |
+
"conv_dim": [
|
12 |
+
512,
|
13 |
+
512,
|
14 |
+
512,
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512
|
19 |
+
],
|
20 |
+
"conv_kernel": [
|
21 |
+
10,
|
22 |
+
3,
|
23 |
+
3,
|
24 |
+
3,
|
25 |
+
3,
|
26 |
+
2,
|
27 |
+
2
|
28 |
+
],
|
29 |
+
"conv_stride": [
|
30 |
+
5,
|
31 |
+
2,
|
32 |
+
2,
|
33 |
+
2,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2
|
37 |
+
],
|
38 |
+
"ctc_loss_reduction": "mean",
|
39 |
+
"ctc_zero_infinity": false,
|
40 |
+
"do_stable_layer_norm": true,
|
41 |
+
"eos_token_id": 2,
|
42 |
+
"feat_extract_activation": "gelu",
|
43 |
+
"feat_extract_dropout": 0.0,
|
44 |
+
"feat_extract_norm": "layer",
|
45 |
+
"feat_proj_dropout": 0.0,
|
46 |
+
"final_dropout": 0.0,
|
47 |
+
"gradient_checkpointing": true,
|
48 |
+
"hidden_act": "gelu",
|
49 |
+
"hidden_dropout": 0.1,
|
50 |
+
"hidden_size": 1024,
|
51 |
+
"initializer_range": 0.02,
|
52 |
+
"intermediate_size": 4096,
|
53 |
+
"layer_norm_eps": 1e-05,
|
54 |
+
"layerdrop": 0.1,
|
55 |
+
"mask_channel_length": 10,
|
56 |
+
"mask_channel_min_space": 1,
|
57 |
+
"mask_channel_other": 0.0,
|
58 |
+
"mask_channel_prob": 0.0,
|
59 |
+
"mask_channel_selection": "static",
|
60 |
+
"mask_feature_length": 10,
|
61 |
+
"mask_feature_prob": 0.0,
|
62 |
+
"mask_time_length": 10,
|
63 |
+
"mask_time_min_space": 1,
|
64 |
+
"mask_time_other": 0.0,
|
65 |
+
"mask_time_prob": 0.05,
|
66 |
+
"mask_time_selection": "static",
|
67 |
+
"model_type": "wav2vec2",
|
68 |
+
"num_attention_heads": 16,
|
69 |
+
"num_conv_pos_embedding_groups": 16,
|
70 |
+
"num_conv_pos_embeddings": 128,
|
71 |
+
"num_feat_extract_layers": 7,
|
72 |
+
"num_hidden_layers": 24,
|
73 |
+
"pad_token_id": 53,
|
74 |
+
"transformers_version": "4.4.0",
|
75 |
+
"vocab_size": 54
|
76 |
+
}
|
preprocessor_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"feature_size": 1,
|
4 |
+
"padding_side": "right",
|
5 |
+
"padding_value": 0.0,
|
6 |
+
"return_attention_mask": true,
|
7 |
+
"sampling_rate": 16000
|
8 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:caa727dfccdbdbdb8c02f9829f7c253951664e871bb0d197fc01c6e3c282ded4
|
3 |
+
size 1262150969
|
trainer_state.json
ADDED
@@ -0,0 +1,2718 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 15.793780687397708,
|
5 |
+
"global_step": 19300,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.08,
|
12 |
+
"learning_rate": 0.0001,
|
13 |
+
"loss": 12.1859,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.08,
|
18 |
+
"eval_loss": 3.897115468978882,
|
19 |
+
"eval_runtime": 18.0759,
|
20 |
+
"eval_samples_per_second": 5.532,
|
21 |
+
"eval_wer": 1.0,
|
22 |
+
"step": 100
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"epoch": 0.16,
|
26 |
+
"learning_rate": 9.997271487039564e-05,
|
27 |
+
"loss": 3.3843,
|
28 |
+
"step": 200
|
29 |
+
},
|
30 |
+
{
|
31 |
+
"epoch": 0.16,
|
32 |
+
"eval_loss": 3.5120508670806885,
|
33 |
+
"eval_runtime": 18.1777,
|
34 |
+
"eval_samples_per_second": 5.501,
|
35 |
+
"eval_wer": 1.0,
|
36 |
+
"step": 200
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.25,
|
40 |
+
"learning_rate": 9.994542974079127e-05,
|
41 |
+
"loss": 3.3224,
|
42 |
+
"step": 300
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.25,
|
46 |
+
"eval_loss": 3.539235830307007,
|
47 |
+
"eval_runtime": 18.2559,
|
48 |
+
"eval_samples_per_second": 5.478,
|
49 |
+
"eval_wer": 1.0,
|
50 |
+
"step": 300
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.33,
|
54 |
+
"learning_rate": 9.99181446111869e-05,
|
55 |
+
"loss": 3.34,
|
56 |
+
"step": 400
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.33,
|
60 |
+
"eval_loss": 3.492466926574707,
|
61 |
+
"eval_runtime": 18.4309,
|
62 |
+
"eval_samples_per_second": 5.426,
|
63 |
+
"eval_wer": 1.0,
|
64 |
+
"step": 400
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 0.41,
|
68 |
+
"learning_rate": 9.989085948158254e-05,
|
69 |
+
"loss": 3.2825,
|
70 |
+
"step": 500
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"epoch": 0.41,
|
74 |
+
"eval_loss": 3.376338481903076,
|
75 |
+
"eval_runtime": 18.3713,
|
76 |
+
"eval_samples_per_second": 5.443,
|
77 |
+
"eval_wer": 1.0,
|
78 |
+
"step": 500
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 0.49,
|
82 |
+
"learning_rate": 9.986357435197817e-05,
|
83 |
+
"loss": 3.2728,
|
84 |
+
"step": 600
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.49,
|
88 |
+
"eval_loss": 3.3126320838928223,
|
89 |
+
"eval_runtime": 18.6895,
|
90 |
+
"eval_samples_per_second": 5.351,
|
91 |
+
"eval_wer": 1.0,
|
92 |
+
"step": 600
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.57,
|
96 |
+
"learning_rate": 9.983628922237382e-05,
|
97 |
+
"loss": 3.2327,
|
98 |
+
"step": 700
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.57,
|
102 |
+
"eval_loss": 3.1550354957580566,
|
103 |
+
"eval_runtime": 18.5514,
|
104 |
+
"eval_samples_per_second": 5.39,
|
105 |
+
"eval_wer": 1.0,
|
106 |
+
"step": 700
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.65,
|
110 |
+
"learning_rate": 9.980900409276944e-05,
|
111 |
+
"loss": 2.7258,
|
112 |
+
"step": 800
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.65,
|
116 |
+
"eval_loss": 1.9835596084594727,
|
117 |
+
"eval_runtime": 18.3986,
|
118 |
+
"eval_samples_per_second": 5.435,
|
119 |
+
"eval_wer": 0.9820554649265906,
|
120 |
+
"step": 800
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.74,
|
124 |
+
"learning_rate": 9.978171896316508e-05,
|
125 |
+
"loss": 1.5581,
|
126 |
+
"step": 900
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 0.74,
|
130 |
+
"eval_loss": 1.2180073261260986,
|
131 |
+
"eval_runtime": 18.413,
|
132 |
+
"eval_samples_per_second": 5.431,
|
133 |
+
"eval_wer": 1.0163132137030995,
|
134 |
+
"step": 900
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.82,
|
138 |
+
"learning_rate": 9.975443383356072e-05,
|
139 |
+
"loss": 1.1164,
|
140 |
+
"step": 1000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.82,
|
144 |
+
"eval_loss": 0.966360867023468,
|
145 |
+
"eval_runtime": 18.5352,
|
146 |
+
"eval_samples_per_second": 5.395,
|
147 |
+
"eval_wer": 0.8711256117455138,
|
148 |
+
"step": 1000
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 0.9,
|
152 |
+
"learning_rate": 9.972714870395635e-05,
|
153 |
+
"loss": 0.9322,
|
154 |
+
"step": 1100
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 0.9,
|
158 |
+
"eval_loss": 0.8568554520606995,
|
159 |
+
"eval_runtime": 18.1123,
|
160 |
+
"eval_samples_per_second": 5.521,
|
161 |
+
"eval_wer": 0.831973898858075,
|
162 |
+
"step": 1100
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.98,
|
166 |
+
"learning_rate": 9.969986357435198e-05,
|
167 |
+
"loss": 0.8174,
|
168 |
+
"step": 1200
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.98,
|
172 |
+
"eval_loss": 0.8210877180099487,
|
173 |
+
"eval_runtime": 18.2233,
|
174 |
+
"eval_samples_per_second": 5.487,
|
175 |
+
"eval_wer": 0.8058727569331158,
|
176 |
+
"step": 1200
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 1.06,
|
180 |
+
"learning_rate": 9.967257844474762e-05,
|
181 |
+
"loss": 0.7625,
|
182 |
+
"step": 1300
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 1.06,
|
186 |
+
"eval_loss": 0.8484710454940796,
|
187 |
+
"eval_runtime": 18.3237,
|
188 |
+
"eval_samples_per_second": 5.457,
|
189 |
+
"eval_wer": 0.8058727569331158,
|
190 |
+
"step": 1300
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 1.15,
|
194 |
+
"learning_rate": 9.964529331514325e-05,
|
195 |
+
"loss": 0.7211,
|
196 |
+
"step": 1400
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 1.15,
|
200 |
+
"eval_loss": 0.7573192119598389,
|
201 |
+
"eval_runtime": 18.7142,
|
202 |
+
"eval_samples_per_second": 5.344,
|
203 |
+
"eval_wer": 0.7732463295269169,
|
204 |
+
"step": 1400
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 1.23,
|
208 |
+
"learning_rate": 9.96180081855389e-05,
|
209 |
+
"loss": 0.6892,
|
210 |
+
"step": 1500
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 1.23,
|
214 |
+
"eval_loss": 0.7421054244041443,
|
215 |
+
"eval_runtime": 18.8747,
|
216 |
+
"eval_samples_per_second": 5.298,
|
217 |
+
"eval_wer": 0.768352365415987,
|
218 |
+
"step": 1500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 1.31,
|
222 |
+
"learning_rate": 9.959072305593452e-05,
|
223 |
+
"loss": 0.6573,
|
224 |
+
"step": 1600
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 1.31,
|
228 |
+
"eval_loss": 0.6711151599884033,
|
229 |
+
"eval_runtime": 18.658,
|
230 |
+
"eval_samples_per_second": 5.36,
|
231 |
+
"eval_wer": 0.7422512234910277,
|
232 |
+
"step": 1600
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 1.39,
|
236 |
+
"learning_rate": 9.956343792633015e-05,
|
237 |
+
"loss": 0.6402,
|
238 |
+
"step": 1700
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.39,
|
242 |
+
"eval_loss": 0.6495039463043213,
|
243 |
+
"eval_runtime": 18.212,
|
244 |
+
"eval_samples_per_second": 5.491,
|
245 |
+
"eval_wer": 0.7161500815660685,
|
246 |
+
"step": 1700
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 1.47,
|
250 |
+
"learning_rate": 9.95361527967258e-05,
|
251 |
+
"loss": 0.6129,
|
252 |
+
"step": 1800
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 1.47,
|
256 |
+
"eval_loss": 0.6313944458961487,
|
257 |
+
"eval_runtime": 18.1993,
|
258 |
+
"eval_samples_per_second": 5.495,
|
259 |
+
"eval_wer": 0.6965742251223491,
|
260 |
+
"step": 1800
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.55,
|
264 |
+
"learning_rate": 9.950886766712142e-05,
|
265 |
+
"loss": 0.6074,
|
266 |
+
"step": 1900
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.55,
|
270 |
+
"eval_loss": 0.6353416442871094,
|
271 |
+
"eval_runtime": 18.4085,
|
272 |
+
"eval_samples_per_second": 5.432,
|
273 |
+
"eval_wer": 0.7030995106035889,
|
274 |
+
"step": 1900
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.64,
|
278 |
+
"learning_rate": 9.948158253751706e-05,
|
279 |
+
"loss": 0.5688,
|
280 |
+
"step": 2000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.64,
|
284 |
+
"eval_loss": 0.6229403018951416,
|
285 |
+
"eval_runtime": 18.9507,
|
286 |
+
"eval_samples_per_second": 5.277,
|
287 |
+
"eval_wer": 0.6900489396411092,
|
288 |
+
"step": 2000
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.72,
|
292 |
+
"learning_rate": 9.945429740791268e-05,
|
293 |
+
"loss": 0.5522,
|
294 |
+
"step": 2100
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.72,
|
298 |
+
"eval_loss": 0.6387014985084534,
|
299 |
+
"eval_runtime": 18.1593,
|
300 |
+
"eval_samples_per_second": 5.507,
|
301 |
+
"eval_wer": 0.6916802610114192,
|
302 |
+
"step": 2100
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 1.8,
|
306 |
+
"learning_rate": 9.942701227830833e-05,
|
307 |
+
"loss": 0.5414,
|
308 |
+
"step": 2200
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 1.8,
|
312 |
+
"eval_loss": 0.5740896463394165,
|
313 |
+
"eval_runtime": 18.1093,
|
314 |
+
"eval_samples_per_second": 5.522,
|
315 |
+
"eval_wer": 0.6606851549755302,
|
316 |
+
"step": 2200
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.88,
|
320 |
+
"learning_rate": 9.939972714870396e-05,
|
321 |
+
"loss": 0.5227,
|
322 |
+
"step": 2300
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.88,
|
326 |
+
"eval_loss": 0.5425416231155396,
|
327 |
+
"eval_runtime": 18.0421,
|
328 |
+
"eval_samples_per_second": 5.543,
|
329 |
+
"eval_wer": 0.6427406199021207,
|
330 |
+
"step": 2300
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.96,
|
334 |
+
"learning_rate": 9.93724420190996e-05,
|
335 |
+
"loss": 0.5233,
|
336 |
+
"step": 2400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 1.96,
|
340 |
+
"eval_loss": 0.5257930755615234,
|
341 |
+
"eval_runtime": 17.9989,
|
342 |
+
"eval_samples_per_second": 5.556,
|
343 |
+
"eval_wer": 0.6655791190864601,
|
344 |
+
"step": 2400
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 2.05,
|
348 |
+
"learning_rate": 9.934515688949523e-05,
|
349 |
+
"loss": 0.5019,
|
350 |
+
"step": 2500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 2.05,
|
354 |
+
"eval_loss": 0.5606628656387329,
|
355 |
+
"eval_runtime": 18.0204,
|
356 |
+
"eval_samples_per_second": 5.549,
|
357 |
+
"eval_wer": 0.6427406199021207,
|
358 |
+
"step": 2500
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 2.13,
|
362 |
+
"learning_rate": 9.931787175989086e-05,
|
363 |
+
"loss": 0.472,
|
364 |
+
"step": 2600
|
365 |
+
},
|
366 |
+
{
|
367 |
+
"epoch": 2.13,
|
368 |
+
"eval_loss": 0.5908198952674866,
|
369 |
+
"eval_runtime": 18.0255,
|
370 |
+
"eval_samples_per_second": 5.548,
|
371 |
+
"eval_wer": 0.6378466557911908,
|
372 |
+
"step": 2600
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 2.21,
|
376 |
+
"learning_rate": 9.92905866302865e-05,
|
377 |
+
"loss": 0.4642,
|
378 |
+
"step": 2700
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 2.21,
|
382 |
+
"eval_loss": 0.5597887635231018,
|
383 |
+
"eval_runtime": 18.0187,
|
384 |
+
"eval_samples_per_second": 5.55,
|
385 |
+
"eval_wer": 0.6443719412724307,
|
386 |
+
"step": 2700
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 2.29,
|
390 |
+
"learning_rate": 9.926330150068214e-05,
|
391 |
+
"loss": 0.4331,
|
392 |
+
"step": 2800
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 2.29,
|
396 |
+
"eval_loss": 0.5704432725906372,
|
397 |
+
"eval_runtime": 18.1172,
|
398 |
+
"eval_samples_per_second": 5.52,
|
399 |
+
"eval_wer": 0.6264274061990212,
|
400 |
+
"step": 2800
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 2.37,
|
404 |
+
"learning_rate": 9.923601637107776e-05,
|
405 |
+
"loss": 0.4562,
|
406 |
+
"step": 2900
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 2.37,
|
410 |
+
"eval_loss": 0.544298529624939,
|
411 |
+
"eval_runtime": 17.9762,
|
412 |
+
"eval_samples_per_second": 5.563,
|
413 |
+
"eval_wer": 0.6199021207177814,
|
414 |
+
"step": 2900
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 2.45,
|
418 |
+
"learning_rate": 9.920873124147341e-05,
|
419 |
+
"loss": 0.4491,
|
420 |
+
"step": 3000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 2.45,
|
424 |
+
"eval_loss": 0.5418174266815186,
|
425 |
+
"eval_runtime": 17.9099,
|
426 |
+
"eval_samples_per_second": 5.584,
|
427 |
+
"eval_wer": 0.6296900489396411,
|
428 |
+
"step": 3000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 2.54,
|
432 |
+
"learning_rate": 9.918144611186904e-05,
|
433 |
+
"loss": 0.4278,
|
434 |
+
"step": 3100
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.54,
|
438 |
+
"eval_loss": 0.5669529438018799,
|
439 |
+
"eval_runtime": 18.1431,
|
440 |
+
"eval_samples_per_second": 5.512,
|
441 |
+
"eval_wer": 0.6280587275693311,
|
442 |
+
"step": 3100
|
443 |
+
},
|
444 |
+
{
|
445 |
+
"epoch": 2.62,
|
446 |
+
"learning_rate": 9.915416098226466e-05,
|
447 |
+
"loss": 0.4267,
|
448 |
+
"step": 3200
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 2.62,
|
452 |
+
"eval_loss": 0.5152673721313477,
|
453 |
+
"eval_runtime": 17.9853,
|
454 |
+
"eval_samples_per_second": 5.56,
|
455 |
+
"eval_wer": 0.6182707993474714,
|
456 |
+
"step": 3200
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 2.7,
|
460 |
+
"learning_rate": 9.912687585266031e-05,
|
461 |
+
"loss": 0.4056,
|
462 |
+
"step": 3300
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.7,
|
466 |
+
"eval_loss": 0.5021263360977173,
|
467 |
+
"eval_runtime": 17.9672,
|
468 |
+
"eval_samples_per_second": 5.566,
|
469 |
+
"eval_wer": 0.598694942903752,
|
470 |
+
"step": 3300
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 2.78,
|
474 |
+
"learning_rate": 9.909959072305594e-05,
|
475 |
+
"loss": 0.4121,
|
476 |
+
"step": 3400
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 2.78,
|
480 |
+
"eval_loss": 0.5312536358833313,
|
481 |
+
"eval_runtime": 17.8604,
|
482 |
+
"eval_samples_per_second": 5.599,
|
483 |
+
"eval_wer": 0.600326264274062,
|
484 |
+
"step": 3400
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 2.86,
|
488 |
+
"learning_rate": 9.907230559345157e-05,
|
489 |
+
"loss": 0.4357,
|
490 |
+
"step": 3500
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 2.86,
|
494 |
+
"eval_loss": 0.5214389562606812,
|
495 |
+
"eval_runtime": 18.0662,
|
496 |
+
"eval_samples_per_second": 5.535,
|
497 |
+
"eval_wer": 0.5954323001631321,
|
498 |
+
"step": 3500
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 2.95,
|
502 |
+
"learning_rate": 9.904502046384721e-05,
|
503 |
+
"loss": 0.4174,
|
504 |
+
"step": 3600
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 2.95,
|
508 |
+
"eval_loss": 0.47309502959251404,
|
509 |
+
"eval_runtime": 18.0379,
|
510 |
+
"eval_samples_per_second": 5.544,
|
511 |
+
"eval_wer": 0.5628058727569332,
|
512 |
+
"step": 3600
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 3.03,
|
516 |
+
"learning_rate": 9.901773533424284e-05,
|
517 |
+
"loss": 0.3911,
|
518 |
+
"step": 3700
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 3.03,
|
522 |
+
"eval_loss": 0.46288207173347473,
|
523 |
+
"eval_runtime": 17.9748,
|
524 |
+
"eval_samples_per_second": 5.563,
|
525 |
+
"eval_wer": 0.566068515497553,
|
526 |
+
"step": 3700
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 3.11,
|
530 |
+
"learning_rate": 9.899045020463847e-05,
|
531 |
+
"loss": 0.3817,
|
532 |
+
"step": 3800
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 3.11,
|
536 |
+
"eval_loss": 0.5229946374893188,
|
537 |
+
"eval_runtime": 18.101,
|
538 |
+
"eval_samples_per_second": 5.525,
|
539 |
+
"eval_wer": 0.5840130505709625,
|
540 |
+
"step": 3800
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 3.19,
|
544 |
+
"learning_rate": 9.896316507503411e-05,
|
545 |
+
"loss": 0.3504,
|
546 |
+
"step": 3900
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 3.19,
|
550 |
+
"eval_loss": 0.459197074174881,
|
551 |
+
"eval_runtime": 17.9054,
|
552 |
+
"eval_samples_per_second": 5.585,
|
553 |
+
"eval_wer": 0.5448613376835236,
|
554 |
+
"step": 3900
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 3.27,
|
558 |
+
"learning_rate": 9.893587994542974e-05,
|
559 |
+
"loss": 0.3741,
|
560 |
+
"step": 4000
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 3.27,
|
564 |
+
"eval_loss": 0.4944967031478882,
|
565 |
+
"eval_runtime": 18.5172,
|
566 |
+
"eval_samples_per_second": 5.4,
|
567 |
+
"eval_wer": 0.5562805872756933,
|
568 |
+
"step": 4000
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 3.36,
|
572 |
+
"learning_rate": 9.890859481582539e-05,
|
573 |
+
"loss": 0.3451,
|
574 |
+
"step": 4100
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 3.36,
|
578 |
+
"eval_loss": 0.49022501707077026,
|
579 |
+
"eval_runtime": 18.2792,
|
580 |
+
"eval_samples_per_second": 5.471,
|
581 |
+
"eval_wer": 0.5562805872756933,
|
582 |
+
"step": 4100
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 3.44,
|
586 |
+
"learning_rate": 9.888130968622101e-05,
|
587 |
+
"loss": 0.356,
|
588 |
+
"step": 4200
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 3.44,
|
592 |
+
"eval_loss": 0.4645284414291382,
|
593 |
+
"eval_runtime": 18.1691,
|
594 |
+
"eval_samples_per_second": 5.504,
|
595 |
+
"eval_wer": 0.5513866231647635,
|
596 |
+
"step": 4200
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 3.52,
|
600 |
+
"learning_rate": 9.885402455661665e-05,
|
601 |
+
"loss": 0.3679,
|
602 |
+
"step": 4300
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 3.52,
|
606 |
+
"eval_loss": 0.4742129147052765,
|
607 |
+
"eval_runtime": 18.0285,
|
608 |
+
"eval_samples_per_second": 5.547,
|
609 |
+
"eval_wer": 0.5791190864600326,
|
610 |
+
"step": 4300
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 3.6,
|
614 |
+
"learning_rate": 9.882673942701229e-05,
|
615 |
+
"loss": 0.343,
|
616 |
+
"step": 4400
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 3.6,
|
620 |
+
"eval_loss": 0.45692238211631775,
|
621 |
+
"eval_runtime": 18.0518,
|
622 |
+
"eval_samples_per_second": 5.54,
|
623 |
+
"eval_wer": 0.566068515497553,
|
624 |
+
"step": 4400
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 3.68,
|
628 |
+
"learning_rate": 9.879945429740792e-05,
|
629 |
+
"loss": 0.39,
|
630 |
+
"step": 4500
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 3.68,
|
634 |
+
"eval_loss": 0.4621705114841461,
|
635 |
+
"eval_runtime": 18.0928,
|
636 |
+
"eval_samples_per_second": 5.527,
|
637 |
+
"eval_wer": 0.5562805872756933,
|
638 |
+
"step": 4500
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 3.76,
|
642 |
+
"learning_rate": 9.877216916780355e-05,
|
643 |
+
"loss": 0.3434,
|
644 |
+
"step": 4600
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 3.76,
|
648 |
+
"eval_loss": 0.5067647695541382,
|
649 |
+
"eval_runtime": 18.1879,
|
650 |
+
"eval_samples_per_second": 5.498,
|
651 |
+
"eval_wer": 0.5693311582381729,
|
652 |
+
"step": 4600
|
653 |
+
},
|
654 |
+
{
|
655 |
+
"epoch": 3.85,
|
656 |
+
"learning_rate": 9.874488403819919e-05,
|
657 |
+
"loss": 0.3402,
|
658 |
+
"step": 4700
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 3.85,
|
662 |
+
"eval_loss": 0.4918912947177887,
|
663 |
+
"eval_runtime": 18.1428,
|
664 |
+
"eval_samples_per_second": 5.512,
|
665 |
+
"eval_wer": 0.5693311582381729,
|
666 |
+
"step": 4700
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 3.93,
|
670 |
+
"learning_rate": 9.871759890859482e-05,
|
671 |
+
"loss": 0.3493,
|
672 |
+
"step": 4800
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 3.93,
|
676 |
+
"eval_loss": 0.4792294204235077,
|
677 |
+
"eval_runtime": 18.131,
|
678 |
+
"eval_samples_per_second": 5.515,
|
679 |
+
"eval_wer": 0.5709624796084829,
|
680 |
+
"step": 4800
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 4.01,
|
684 |
+
"learning_rate": 9.869031377899045e-05,
|
685 |
+
"loss": 0.3367,
|
686 |
+
"step": 4900
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 4.01,
|
690 |
+
"eval_loss": 0.5306392908096313,
|
691 |
+
"eval_runtime": 18.1096,
|
692 |
+
"eval_samples_per_second": 5.522,
|
693 |
+
"eval_wer": 0.5774877650897227,
|
694 |
+
"step": 4900
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 4.09,
|
698 |
+
"learning_rate": 9.866302864938609e-05,
|
699 |
+
"loss": 0.3163,
|
700 |
+
"step": 5000
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 4.09,
|
704 |
+
"eval_loss": 0.49867820739746094,
|
705 |
+
"eval_runtime": 18.1564,
|
706 |
+
"eval_samples_per_second": 5.508,
|
707 |
+
"eval_wer": 0.5709624796084829,
|
708 |
+
"step": 5000
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 4.17,
|
712 |
+
"learning_rate": 9.863574351978172e-05,
|
713 |
+
"loss": 0.3025,
|
714 |
+
"step": 5100
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 4.17,
|
718 |
+
"eval_loss": 0.49608245491981506,
|
719 |
+
"eval_runtime": 18.3468,
|
720 |
+
"eval_samples_per_second": 5.451,
|
721 |
+
"eval_wer": 0.5383360522022839,
|
722 |
+
"step": 5100
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 4.26,
|
726 |
+
"learning_rate": 9.860845839017737e-05,
|
727 |
+
"loss": 0.3017,
|
728 |
+
"step": 5200
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 4.26,
|
732 |
+
"eval_loss": 0.519436776638031,
|
733 |
+
"eval_runtime": 18.1576,
|
734 |
+
"eval_samples_per_second": 5.507,
|
735 |
+
"eval_wer": 0.5595432300163132,
|
736 |
+
"step": 5200
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 4.34,
|
740 |
+
"learning_rate": 9.858117326057299e-05,
|
741 |
+
"loss": 0.318,
|
742 |
+
"step": 5300
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 4.34,
|
746 |
+
"eval_loss": 0.49539458751678467,
|
747 |
+
"eval_runtime": 18.1275,
|
748 |
+
"eval_samples_per_second": 5.516,
|
749 |
+
"eval_wer": 0.5530179445350734,
|
750 |
+
"step": 5300
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 4.42,
|
754 |
+
"learning_rate": 9.855388813096863e-05,
|
755 |
+
"loss": 0.3297,
|
756 |
+
"step": 5400
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 4.42,
|
760 |
+
"eval_loss": 0.5326051115989685,
|
761 |
+
"eval_runtime": 18.044,
|
762 |
+
"eval_samples_per_second": 5.542,
|
763 |
+
"eval_wer": 0.5595432300163132,
|
764 |
+
"step": 5400
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 4.5,
|
768 |
+
"learning_rate": 9.852660300136425e-05,
|
769 |
+
"loss": 0.3045,
|
770 |
+
"step": 5500
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 4.5,
|
774 |
+
"eval_loss": 0.48108261823654175,
|
775 |
+
"eval_runtime": 18.0697,
|
776 |
+
"eval_samples_per_second": 5.534,
|
777 |
+
"eval_wer": 0.5415986949429038,
|
778 |
+
"step": 5500
|
779 |
+
},
|
780 |
+
{
|
781 |
+
"epoch": 4.58,
|
782 |
+
"learning_rate": 9.84993178717599e-05,
|
783 |
+
"loss": 0.2975,
|
784 |
+
"step": 5600
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 4.58,
|
788 |
+
"eval_loss": 0.5604788064956665,
|
789 |
+
"eval_runtime": 18.1396,
|
790 |
+
"eval_samples_per_second": 5.513,
|
791 |
+
"eval_wer": 0.5725938009787929,
|
792 |
+
"step": 5600
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 4.66,
|
796 |
+
"learning_rate": 9.847203274215553e-05,
|
797 |
+
"loss": 0.2851,
|
798 |
+
"step": 5700
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 4.66,
|
802 |
+
"eval_loss": 0.52396559715271,
|
803 |
+
"eval_runtime": 18.0267,
|
804 |
+
"eval_samples_per_second": 5.547,
|
805 |
+
"eval_wer": 0.5464926590538336,
|
806 |
+
"step": 5700
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 4.75,
|
810 |
+
"learning_rate": 9.844474761255117e-05,
|
811 |
+
"loss": 0.3037,
|
812 |
+
"step": 5800
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 4.75,
|
816 |
+
"eval_loss": 0.4817478060722351,
|
817 |
+
"eval_runtime": 18.1198,
|
818 |
+
"eval_samples_per_second": 5.519,
|
819 |
+
"eval_wer": 0.5415986949429038,
|
820 |
+
"step": 5800
|
821 |
+
},
|
822 |
+
{
|
823 |
+
"epoch": 4.83,
|
824 |
+
"learning_rate": 9.84174624829468e-05,
|
825 |
+
"loss": 0.3032,
|
826 |
+
"step": 5900
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 4.83,
|
830 |
+
"eval_loss": 0.5403196215629578,
|
831 |
+
"eval_runtime": 18.0777,
|
832 |
+
"eval_samples_per_second": 5.532,
|
833 |
+
"eval_wer": 0.5709624796084829,
|
834 |
+
"step": 5900
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 4.91,
|
838 |
+
"learning_rate": 9.839017735334243e-05,
|
839 |
+
"loss": 0.2947,
|
840 |
+
"step": 6000
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 4.91,
|
844 |
+
"eval_loss": 0.4690156877040863,
|
845 |
+
"eval_runtime": 18.4111,
|
846 |
+
"eval_samples_per_second": 5.432,
|
847 |
+
"eval_wer": 0.5579119086460033,
|
848 |
+
"step": 6000
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 4.99,
|
852 |
+
"learning_rate": 9.836289222373807e-05,
|
853 |
+
"loss": 0.3041,
|
854 |
+
"step": 6100
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 4.99,
|
858 |
+
"eval_loss": 0.5112998485565186,
|
859 |
+
"eval_runtime": 18.0574,
|
860 |
+
"eval_samples_per_second": 5.538,
|
861 |
+
"eval_wer": 0.5595432300163132,
|
862 |
+
"step": 6100
|
863 |
+
},
|
864 |
+
{
|
865 |
+
"epoch": 5.07,
|
866 |
+
"learning_rate": 9.83356070941337e-05,
|
867 |
+
"loss": 0.2659,
|
868 |
+
"step": 6200
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 5.07,
|
872 |
+
"eval_loss": 0.5093588829040527,
|
873 |
+
"eval_runtime": 18.1339,
|
874 |
+
"eval_samples_per_second": 5.515,
|
875 |
+
"eval_wer": 0.5448613376835236,
|
876 |
+
"step": 6200
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 5.16,
|
880 |
+
"learning_rate": 9.830832196452933e-05,
|
881 |
+
"loss": 0.2621,
|
882 |
+
"step": 6300
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 5.16,
|
886 |
+
"eval_loss": 0.5168948173522949,
|
887 |
+
"eval_runtime": 18.1269,
|
888 |
+
"eval_samples_per_second": 5.517,
|
889 |
+
"eval_wer": 0.5758564437194127,
|
890 |
+
"step": 6300
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 5.24,
|
894 |
+
"learning_rate": 9.828103683492497e-05,
|
895 |
+
"loss": 0.264,
|
896 |
+
"step": 6400
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 5.24,
|
900 |
+
"eval_loss": 0.5183637142181396,
|
901 |
+
"eval_runtime": 18.0896,
|
902 |
+
"eval_samples_per_second": 5.528,
|
903 |
+
"eval_wer": 0.5399673735725938,
|
904 |
+
"step": 6400
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 5.32,
|
908 |
+
"learning_rate": 9.825375170532061e-05,
|
909 |
+
"loss": 0.2725,
|
910 |
+
"step": 6500
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 5.32,
|
914 |
+
"eval_loss": 0.5389344096183777,
|
915 |
+
"eval_runtime": 18.0714,
|
916 |
+
"eval_samples_per_second": 5.534,
|
917 |
+
"eval_wer": 0.5415986949429038,
|
918 |
+
"step": 6500
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 5.4,
|
922 |
+
"learning_rate": 9.822646657571623e-05,
|
923 |
+
"loss": 0.273,
|
924 |
+
"step": 6600
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 5.4,
|
928 |
+
"eval_loss": 0.5070094466209412,
|
929 |
+
"eval_runtime": 18.2125,
|
930 |
+
"eval_samples_per_second": 5.491,
|
931 |
+
"eval_wer": 0.5595432300163132,
|
932 |
+
"step": 6600
|
933 |
+
},
|
934 |
+
{
|
935 |
+
"epoch": 5.48,
|
936 |
+
"learning_rate": 9.819918144611188e-05,
|
937 |
+
"loss": 0.2889,
|
938 |
+
"step": 6700
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 5.48,
|
942 |
+
"eval_loss": 0.5503917932510376,
|
943 |
+
"eval_runtime": 18.0283,
|
944 |
+
"eval_samples_per_second": 5.547,
|
945 |
+
"eval_wer": 0.5464926590538336,
|
946 |
+
"step": 6700
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 5.56,
|
950 |
+
"learning_rate": 9.817189631650751e-05,
|
951 |
+
"loss": 0.2728,
|
952 |
+
"step": 6800
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 5.56,
|
956 |
+
"eval_loss": 0.6155907511711121,
|
957 |
+
"eval_runtime": 18.1183,
|
958 |
+
"eval_samples_per_second": 5.519,
|
959 |
+
"eval_wer": 0.5889070146818923,
|
960 |
+
"step": 6800
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 5.65,
|
964 |
+
"learning_rate": 9.814461118690314e-05,
|
965 |
+
"loss": 0.2914,
|
966 |
+
"step": 6900
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 5.65,
|
970 |
+
"eval_loss": 0.521608293056488,
|
971 |
+
"eval_runtime": 18.3112,
|
972 |
+
"eval_samples_per_second": 5.461,
|
973 |
+
"eval_wer": 0.5383360522022839,
|
974 |
+
"step": 6900
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 5.73,
|
978 |
+
"learning_rate": 9.811732605729878e-05,
|
979 |
+
"loss": 0.2639,
|
980 |
+
"step": 7000
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 5.73,
|
984 |
+
"eval_loss": 0.6134706139564514,
|
985 |
+
"eval_runtime": 18.0647,
|
986 |
+
"eval_samples_per_second": 5.536,
|
987 |
+
"eval_wer": 0.5481239804241436,
|
988 |
+
"step": 7000
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 5.81,
|
992 |
+
"learning_rate": 9.809004092769441e-05,
|
993 |
+
"loss": 0.2688,
|
994 |
+
"step": 7100
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 5.81,
|
998 |
+
"eval_loss": 0.5232261419296265,
|
999 |
+
"eval_runtime": 18.1754,
|
1000 |
+
"eval_samples_per_second": 5.502,
|
1001 |
+
"eval_wer": 0.5464926590538336,
|
1002 |
+
"step": 7100
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 5.89,
|
1006 |
+
"learning_rate": 9.806275579809004e-05,
|
1007 |
+
"loss": 0.2652,
|
1008 |
+
"step": 7200
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 5.89,
|
1012 |
+
"eval_loss": 0.5219370126724243,
|
1013 |
+
"eval_runtime": 18.2531,
|
1014 |
+
"eval_samples_per_second": 5.479,
|
1015 |
+
"eval_wer": 0.533442088091354,
|
1016 |
+
"step": 7200
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 5.97,
|
1020 |
+
"learning_rate": 9.803547066848569e-05,
|
1021 |
+
"loss": 0.2722,
|
1022 |
+
"step": 7300
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 5.97,
|
1026 |
+
"eval_loss": 0.5088679790496826,
|
1027 |
+
"eval_runtime": 18.2641,
|
1028 |
+
"eval_samples_per_second": 5.475,
|
1029 |
+
"eval_wer": 0.5236541598694943,
|
1030 |
+
"step": 7300
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 6.06,
|
1034 |
+
"learning_rate": 9.800818553888131e-05,
|
1035 |
+
"loss": 0.2575,
|
1036 |
+
"step": 7400
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 6.06,
|
1040 |
+
"eval_loss": 0.509792149066925,
|
1041 |
+
"eval_runtime": 18.1448,
|
1042 |
+
"eval_samples_per_second": 5.511,
|
1043 |
+
"eval_wer": 0.5220228384991843,
|
1044 |
+
"step": 7400
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 6.14,
|
1048 |
+
"learning_rate": 9.798090040927694e-05,
|
1049 |
+
"loss": 0.2397,
|
1050 |
+
"step": 7500
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 6.14,
|
1054 |
+
"eval_loss": 0.5148301124572754,
|
1055 |
+
"eval_runtime": 18.0794,
|
1056 |
+
"eval_samples_per_second": 5.531,
|
1057 |
+
"eval_wer": 0.5579119086460033,
|
1058 |
+
"step": 7500
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 6.22,
|
1062 |
+
"learning_rate": 9.795361527967258e-05,
|
1063 |
+
"loss": 0.2464,
|
1064 |
+
"step": 7600
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 6.22,
|
1068 |
+
"eval_loss": 0.5043299198150635,
|
1069 |
+
"eval_runtime": 18.0544,
|
1070 |
+
"eval_samples_per_second": 5.539,
|
1071 |
+
"eval_wer": 0.5513866231647635,
|
1072 |
+
"step": 7600
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 6.3,
|
1076 |
+
"learning_rate": 9.792633015006821e-05,
|
1077 |
+
"loss": 0.2485,
|
1078 |
+
"step": 7700
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 6.3,
|
1082 |
+
"eval_loss": 0.5194571018218994,
|
1083 |
+
"eval_runtime": 18.0679,
|
1084 |
+
"eval_samples_per_second": 5.535,
|
1085 |
+
"eval_wer": 0.5823817292006526,
|
1086 |
+
"step": 7700
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 6.38,
|
1090 |
+
"learning_rate": 9.789904502046386e-05,
|
1091 |
+
"loss": 0.2299,
|
1092 |
+
"step": 7800
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 6.38,
|
1096 |
+
"eval_loss": 0.5433354377746582,
|
1097 |
+
"eval_runtime": 18.0004,
|
1098 |
+
"eval_samples_per_second": 5.555,
|
1099 |
+
"eval_wer": 0.5399673735725938,
|
1100 |
+
"step": 7800
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 6.46,
|
1104 |
+
"learning_rate": 9.787175989085948e-05,
|
1105 |
+
"loss": 0.244,
|
1106 |
+
"step": 7900
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 6.46,
|
1110 |
+
"eval_loss": 0.5765259861946106,
|
1111 |
+
"eval_runtime": 18.2353,
|
1112 |
+
"eval_samples_per_second": 5.484,
|
1113 |
+
"eval_wer": 0.531810766721044,
|
1114 |
+
"step": 7900
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 6.55,
|
1118 |
+
"learning_rate": 9.784447476125512e-05,
|
1119 |
+
"loss": 0.2394,
|
1120 |
+
"step": 8000
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 6.55,
|
1124 |
+
"eval_loss": 0.6303309202194214,
|
1125 |
+
"eval_runtime": 18.0232,
|
1126 |
+
"eval_samples_per_second": 5.548,
|
1127 |
+
"eval_wer": 0.5530179445350734,
|
1128 |
+
"step": 8000
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 6.63,
|
1132 |
+
"learning_rate": 9.781718963165076e-05,
|
1133 |
+
"loss": 0.2423,
|
1134 |
+
"step": 8100
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 6.63,
|
1138 |
+
"eval_loss": 0.5469087362289429,
|
1139 |
+
"eval_runtime": 18.1188,
|
1140 |
+
"eval_samples_per_second": 5.519,
|
1141 |
+
"eval_wer": 0.533442088091354,
|
1142 |
+
"step": 8100
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 6.71,
|
1146 |
+
"learning_rate": 9.778990450204639e-05,
|
1147 |
+
"loss": 0.2363,
|
1148 |
+
"step": 8200
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 6.71,
|
1152 |
+
"eval_loss": 0.5919773578643799,
|
1153 |
+
"eval_runtime": 18.0302,
|
1154 |
+
"eval_samples_per_second": 5.546,
|
1155 |
+
"eval_wer": 0.567699836867863,
|
1156 |
+
"step": 8200
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 6.79,
|
1160 |
+
"learning_rate": 9.776261937244202e-05,
|
1161 |
+
"loss": 0.231,
|
1162 |
+
"step": 8300
|
1163 |
+
},
|
1164 |
+
{
|
1165 |
+
"epoch": 6.79,
|
1166 |
+
"eval_loss": 0.49973785877227783,
|
1167 |
+
"eval_runtime": 18.1196,
|
1168 |
+
"eval_samples_per_second": 5.519,
|
1169 |
+
"eval_wer": 0.5464926590538336,
|
1170 |
+
"step": 8300
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 6.87,
|
1174 |
+
"learning_rate": 9.773533424283766e-05,
|
1175 |
+
"loss": 0.244,
|
1176 |
+
"step": 8400
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 6.87,
|
1180 |
+
"eval_loss": 0.5464155673980713,
|
1181 |
+
"eval_runtime": 18.0797,
|
1182 |
+
"eval_samples_per_second": 5.531,
|
1183 |
+
"eval_wer": 0.5513866231647635,
|
1184 |
+
"step": 8400
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 6.96,
|
1188 |
+
"learning_rate": 9.770804911323329e-05,
|
1189 |
+
"loss": 0.2332,
|
1190 |
+
"step": 8500
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 6.96,
|
1194 |
+
"eval_loss": 0.526359498500824,
|
1195 |
+
"eval_runtime": 18.1851,
|
1196 |
+
"eval_samples_per_second": 5.499,
|
1197 |
+
"eval_wer": 0.5709624796084829,
|
1198 |
+
"step": 8500
|
1199 |
+
},
|
1200 |
+
{
|
1201 |
+
"epoch": 7.04,
|
1202 |
+
"learning_rate": 9.768076398362894e-05,
|
1203 |
+
"loss": 0.2421,
|
1204 |
+
"step": 8600
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 7.04,
|
1208 |
+
"eval_loss": 0.5105239748954773,
|
1209 |
+
"eval_runtime": 18.1953,
|
1210 |
+
"eval_samples_per_second": 5.496,
|
1211 |
+
"eval_wer": 0.5562805872756933,
|
1212 |
+
"step": 8600
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 7.12,
|
1216 |
+
"learning_rate": 9.765347885402456e-05,
|
1217 |
+
"loss": 0.2236,
|
1218 |
+
"step": 8700
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 7.12,
|
1222 |
+
"eval_loss": 0.5103984475135803,
|
1223 |
+
"eval_runtime": 18.0762,
|
1224 |
+
"eval_samples_per_second": 5.532,
|
1225 |
+
"eval_wer": 0.5367047308319739,
|
1226 |
+
"step": 8700
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 7.2,
|
1230 |
+
"learning_rate": 9.76261937244202e-05,
|
1231 |
+
"loss": 0.2075,
|
1232 |
+
"step": 8800
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 7.2,
|
1236 |
+
"eval_loss": 0.5242435336112976,
|
1237 |
+
"eval_runtime": 18.142,
|
1238 |
+
"eval_samples_per_second": 5.512,
|
1239 |
+
"eval_wer": 0.5106035889070146,
|
1240 |
+
"step": 8800
|
1241 |
+
},
|
1242 |
+
{
|
1243 |
+
"epoch": 7.28,
|
1244 |
+
"learning_rate": 9.759890859481584e-05,
|
1245 |
+
"loss": 0.2322,
|
1246 |
+
"step": 8900
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 7.28,
|
1250 |
+
"eval_loss": 0.5415535569190979,
|
1251 |
+
"eval_runtime": 18.1455,
|
1252 |
+
"eval_samples_per_second": 5.511,
|
1253 |
+
"eval_wer": 0.5350734094616639,
|
1254 |
+
"step": 8900
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 7.36,
|
1258 |
+
"learning_rate": 9.757162346521146e-05,
|
1259 |
+
"loss": 0.209,
|
1260 |
+
"step": 9000
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 7.36,
|
1264 |
+
"eval_loss": 0.45015835762023926,
|
1265 |
+
"eval_runtime": 18.2899,
|
1266 |
+
"eval_samples_per_second": 5.467,
|
1267 |
+
"eval_wer": 0.5154975530179445,
|
1268 |
+
"step": 9000
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 7.45,
|
1272 |
+
"learning_rate": 9.75443383356071e-05,
|
1273 |
+
"loss": 0.2161,
|
1274 |
+
"step": 9100
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 7.45,
|
1278 |
+
"eval_loss": 0.435876727104187,
|
1279 |
+
"eval_runtime": 18.1085,
|
1280 |
+
"eval_samples_per_second": 5.522,
|
1281 |
+
"eval_wer": 0.5203915171288744,
|
1282 |
+
"step": 9100
|
1283 |
+
},
|
1284 |
+
{
|
1285 |
+
"epoch": 7.53,
|
1286 |
+
"learning_rate": 9.751705320600272e-05,
|
1287 |
+
"loss": 0.2239,
|
1288 |
+
"step": 9200
|
1289 |
+
},
|
1290 |
+
{
|
1291 |
+
"epoch": 7.53,
|
1292 |
+
"eval_loss": 0.5517589449882507,
|
1293 |
+
"eval_runtime": 18.1314,
|
1294 |
+
"eval_samples_per_second": 5.515,
|
1295 |
+
"eval_wer": 0.5513866231647635,
|
1296 |
+
"step": 9200
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 7.61,
|
1300 |
+
"learning_rate": 9.748976807639837e-05,
|
1301 |
+
"loss": 0.2135,
|
1302 |
+
"step": 9300
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 7.61,
|
1306 |
+
"eval_loss": 0.5722827315330505,
|
1307 |
+
"eval_runtime": 18.2661,
|
1308 |
+
"eval_samples_per_second": 5.475,
|
1309 |
+
"eval_wer": 0.5383360522022839,
|
1310 |
+
"step": 9300
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 7.69,
|
1314 |
+
"learning_rate": 9.7462482946794e-05,
|
1315 |
+
"loss": 0.2193,
|
1316 |
+
"step": 9400
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 7.69,
|
1320 |
+
"eval_loss": 0.5303468704223633,
|
1321 |
+
"eval_runtime": 18.1001,
|
1322 |
+
"eval_samples_per_second": 5.525,
|
1323 |
+
"eval_wer": 0.5073409461663948,
|
1324 |
+
"step": 9400
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 7.77,
|
1328 |
+
"learning_rate": 9.743519781718964e-05,
|
1329 |
+
"loss": 0.239,
|
1330 |
+
"step": 9500
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 7.77,
|
1334 |
+
"eval_loss": 0.5649023056030273,
|
1335 |
+
"eval_runtime": 18.115,
|
1336 |
+
"eval_samples_per_second": 5.52,
|
1337 |
+
"eval_wer": 0.5089722675367048,
|
1338 |
+
"step": 9500
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 7.86,
|
1342 |
+
"learning_rate": 9.740791268758527e-05,
|
1343 |
+
"loss": 0.2266,
|
1344 |
+
"step": 9600
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 7.86,
|
1348 |
+
"eval_loss": 0.4942982792854309,
|
1349 |
+
"eval_runtime": 18.1724,
|
1350 |
+
"eval_samples_per_second": 5.503,
|
1351 |
+
"eval_wer": 0.5187601957585645,
|
1352 |
+
"step": 9600
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 7.94,
|
1356 |
+
"learning_rate": 9.73806275579809e-05,
|
1357 |
+
"loss": 0.2328,
|
1358 |
+
"step": 9700
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 7.94,
|
1362 |
+
"eval_loss": 0.4808942377567291,
|
1363 |
+
"eval_runtime": 18.123,
|
1364 |
+
"eval_samples_per_second": 5.518,
|
1365 |
+
"eval_wer": 0.5171288743882545,
|
1366 |
+
"step": 9700
|
1367 |
+
},
|
1368 |
+
{
|
1369 |
+
"epoch": 8.02,
|
1370 |
+
"learning_rate": 9.735334242837654e-05,
|
1371 |
+
"loss": 0.2219,
|
1372 |
+
"step": 9800
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 8.02,
|
1376 |
+
"eval_loss": 0.5041812062263489,
|
1377 |
+
"eval_runtime": 18.2686,
|
1378 |
+
"eval_samples_per_second": 5.474,
|
1379 |
+
"eval_wer": 0.5220228384991843,
|
1380 |
+
"step": 9800
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 8.1,
|
1384 |
+
"learning_rate": 9.732605729877218e-05,
|
1385 |
+
"loss": 0.2166,
|
1386 |
+
"step": 9900
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 8.1,
|
1390 |
+
"eval_loss": 0.49035871028900146,
|
1391 |
+
"eval_runtime": 18.2214,
|
1392 |
+
"eval_samples_per_second": 5.488,
|
1393 |
+
"eval_wer": 0.533442088091354,
|
1394 |
+
"step": 9900
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 8.18,
|
1398 |
+
"learning_rate": 9.72987721691678e-05,
|
1399 |
+
"loss": 0.2036,
|
1400 |
+
"step": 10000
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 8.18,
|
1404 |
+
"eval_loss": 0.4532802104949951,
|
1405 |
+
"eval_runtime": 18.0465,
|
1406 |
+
"eval_samples_per_second": 5.541,
|
1407 |
+
"eval_wer": 0.5154975530179445,
|
1408 |
+
"step": 10000
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 8.27,
|
1412 |
+
"learning_rate": 9.727148703956345e-05,
|
1413 |
+
"loss": 0.2056,
|
1414 |
+
"step": 10100
|
1415 |
+
},
|
1416 |
+
{
|
1417 |
+
"epoch": 8.27,
|
1418 |
+
"eval_loss": 0.49766284227371216,
|
1419 |
+
"eval_runtime": 18.3478,
|
1420 |
+
"eval_samples_per_second": 5.45,
|
1421 |
+
"eval_wer": 0.5220228384991843,
|
1422 |
+
"step": 10100
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 8.35,
|
1426 |
+
"learning_rate": 9.724420190995908e-05,
|
1427 |
+
"loss": 0.1897,
|
1428 |
+
"step": 10200
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 8.35,
|
1432 |
+
"eval_loss": 0.5390677452087402,
|
1433 |
+
"eval_runtime": 18.1818,
|
1434 |
+
"eval_samples_per_second": 5.5,
|
1435 |
+
"eval_wer": 0.5269168026101142,
|
1436 |
+
"step": 10200
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 8.43,
|
1440 |
+
"learning_rate": 9.721691678035471e-05,
|
1441 |
+
"loss": 0.1957,
|
1442 |
+
"step": 10300
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 8.43,
|
1446 |
+
"eval_loss": 0.5764414072036743,
|
1447 |
+
"eval_runtime": 18.1005,
|
1448 |
+
"eval_samples_per_second": 5.525,
|
1449 |
+
"eval_wer": 0.533442088091354,
|
1450 |
+
"step": 10300
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 8.51,
|
1454 |
+
"learning_rate": 9.718963165075035e-05,
|
1455 |
+
"loss": 0.1985,
|
1456 |
+
"step": 10400
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 8.51,
|
1460 |
+
"eval_loss": 0.5892696976661682,
|
1461 |
+
"eval_runtime": 18.12,
|
1462 |
+
"eval_samples_per_second": 5.519,
|
1463 |
+
"eval_wer": 0.5399673735725938,
|
1464 |
+
"step": 10400
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 8.59,
|
1468 |
+
"learning_rate": 9.716234652114598e-05,
|
1469 |
+
"loss": 0.208,
|
1470 |
+
"step": 10500
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 8.59,
|
1474 |
+
"eval_loss": 0.46923932433128357,
|
1475 |
+
"eval_runtime": 18.0696,
|
1476 |
+
"eval_samples_per_second": 5.534,
|
1477 |
+
"eval_wer": 0.5122349102773246,
|
1478 |
+
"step": 10500
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 8.67,
|
1482 |
+
"learning_rate": 9.713506139154161e-05,
|
1483 |
+
"loss": 0.2055,
|
1484 |
+
"step": 10600
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 8.67,
|
1488 |
+
"eval_loss": 0.5385192036628723,
|
1489 |
+
"eval_runtime": 18.086,
|
1490 |
+
"eval_samples_per_second": 5.529,
|
1491 |
+
"eval_wer": 0.5220228384991843,
|
1492 |
+
"step": 10600
|
1493 |
+
},
|
1494 |
+
{
|
1495 |
+
"epoch": 8.76,
|
1496 |
+
"learning_rate": 9.710777626193725e-05,
|
1497 |
+
"loss": 0.205,
|
1498 |
+
"step": 10700
|
1499 |
+
},
|
1500 |
+
{
|
1501 |
+
"epoch": 8.76,
|
1502 |
+
"eval_loss": 0.5977317690849304,
|
1503 |
+
"eval_runtime": 18.0762,
|
1504 |
+
"eval_samples_per_second": 5.532,
|
1505 |
+
"eval_wer": 0.5252854812398042,
|
1506 |
+
"step": 10700
|
1507 |
+
},
|
1508 |
+
{
|
1509 |
+
"epoch": 8.84,
|
1510 |
+
"learning_rate": 9.708049113233288e-05,
|
1511 |
+
"loss": 0.2153,
|
1512 |
+
"step": 10800
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 8.84,
|
1516 |
+
"eval_loss": 0.5502565503120422,
|
1517 |
+
"eval_runtime": 18.0402,
|
1518 |
+
"eval_samples_per_second": 5.543,
|
1519 |
+
"eval_wer": 0.5350734094616639,
|
1520 |
+
"step": 10800
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 8.92,
|
1524 |
+
"learning_rate": 9.705320600272851e-05,
|
1525 |
+
"loss": 0.2221,
|
1526 |
+
"step": 10900
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 8.92,
|
1530 |
+
"eval_loss": 0.5062658786773682,
|
1531 |
+
"eval_runtime": 18.1668,
|
1532 |
+
"eval_samples_per_second": 5.505,
|
1533 |
+
"eval_wer": 0.5057096247960848,
|
1534 |
+
"step": 10900
|
1535 |
+
},
|
1536 |
+
{
|
1537 |
+
"epoch": 9.0,
|
1538 |
+
"learning_rate": 9.702592087312416e-05,
|
1539 |
+
"loss": 0.2228,
|
1540 |
+
"step": 11000
|
1541 |
+
},
|
1542 |
+
{
|
1543 |
+
"epoch": 9.0,
|
1544 |
+
"eval_loss": 0.4952687919139862,
|
1545 |
+
"eval_runtime": 18.0738,
|
1546 |
+
"eval_samples_per_second": 5.533,
|
1547 |
+
"eval_wer": 0.5187601957585645,
|
1548 |
+
"step": 11000
|
1549 |
+
},
|
1550 |
+
{
|
1551 |
+
"epoch": 9.08,
|
1552 |
+
"learning_rate": 9.699863574351978e-05,
|
1553 |
+
"loss": 0.1888,
|
1554 |
+
"step": 11100
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 9.08,
|
1558 |
+
"eval_loss": 0.5036348700523376,
|
1559 |
+
"eval_runtime": 18.1191,
|
1560 |
+
"eval_samples_per_second": 5.519,
|
1561 |
+
"eval_wer": 0.5089722675367048,
|
1562 |
+
"step": 11100
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 9.17,
|
1566 |
+
"learning_rate": 9.697135061391543e-05,
|
1567 |
+
"loss": 0.1827,
|
1568 |
+
"step": 11200
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 9.17,
|
1572 |
+
"eval_loss": 0.5076143741607666,
|
1573 |
+
"eval_runtime": 18.3508,
|
1574 |
+
"eval_samples_per_second": 5.449,
|
1575 |
+
"eval_wer": 0.5008156606851549,
|
1576 |
+
"step": 11200
|
1577 |
+
},
|
1578 |
+
{
|
1579 |
+
"epoch": 9.25,
|
1580 |
+
"learning_rate": 9.694406548431105e-05,
|
1581 |
+
"loss": 0.1881,
|
1582 |
+
"step": 11300
|
1583 |
+
},
|
1584 |
+
{
|
1585 |
+
"epoch": 9.25,
|
1586 |
+
"eval_loss": 0.48387715220451355,
|
1587 |
+
"eval_runtime": 18.1083,
|
1588 |
+
"eval_samples_per_second": 5.522,
|
1589 |
+
"eval_wer": 0.49592169657422513,
|
1590 |
+
"step": 11300
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 9.33,
|
1594 |
+
"learning_rate": 9.69167803547067e-05,
|
1595 |
+
"loss": 0.1889,
|
1596 |
+
"step": 11400
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 9.33,
|
1600 |
+
"eval_loss": 0.5388205051422119,
|
1601 |
+
"eval_runtime": 18.0209,
|
1602 |
+
"eval_samples_per_second": 5.549,
|
1603 |
+
"eval_wer": 0.5008156606851549,
|
1604 |
+
"step": 11400
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 9.41,
|
1608 |
+
"learning_rate": 9.688949522510233e-05,
|
1609 |
+
"loss": 0.1845,
|
1610 |
+
"step": 11500
|
1611 |
+
},
|
1612 |
+
{
|
1613 |
+
"epoch": 9.41,
|
1614 |
+
"eval_loss": 0.4822709262371063,
|
1615 |
+
"eval_runtime": 18.0517,
|
1616 |
+
"eval_samples_per_second": 5.54,
|
1617 |
+
"eval_wer": 0.4893964110929853,
|
1618 |
+
"step": 11500
|
1619 |
+
},
|
1620 |
+
{
|
1621 |
+
"epoch": 9.49,
|
1622 |
+
"learning_rate": 9.686221009549796e-05,
|
1623 |
+
"loss": 0.2032,
|
1624 |
+
"step": 11600
|
1625 |
+
},
|
1626 |
+
{
|
1627 |
+
"epoch": 9.49,
|
1628 |
+
"eval_loss": 0.50101637840271,
|
1629 |
+
"eval_runtime": 18.1112,
|
1630 |
+
"eval_samples_per_second": 5.521,
|
1631 |
+
"eval_wer": 0.5350734094616639,
|
1632 |
+
"step": 11600
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 9.57,
|
1636 |
+
"learning_rate": 9.68349249658936e-05,
|
1637 |
+
"loss": 0.2148,
|
1638 |
+
"step": 11700
|
1639 |
+
},
|
1640 |
+
{
|
1641 |
+
"epoch": 9.57,
|
1642 |
+
"eval_loss": 0.49152928590774536,
|
1643 |
+
"eval_runtime": 18.1082,
|
1644 |
+
"eval_samples_per_second": 5.522,
|
1645 |
+
"eval_wer": 0.5122349102773246,
|
1646 |
+
"step": 11700
|
1647 |
+
},
|
1648 |
+
{
|
1649 |
+
"epoch": 9.66,
|
1650 |
+
"learning_rate": 9.680763983628923e-05,
|
1651 |
+
"loss": 0.2014,
|
1652 |
+
"step": 11800
|
1653 |
+
},
|
1654 |
+
{
|
1655 |
+
"epoch": 9.66,
|
1656 |
+
"eval_loss": 0.490098774433136,
|
1657 |
+
"eval_runtime": 18.2244,
|
1658 |
+
"eval_samples_per_second": 5.487,
|
1659 |
+
"eval_wer": 0.5187601957585645,
|
1660 |
+
"step": 11800
|
1661 |
+
},
|
1662 |
+
{
|
1663 |
+
"epoch": 9.74,
|
1664 |
+
"learning_rate": 9.678035470668486e-05,
|
1665 |
+
"loss": 0.2006,
|
1666 |
+
"step": 11900
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 9.74,
|
1670 |
+
"eval_loss": 0.5543618202209473,
|
1671 |
+
"eval_runtime": 18.1457,
|
1672 |
+
"eval_samples_per_second": 5.511,
|
1673 |
+
"eval_wer": 0.5122349102773246,
|
1674 |
+
"step": 11900
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 9.82,
|
1678 |
+
"learning_rate": 9.675306957708049e-05,
|
1679 |
+
"loss": 0.2015,
|
1680 |
+
"step": 12000
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 9.82,
|
1684 |
+
"eval_loss": 0.46993693709373474,
|
1685 |
+
"eval_runtime": 18.1108,
|
1686 |
+
"eval_samples_per_second": 5.522,
|
1687 |
+
"eval_wer": 0.49755301794453505,
|
1688 |
+
"step": 12000
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 9.9,
|
1692 |
+
"learning_rate": 9.672578444747613e-05,
|
1693 |
+
"loss": 0.1955,
|
1694 |
+
"step": 12100
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 9.9,
|
1698 |
+
"eval_loss": 0.5119531750679016,
|
1699 |
+
"eval_runtime": 18.1088,
|
1700 |
+
"eval_samples_per_second": 5.522,
|
1701 |
+
"eval_wer": 0.5171288743882545,
|
1702 |
+
"step": 12100
|
1703 |
+
},
|
1704 |
+
{
|
1705 |
+
"epoch": 9.98,
|
1706 |
+
"learning_rate": 9.669849931787176e-05,
|
1707 |
+
"loss": 0.1992,
|
1708 |
+
"step": 12200
|
1709 |
+
},
|
1710 |
+
{
|
1711 |
+
"epoch": 9.98,
|
1712 |
+
"eval_loss": 0.5462976098060608,
|
1713 |
+
"eval_runtime": 18.1437,
|
1714 |
+
"eval_samples_per_second": 5.512,
|
1715 |
+
"eval_wer": 0.5122349102773246,
|
1716 |
+
"step": 12200
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 10.07,
|
1720 |
+
"learning_rate": 9.66712141882674e-05,
|
1721 |
+
"loss": 0.1858,
|
1722 |
+
"step": 12300
|
1723 |
+
},
|
1724 |
+
{
|
1725 |
+
"epoch": 10.07,
|
1726 |
+
"eval_loss": 0.5365936756134033,
|
1727 |
+
"eval_runtime": 18.1212,
|
1728 |
+
"eval_samples_per_second": 5.518,
|
1729 |
+
"eval_wer": 0.5057096247960848,
|
1730 |
+
"step": 12300
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 10.15,
|
1734 |
+
"learning_rate": 9.664392905866303e-05,
|
1735 |
+
"loss": 0.1695,
|
1736 |
+
"step": 12400
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 10.15,
|
1740 |
+
"eval_loss": 0.5218992829322815,
|
1741 |
+
"eval_runtime": 18.1427,
|
1742 |
+
"eval_samples_per_second": 5.512,
|
1743 |
+
"eval_wer": 0.5138662316476346,
|
1744 |
+
"step": 12400
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 10.23,
|
1748 |
+
"learning_rate": 9.661664392905867e-05,
|
1749 |
+
"loss": 0.198,
|
1750 |
+
"step": 12500
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 10.23,
|
1754 |
+
"eval_loss": 0.5381004214286804,
|
1755 |
+
"eval_runtime": 18.0861,
|
1756 |
+
"eval_samples_per_second": 5.529,
|
1757 |
+
"eval_wer": 0.5220228384991843,
|
1758 |
+
"step": 12500
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 10.31,
|
1762 |
+
"learning_rate": 9.65893587994543e-05,
|
1763 |
+
"loss": 0.1741,
|
1764 |
+
"step": 12600
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 10.31,
|
1768 |
+
"eval_loss": 0.549069344997406,
|
1769 |
+
"eval_runtime": 18.1945,
|
1770 |
+
"eval_samples_per_second": 5.496,
|
1771 |
+
"eval_wer": 0.5171288743882545,
|
1772 |
+
"step": 12600
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 10.39,
|
1776 |
+
"learning_rate": 9.656207366984994e-05,
|
1777 |
+
"loss": 0.1941,
|
1778 |
+
"step": 12700
|
1779 |
+
},
|
1780 |
+
{
|
1781 |
+
"epoch": 10.39,
|
1782 |
+
"eval_loss": 0.5582250356674194,
|
1783 |
+
"eval_runtime": 18.1284,
|
1784 |
+
"eval_samples_per_second": 5.516,
|
1785 |
+
"eval_wer": 0.5187601957585645,
|
1786 |
+
"step": 12700
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 10.47,
|
1790 |
+
"learning_rate": 9.653478854024557e-05,
|
1791 |
+
"loss": 0.197,
|
1792 |
+
"step": 12800
|
1793 |
+
},
|
1794 |
+
{
|
1795 |
+
"epoch": 10.47,
|
1796 |
+
"eval_loss": 0.5502349138259888,
|
1797 |
+
"eval_runtime": 18.1653,
|
1798 |
+
"eval_samples_per_second": 5.505,
|
1799 |
+
"eval_wer": 0.5040783034257749,
|
1800 |
+
"step": 12800
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 10.56,
|
1804 |
+
"learning_rate": 9.65075034106412e-05,
|
1805 |
+
"loss": 0.1758,
|
1806 |
+
"step": 12900
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 10.56,
|
1810 |
+
"eval_loss": 0.5512282848358154,
|
1811 |
+
"eval_runtime": 18.1381,
|
1812 |
+
"eval_samples_per_second": 5.513,
|
1813 |
+
"eval_wer": 0.49592169657422513,
|
1814 |
+
"step": 12900
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 10.64,
|
1818 |
+
"learning_rate": 9.648021828103684e-05,
|
1819 |
+
"loss": 0.1795,
|
1820 |
+
"step": 13000
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 10.64,
|
1824 |
+
"eval_loss": 0.5705503225326538,
|
1825 |
+
"eval_runtime": 18.1085,
|
1826 |
+
"eval_samples_per_second": 5.522,
|
1827 |
+
"eval_wer": 0.4926590538336052,
|
1828 |
+
"step": 13000
|
1829 |
+
},
|
1830 |
+
{
|
1831 |
+
"epoch": 10.72,
|
1832 |
+
"learning_rate": 9.645293315143247e-05,
|
1833 |
+
"loss": 0.1908,
|
1834 |
+
"step": 13100
|
1835 |
+
},
|
1836 |
+
{
|
1837 |
+
"epoch": 10.72,
|
1838 |
+
"eval_loss": 0.5005561113357544,
|
1839 |
+
"eval_runtime": 18.2399,
|
1840 |
+
"eval_samples_per_second": 5.482,
|
1841 |
+
"eval_wer": 0.499184339314845,
|
1842 |
+
"step": 13100
|
1843 |
+
},
|
1844 |
+
{
|
1845 |
+
"epoch": 10.8,
|
1846 |
+
"learning_rate": 9.64256480218281e-05,
|
1847 |
+
"loss": 0.1849,
|
1848 |
+
"step": 13200
|
1849 |
+
},
|
1850 |
+
{
|
1851 |
+
"epoch": 10.8,
|
1852 |
+
"eval_loss": 0.4815738797187805,
|
1853 |
+
"eval_runtime": 18.0612,
|
1854 |
+
"eval_samples_per_second": 5.537,
|
1855 |
+
"eval_wer": 0.49755301794453505,
|
1856 |
+
"step": 13200
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 10.88,
|
1860 |
+
"learning_rate": 9.639836289222375e-05,
|
1861 |
+
"loss": 0.1746,
|
1862 |
+
"step": 13300
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 10.88,
|
1866 |
+
"eval_loss": 0.5251672863960266,
|
1867 |
+
"eval_runtime": 18.1744,
|
1868 |
+
"eval_samples_per_second": 5.502,
|
1869 |
+
"eval_wer": 0.5057096247960848,
|
1870 |
+
"step": 13300
|
1871 |
+
},
|
1872 |
+
{
|
1873 |
+
"epoch": 10.97,
|
1874 |
+
"learning_rate": 9.637107776261937e-05,
|
1875 |
+
"loss": 0.1805,
|
1876 |
+
"step": 13400
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 10.97,
|
1880 |
+
"eval_loss": 0.6231664419174194,
|
1881 |
+
"eval_runtime": 18.1307,
|
1882 |
+
"eval_samples_per_second": 5.515,
|
1883 |
+
"eval_wer": 0.5301794453507341,
|
1884 |
+
"step": 13400
|
1885 |
+
},
|
1886 |
+
{
|
1887 |
+
"epoch": 11.05,
|
1888 |
+
"learning_rate": 9.6343792633015e-05,
|
1889 |
+
"loss": 0.1585,
|
1890 |
+
"step": 13500
|
1891 |
+
},
|
1892 |
+
{
|
1893 |
+
"epoch": 11.05,
|
1894 |
+
"eval_loss": 0.639666736125946,
|
1895 |
+
"eval_runtime": 18.0378,
|
1896 |
+
"eval_samples_per_second": 5.544,
|
1897 |
+
"eval_wer": 0.5024469820554649,
|
1898 |
+
"step": 13500
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 11.13,
|
1902 |
+
"learning_rate": 9.631650750341065e-05,
|
1903 |
+
"loss": 0.1676,
|
1904 |
+
"step": 13600
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 11.13,
|
1908 |
+
"eval_loss": 0.5459147095680237,
|
1909 |
+
"eval_runtime": 18.1234,
|
1910 |
+
"eval_samples_per_second": 5.518,
|
1911 |
+
"eval_wer": 0.5008156606851549,
|
1912 |
+
"step": 13600
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 11.21,
|
1916 |
+
"learning_rate": 9.628922237380627e-05,
|
1917 |
+
"loss": 0.1797,
|
1918 |
+
"step": 13700
|
1919 |
+
},
|
1920 |
+
{
|
1921 |
+
"epoch": 11.21,
|
1922 |
+
"eval_loss": 0.542678952217102,
|
1923 |
+
"eval_runtime": 18.1227,
|
1924 |
+
"eval_samples_per_second": 5.518,
|
1925 |
+
"eval_wer": 0.5089722675367048,
|
1926 |
+
"step": 13700
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 11.29,
|
1930 |
+
"learning_rate": 9.626193724420192e-05,
|
1931 |
+
"loss": 0.1587,
|
1932 |
+
"step": 13800
|
1933 |
+
},
|
1934 |
+
{
|
1935 |
+
"epoch": 11.29,
|
1936 |
+
"eval_loss": 0.5375781059265137,
|
1937 |
+
"eval_runtime": 18.0437,
|
1938 |
+
"eval_samples_per_second": 5.542,
|
1939 |
+
"eval_wer": 0.5106035889070146,
|
1940 |
+
"step": 13800
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 11.37,
|
1944 |
+
"learning_rate": 9.623465211459755e-05,
|
1945 |
+
"loss": 0.1786,
|
1946 |
+
"step": 13900
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 11.37,
|
1950 |
+
"eval_loss": 0.5375598669052124,
|
1951 |
+
"eval_runtime": 18.3192,
|
1952 |
+
"eval_samples_per_second": 5.459,
|
1953 |
+
"eval_wer": 0.5008156606851549,
|
1954 |
+
"step": 13900
|
1955 |
+
},
|
1956 |
+
{
|
1957 |
+
"epoch": 11.46,
|
1958 |
+
"learning_rate": 9.620736698499318e-05,
|
1959 |
+
"loss": 0.1771,
|
1960 |
+
"step": 14000
|
1961 |
+
},
|
1962 |
+
{
|
1963 |
+
"epoch": 11.46,
|
1964 |
+
"eval_loss": 0.5227739214897156,
|
1965 |
+
"eval_runtime": 18.3436,
|
1966 |
+
"eval_samples_per_second": 5.452,
|
1967 |
+
"eval_wer": 0.4910277324632953,
|
1968 |
+
"step": 14000
|
1969 |
+
},
|
1970 |
+
{
|
1971 |
+
"epoch": 11.54,
|
1972 |
+
"learning_rate": 9.618008185538882e-05,
|
1973 |
+
"loss": 0.1668,
|
1974 |
+
"step": 14100
|
1975 |
+
},
|
1976 |
+
{
|
1977 |
+
"epoch": 11.54,
|
1978 |
+
"eval_loss": 0.54190593957901,
|
1979 |
+
"eval_runtime": 18.2145,
|
1980 |
+
"eval_samples_per_second": 5.49,
|
1981 |
+
"eval_wer": 0.499184339314845,
|
1982 |
+
"step": 14100
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 11.62,
|
1986 |
+
"learning_rate": 9.615279672578445e-05,
|
1987 |
+
"loss": 0.1628,
|
1988 |
+
"step": 14200
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 11.62,
|
1992 |
+
"eval_loss": 0.49297595024108887,
|
1993 |
+
"eval_runtime": 18.1064,
|
1994 |
+
"eval_samples_per_second": 5.523,
|
1995 |
+
"eval_wer": 0.48613376835236544,
|
1996 |
+
"step": 14200
|
1997 |
+
},
|
1998 |
+
{
|
1999 |
+
"epoch": 11.7,
|
2000 |
+
"learning_rate": 9.612551159618008e-05,
|
2001 |
+
"loss": 0.1728,
|
2002 |
+
"step": 14300
|
2003 |
+
},
|
2004 |
+
{
|
2005 |
+
"epoch": 11.7,
|
2006 |
+
"eval_loss": 0.4872676432132721,
|
2007 |
+
"eval_runtime": 18.2245,
|
2008 |
+
"eval_samples_per_second": 5.487,
|
2009 |
+
"eval_wer": 0.48776508972267535,
|
2010 |
+
"step": 14300
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 11.78,
|
2014 |
+
"learning_rate": 9.609822646657573e-05,
|
2015 |
+
"loss": 0.1876,
|
2016 |
+
"step": 14400
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 11.78,
|
2020 |
+
"eval_loss": 0.5935976505279541,
|
2021 |
+
"eval_runtime": 18.1277,
|
2022 |
+
"eval_samples_per_second": 5.516,
|
2023 |
+
"eval_wer": 0.5220228384991843,
|
2024 |
+
"step": 14400
|
2025 |
+
},
|
2026 |
+
{
|
2027 |
+
"epoch": 11.87,
|
2028 |
+
"learning_rate": 9.607094133697135e-05,
|
2029 |
+
"loss": 0.1732,
|
2030 |
+
"step": 14500
|
2031 |
+
},
|
2032 |
+
{
|
2033 |
+
"epoch": 11.87,
|
2034 |
+
"eval_loss": 0.5435391068458557,
|
2035 |
+
"eval_runtime": 18.2852,
|
2036 |
+
"eval_samples_per_second": 5.469,
|
2037 |
+
"eval_wer": 0.49429037520391517,
|
2038 |
+
"step": 14500
|
2039 |
+
},
|
2040 |
+
{
|
2041 |
+
"epoch": 11.95,
|
2042 |
+
"learning_rate": 9.6043656207367e-05,
|
2043 |
+
"loss": 0.1794,
|
2044 |
+
"step": 14600
|
2045 |
+
},
|
2046 |
+
{
|
2047 |
+
"epoch": 11.95,
|
2048 |
+
"eval_loss": 0.5519204139709473,
|
2049 |
+
"eval_runtime": 18.2453,
|
2050 |
+
"eval_samples_per_second": 5.481,
|
2051 |
+
"eval_wer": 0.5138662316476346,
|
2052 |
+
"step": 14600
|
2053 |
+
},
|
2054 |
+
{
|
2055 |
+
"epoch": 12.03,
|
2056 |
+
"learning_rate": 9.601637107776262e-05,
|
2057 |
+
"loss": 0.1696,
|
2058 |
+
"step": 14700
|
2059 |
+
},
|
2060 |
+
{
|
2061 |
+
"epoch": 12.03,
|
2062 |
+
"eval_loss": 0.5955798029899597,
|
2063 |
+
"eval_runtime": 18.1714,
|
2064 |
+
"eval_samples_per_second": 5.503,
|
2065 |
+
"eval_wer": 0.5252854812398042,
|
2066 |
+
"step": 14700
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 12.11,
|
2070 |
+
"learning_rate": 9.598908594815826e-05,
|
2071 |
+
"loss": 0.1665,
|
2072 |
+
"step": 14800
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 12.11,
|
2076 |
+
"eval_loss": 0.5391552448272705,
|
2077 |
+
"eval_runtime": 18.128,
|
2078 |
+
"eval_samples_per_second": 5.516,
|
2079 |
+
"eval_wer": 0.49592169657422513,
|
2080 |
+
"step": 14800
|
2081 |
+
},
|
2082 |
+
{
|
2083 |
+
"epoch": 12.19,
|
2084 |
+
"learning_rate": 9.59618008185539e-05,
|
2085 |
+
"loss": 0.1668,
|
2086 |
+
"step": 14900
|
2087 |
+
},
|
2088 |
+
{
|
2089 |
+
"epoch": 12.19,
|
2090 |
+
"eval_loss": 0.523460865020752,
|
2091 |
+
"eval_runtime": 18.2149,
|
2092 |
+
"eval_samples_per_second": 5.49,
|
2093 |
+
"eval_wer": 0.499184339314845,
|
2094 |
+
"step": 14900
|
2095 |
+
},
|
2096 |
+
{
|
2097 |
+
"epoch": 12.27,
|
2098 |
+
"learning_rate": 9.593451568894952e-05,
|
2099 |
+
"loss": 0.1656,
|
2100 |
+
"step": 15000
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 12.27,
|
2104 |
+
"eval_loss": 0.6226036548614502,
|
2105 |
+
"eval_runtime": 18.3111,
|
2106 |
+
"eval_samples_per_second": 5.461,
|
2107 |
+
"eval_wer": 0.5089722675367048,
|
2108 |
+
"step": 15000
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 12.36,
|
2112 |
+
"learning_rate": 9.590723055934516e-05,
|
2113 |
+
"loss": 0.1652,
|
2114 |
+
"step": 15100
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 12.36,
|
2118 |
+
"eval_loss": 0.6167003512382507,
|
2119 |
+
"eval_runtime": 18.2279,
|
2120 |
+
"eval_samples_per_second": 5.486,
|
2121 |
+
"eval_wer": 0.5138662316476346,
|
2122 |
+
"step": 15100
|
2123 |
+
},
|
2124 |
+
{
|
2125 |
+
"epoch": 12.44,
|
2126 |
+
"learning_rate": 9.58799454297408e-05,
|
2127 |
+
"loss": 0.1576,
|
2128 |
+
"step": 15200
|
2129 |
+
},
|
2130 |
+
{
|
2131 |
+
"epoch": 12.44,
|
2132 |
+
"eval_loss": 0.561502993106842,
|
2133 |
+
"eval_runtime": 18.1886,
|
2134 |
+
"eval_samples_per_second": 5.498,
|
2135 |
+
"eval_wer": 0.5106035889070146,
|
2136 |
+
"step": 15200
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 12.52,
|
2140 |
+
"learning_rate": 9.585266030013643e-05,
|
2141 |
+
"loss": 0.1483,
|
2142 |
+
"step": 15300
|
2143 |
+
},
|
2144 |
+
{
|
2145 |
+
"epoch": 12.52,
|
2146 |
+
"eval_loss": 0.5715560913085938,
|
2147 |
+
"eval_runtime": 18.3019,
|
2148 |
+
"eval_samples_per_second": 5.464,
|
2149 |
+
"eval_wer": 0.5252854812398042,
|
2150 |
+
"step": 15300
|
2151 |
+
},
|
2152 |
+
{
|
2153 |
+
"epoch": 12.6,
|
2154 |
+
"learning_rate": 9.582537517053206e-05,
|
2155 |
+
"loss": 0.1713,
|
2156 |
+
"step": 15400
|
2157 |
+
},
|
2158 |
+
{
|
2159 |
+
"epoch": 12.6,
|
2160 |
+
"eval_loss": 0.5943311452865601,
|
2161 |
+
"eval_runtime": 18.2667,
|
2162 |
+
"eval_samples_per_second": 5.474,
|
2163 |
+
"eval_wer": 0.5024469820554649,
|
2164 |
+
"step": 15400
|
2165 |
+
},
|
2166 |
+
{
|
2167 |
+
"epoch": 12.68,
|
2168 |
+
"learning_rate": 9.57980900409277e-05,
|
2169 |
+
"loss": 0.1496,
|
2170 |
+
"step": 15500
|
2171 |
+
},
|
2172 |
+
{
|
2173 |
+
"epoch": 12.68,
|
2174 |
+
"eval_loss": 0.5450413227081299,
|
2175 |
+
"eval_runtime": 19.1731,
|
2176 |
+
"eval_samples_per_second": 5.216,
|
2177 |
+
"eval_wer": 0.49755301794453505,
|
2178 |
+
"step": 15500
|
2179 |
+
},
|
2180 |
+
{
|
2181 |
+
"epoch": 12.77,
|
2182 |
+
"learning_rate": 9.577080491132333e-05,
|
2183 |
+
"loss": 0.1673,
|
2184 |
+
"step": 15600
|
2185 |
+
},
|
2186 |
+
{
|
2187 |
+
"epoch": 12.77,
|
2188 |
+
"eval_loss": 0.5580278635025024,
|
2189 |
+
"eval_runtime": 18.1472,
|
2190 |
+
"eval_samples_per_second": 5.51,
|
2191 |
+
"eval_wer": 0.49429037520391517,
|
2192 |
+
"step": 15600
|
2193 |
+
},
|
2194 |
+
{
|
2195 |
+
"epoch": 12.85,
|
2196 |
+
"learning_rate": 9.574351978171898e-05,
|
2197 |
+
"loss": 0.1596,
|
2198 |
+
"step": 15700
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 12.85,
|
2202 |
+
"eval_loss": 0.5721789002418518,
|
2203 |
+
"eval_runtime": 18.2542,
|
2204 |
+
"eval_samples_per_second": 5.478,
|
2205 |
+
"eval_wer": 0.5073409461663948,
|
2206 |
+
"step": 15700
|
2207 |
+
},
|
2208 |
+
{
|
2209 |
+
"epoch": 12.93,
|
2210 |
+
"learning_rate": 9.57162346521146e-05,
|
2211 |
+
"loss": 0.1588,
|
2212 |
+
"step": 15800
|
2213 |
+
},
|
2214 |
+
{
|
2215 |
+
"epoch": 12.93,
|
2216 |
+
"eval_loss": 0.5455964803695679,
|
2217 |
+
"eval_runtime": 18.234,
|
2218 |
+
"eval_samples_per_second": 5.484,
|
2219 |
+
"eval_wer": 0.5073409461663948,
|
2220 |
+
"step": 15800
|
2221 |
+
},
|
2222 |
+
{
|
2223 |
+
"epoch": 13.01,
|
2224 |
+
"learning_rate": 9.568894952251024e-05,
|
2225 |
+
"loss": 0.1737,
|
2226 |
+
"step": 15900
|
2227 |
+
},
|
2228 |
+
{
|
2229 |
+
"epoch": 13.01,
|
2230 |
+
"eval_loss": 0.5726311206817627,
|
2231 |
+
"eval_runtime": 18.2177,
|
2232 |
+
"eval_samples_per_second": 5.489,
|
2233 |
+
"eval_wer": 0.5040783034257749,
|
2234 |
+
"step": 15900
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 13.09,
|
2238 |
+
"learning_rate": 9.566166439290588e-05,
|
2239 |
+
"loss": 0.1571,
|
2240 |
+
"step": 16000
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 13.09,
|
2244 |
+
"eval_loss": 0.579778790473938,
|
2245 |
+
"eval_runtime": 18.1586,
|
2246 |
+
"eval_samples_per_second": 5.507,
|
2247 |
+
"eval_wer": 0.5122349102773246,
|
2248 |
+
"step": 16000
|
2249 |
+
},
|
2250 |
+
{
|
2251 |
+
"epoch": 13.18,
|
2252 |
+
"learning_rate": 9.563437926330151e-05,
|
2253 |
+
"loss": 0.1486,
|
2254 |
+
"step": 16100
|
2255 |
+
},
|
2256 |
+
{
|
2257 |
+
"epoch": 13.18,
|
2258 |
+
"eval_loss": 0.5972344279289246,
|
2259 |
+
"eval_runtime": 18.1733,
|
2260 |
+
"eval_samples_per_second": 5.503,
|
2261 |
+
"eval_wer": 0.5106035889070146,
|
2262 |
+
"step": 16100
|
2263 |
+
},
|
2264 |
+
{
|
2265 |
+
"epoch": 13.26,
|
2266 |
+
"learning_rate": 9.560709413369714e-05,
|
2267 |
+
"loss": 0.1551,
|
2268 |
+
"step": 16200
|
2269 |
+
},
|
2270 |
+
{
|
2271 |
+
"epoch": 13.26,
|
2272 |
+
"eval_loss": 0.6507982015609741,
|
2273 |
+
"eval_runtime": 18.1445,
|
2274 |
+
"eval_samples_per_second": 5.511,
|
2275 |
+
"eval_wer": 0.5171288743882545,
|
2276 |
+
"step": 16200
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 13.34,
|
2280 |
+
"learning_rate": 9.557980900409276e-05,
|
2281 |
+
"loss": 0.1655,
|
2282 |
+
"step": 16300
|
2283 |
+
},
|
2284 |
+
{
|
2285 |
+
"epoch": 13.34,
|
2286 |
+
"eval_loss": 0.5814386606216431,
|
2287 |
+
"eval_runtime": 18.3148,
|
2288 |
+
"eval_samples_per_second": 5.46,
|
2289 |
+
"eval_wer": 0.5154975530179445,
|
2290 |
+
"step": 16300
|
2291 |
+
},
|
2292 |
+
{
|
2293 |
+
"epoch": 13.42,
|
2294 |
+
"learning_rate": 9.555252387448841e-05,
|
2295 |
+
"loss": 0.1673,
|
2296 |
+
"step": 16400
|
2297 |
+
},
|
2298 |
+
{
|
2299 |
+
"epoch": 13.42,
|
2300 |
+
"eval_loss": 0.5608506202697754,
|
2301 |
+
"eval_runtime": 18.1216,
|
2302 |
+
"eval_samples_per_second": 5.518,
|
2303 |
+
"eval_wer": 0.49429037520391517,
|
2304 |
+
"step": 16400
|
2305 |
+
},
|
2306 |
+
{
|
2307 |
+
"epoch": 13.5,
|
2308 |
+
"learning_rate": 9.552523874488404e-05,
|
2309 |
+
"loss": 0.1453,
|
2310 |
+
"step": 16500
|
2311 |
+
},
|
2312 |
+
{
|
2313 |
+
"epoch": 13.5,
|
2314 |
+
"eval_loss": 0.5696703195571899,
|
2315 |
+
"eval_runtime": 18.192,
|
2316 |
+
"eval_samples_per_second": 5.497,
|
2317 |
+
"eval_wer": 0.499184339314845,
|
2318 |
+
"step": 16500
|
2319 |
+
},
|
2320 |
+
{
|
2321 |
+
"epoch": 13.58,
|
2322 |
+
"learning_rate": 9.549795361527968e-05,
|
2323 |
+
"loss": 0.1497,
|
2324 |
+
"step": 16600
|
2325 |
+
},
|
2326 |
+
{
|
2327 |
+
"epoch": 13.58,
|
2328 |
+
"eval_loss": 0.5988049507141113,
|
2329 |
+
"eval_runtime": 18.1244,
|
2330 |
+
"eval_samples_per_second": 5.517,
|
2331 |
+
"eval_wer": 0.5057096247960848,
|
2332 |
+
"step": 16600
|
2333 |
+
},
|
2334 |
+
{
|
2335 |
+
"epoch": 13.67,
|
2336 |
+
"learning_rate": 9.547066848567531e-05,
|
2337 |
+
"loss": 0.1648,
|
2338 |
+
"step": 16700
|
2339 |
+
},
|
2340 |
+
{
|
2341 |
+
"epoch": 13.67,
|
2342 |
+
"eval_loss": 0.5495877861976624,
|
2343 |
+
"eval_runtime": 18.2527,
|
2344 |
+
"eval_samples_per_second": 5.479,
|
2345 |
+
"eval_wer": 0.4926590538336052,
|
2346 |
+
"step": 16700
|
2347 |
+
},
|
2348 |
+
{
|
2349 |
+
"epoch": 13.75,
|
2350 |
+
"learning_rate": 9.544338335607094e-05,
|
2351 |
+
"loss": 0.1521,
|
2352 |
+
"step": 16800
|
2353 |
+
},
|
2354 |
+
{
|
2355 |
+
"epoch": 13.75,
|
2356 |
+
"eval_loss": 0.5653265118598938,
|
2357 |
+
"eval_runtime": 18.0473,
|
2358 |
+
"eval_samples_per_second": 5.541,
|
2359 |
+
"eval_wer": 0.49592169657422513,
|
2360 |
+
"step": 16800
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 13.83,
|
2364 |
+
"learning_rate": 9.541609822646657e-05,
|
2365 |
+
"loss": 0.1485,
|
2366 |
+
"step": 16900
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 13.83,
|
2370 |
+
"eval_loss": 0.5756533145904541,
|
2371 |
+
"eval_runtime": 18.163,
|
2372 |
+
"eval_samples_per_second": 5.506,
|
2373 |
+
"eval_wer": 0.5040783034257749,
|
2374 |
+
"step": 16900
|
2375 |
+
},
|
2376 |
+
{
|
2377 |
+
"epoch": 13.91,
|
2378 |
+
"learning_rate": 9.538881309686222e-05,
|
2379 |
+
"loss": 0.1662,
|
2380 |
+
"step": 17000
|
2381 |
+
},
|
2382 |
+
{
|
2383 |
+
"epoch": 13.91,
|
2384 |
+
"eval_loss": 0.5320442318916321,
|
2385 |
+
"eval_runtime": 18.1679,
|
2386 |
+
"eval_samples_per_second": 5.504,
|
2387 |
+
"eval_wer": 0.5040783034257749,
|
2388 |
+
"step": 17000
|
2389 |
+
},
|
2390 |
+
{
|
2391 |
+
"epoch": 13.99,
|
2392 |
+
"learning_rate": 9.536152796725784e-05,
|
2393 |
+
"loss": 0.167,
|
2394 |
+
"step": 17100
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 13.99,
|
2398 |
+
"eval_loss": 0.5141332745552063,
|
2399 |
+
"eval_runtime": 18.227,
|
2400 |
+
"eval_samples_per_second": 5.486,
|
2401 |
+
"eval_wer": 0.5008156606851549,
|
2402 |
+
"step": 17100
|
2403 |
+
},
|
2404 |
+
{
|
2405 |
+
"epoch": 14.08,
|
2406 |
+
"learning_rate": 9.533424283765349e-05,
|
2407 |
+
"loss": 0.1588,
|
2408 |
+
"step": 17200
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 14.08,
|
2412 |
+
"eval_loss": 0.5980657339096069,
|
2413 |
+
"eval_runtime": 18.0668,
|
2414 |
+
"eval_samples_per_second": 5.535,
|
2415 |
+
"eval_wer": 0.5089722675367048,
|
2416 |
+
"step": 17200
|
2417 |
+
},
|
2418 |
+
{
|
2419 |
+
"epoch": 14.16,
|
2420 |
+
"learning_rate": 9.530695770804912e-05,
|
2421 |
+
"loss": 0.1512,
|
2422 |
+
"step": 17300
|
2423 |
+
},
|
2424 |
+
{
|
2425 |
+
"epoch": 14.16,
|
2426 |
+
"eval_loss": 0.5132100582122803,
|
2427 |
+
"eval_runtime": 18.2884,
|
2428 |
+
"eval_samples_per_second": 5.468,
|
2429 |
+
"eval_wer": 0.5122349102773246,
|
2430 |
+
"step": 17300
|
2431 |
+
},
|
2432 |
+
{
|
2433 |
+
"epoch": 14.24,
|
2434 |
+
"learning_rate": 9.527967257844475e-05,
|
2435 |
+
"loss": 0.1407,
|
2436 |
+
"step": 17400
|
2437 |
+
},
|
2438 |
+
{
|
2439 |
+
"epoch": 14.24,
|
2440 |
+
"eval_loss": 0.5253714919090271,
|
2441 |
+
"eval_runtime": 18.1261,
|
2442 |
+
"eval_samples_per_second": 5.517,
|
2443 |
+
"eval_wer": 0.49592169657422513,
|
2444 |
+
"step": 17400
|
2445 |
+
},
|
2446 |
+
{
|
2447 |
+
"epoch": 14.32,
|
2448 |
+
"learning_rate": 9.525238744884039e-05,
|
2449 |
+
"loss": 0.1385,
|
2450 |
+
"step": 17500
|
2451 |
+
},
|
2452 |
+
{
|
2453 |
+
"epoch": 14.32,
|
2454 |
+
"eval_loss": 0.516293466091156,
|
2455 |
+
"eval_runtime": 18.1921,
|
2456 |
+
"eval_samples_per_second": 5.497,
|
2457 |
+
"eval_wer": 0.48613376835236544,
|
2458 |
+
"step": 17500
|
2459 |
+
},
|
2460 |
+
{
|
2461 |
+
"epoch": 14.4,
|
2462 |
+
"learning_rate": 9.522510231923602e-05,
|
2463 |
+
"loss": 0.1454,
|
2464 |
+
"step": 17600
|
2465 |
+
},
|
2466 |
+
{
|
2467 |
+
"epoch": 14.4,
|
2468 |
+
"eval_loss": 0.5024062991142273,
|
2469 |
+
"eval_runtime": 18.1823,
|
2470 |
+
"eval_samples_per_second": 5.5,
|
2471 |
+
"eval_wer": 0.4698205546492659,
|
2472 |
+
"step": 17600
|
2473 |
+
},
|
2474 |
+
{
|
2475 |
+
"epoch": 14.48,
|
2476 |
+
"learning_rate": 9.519781718963165e-05,
|
2477 |
+
"loss": 0.1574,
|
2478 |
+
"step": 17700
|
2479 |
+
},
|
2480 |
+
{
|
2481 |
+
"epoch": 14.48,
|
2482 |
+
"eval_loss": 0.5171040296554565,
|
2483 |
+
"eval_runtime": 18.1815,
|
2484 |
+
"eval_samples_per_second": 5.5,
|
2485 |
+
"eval_wer": 0.4926590538336052,
|
2486 |
+
"step": 17700
|
2487 |
+
},
|
2488 |
+
{
|
2489 |
+
"epoch": 14.57,
|
2490 |
+
"learning_rate": 9.517053206002729e-05,
|
2491 |
+
"loss": 0.1481,
|
2492 |
+
"step": 17800
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 14.57,
|
2496 |
+
"eval_loss": 0.5247868895530701,
|
2497 |
+
"eval_runtime": 18.1938,
|
2498 |
+
"eval_samples_per_second": 5.496,
|
2499 |
+
"eval_wer": 0.48776508972267535,
|
2500 |
+
"step": 17800
|
2501 |
+
},
|
2502 |
+
{
|
2503 |
+
"epoch": 14.65,
|
2504 |
+
"learning_rate": 9.514324693042292e-05,
|
2505 |
+
"loss": 0.1382,
|
2506 |
+
"step": 17900
|
2507 |
+
},
|
2508 |
+
{
|
2509 |
+
"epoch": 14.65,
|
2510 |
+
"eval_loss": 0.5333756804466248,
|
2511 |
+
"eval_runtime": 18.3407,
|
2512 |
+
"eval_samples_per_second": 5.452,
|
2513 |
+
"eval_wer": 0.47797716150081565,
|
2514 |
+
"step": 17900
|
2515 |
+
},
|
2516 |
+
{
|
2517 |
+
"epoch": 14.73,
|
2518 |
+
"learning_rate": 9.511596180081855e-05,
|
2519 |
+
"loss": 0.1557,
|
2520 |
+
"step": 18000
|
2521 |
+
},
|
2522 |
+
{
|
2523 |
+
"epoch": 14.73,
|
2524 |
+
"eval_loss": 0.5148436427116394,
|
2525 |
+
"eval_runtime": 18.1439,
|
2526 |
+
"eval_samples_per_second": 5.511,
|
2527 |
+
"eval_wer": 0.4893964110929853,
|
2528 |
+
"step": 18000
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 14.81,
|
2532 |
+
"learning_rate": 9.50886766712142e-05,
|
2533 |
+
"loss": 0.1559,
|
2534 |
+
"step": 18100
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 14.81,
|
2538 |
+
"eval_loss": 0.5419223308563232,
|
2539 |
+
"eval_runtime": 18.3072,
|
2540 |
+
"eval_samples_per_second": 5.462,
|
2541 |
+
"eval_wer": 0.5024469820554649,
|
2542 |
+
"step": 18100
|
2543 |
+
},
|
2544 |
+
{
|
2545 |
+
"epoch": 14.89,
|
2546 |
+
"learning_rate": 9.506139154160982e-05,
|
2547 |
+
"loss": 0.1412,
|
2548 |
+
"step": 18200
|
2549 |
+
},
|
2550 |
+
{
|
2551 |
+
"epoch": 14.89,
|
2552 |
+
"eval_loss": 0.5893602967262268,
|
2553 |
+
"eval_runtime": 18.1299,
|
2554 |
+
"eval_samples_per_second": 5.516,
|
2555 |
+
"eval_wer": 0.49592169657422513,
|
2556 |
+
"step": 18200
|
2557 |
+
},
|
2558 |
+
{
|
2559 |
+
"epoch": 14.98,
|
2560 |
+
"learning_rate": 9.503410641200547e-05,
|
2561 |
+
"loss": 0.1508,
|
2562 |
+
"step": 18300
|
2563 |
+
},
|
2564 |
+
{
|
2565 |
+
"epoch": 14.98,
|
2566 |
+
"eval_loss": 0.5416052937507629,
|
2567 |
+
"eval_runtime": 18.2068,
|
2568 |
+
"eval_samples_per_second": 5.492,
|
2569 |
+
"eval_wer": 0.5057096247960848,
|
2570 |
+
"step": 18300
|
2571 |
+
},
|
2572 |
+
{
|
2573 |
+
"epoch": 15.06,
|
2574 |
+
"learning_rate": 9.500682128240109e-05,
|
2575 |
+
"loss": 0.1461,
|
2576 |
+
"step": 18400
|
2577 |
+
},
|
2578 |
+
{
|
2579 |
+
"epoch": 15.06,
|
2580 |
+
"eval_loss": 0.5962003469467163,
|
2581 |
+
"eval_runtime": 18.1627,
|
2582 |
+
"eval_samples_per_second": 5.506,
|
2583 |
+
"eval_wer": 0.49755301794453505,
|
2584 |
+
"step": 18400
|
2585 |
+
},
|
2586 |
+
{
|
2587 |
+
"epoch": 15.14,
|
2588 |
+
"learning_rate": 9.497953615279673e-05,
|
2589 |
+
"loss": 0.1503,
|
2590 |
+
"step": 18500
|
2591 |
+
},
|
2592 |
+
{
|
2593 |
+
"epoch": 15.14,
|
2594 |
+
"eval_loss": 0.59946209192276,
|
2595 |
+
"eval_runtime": 18.2979,
|
2596 |
+
"eval_samples_per_second": 5.465,
|
2597 |
+
"eval_wer": 0.5073409461663948,
|
2598 |
+
"step": 18500
|
2599 |
+
},
|
2600 |
+
{
|
2601 |
+
"epoch": 15.22,
|
2602 |
+
"learning_rate": 9.495225102319237e-05,
|
2603 |
+
"loss": 0.148,
|
2604 |
+
"step": 18600
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 15.22,
|
2608 |
+
"eval_loss": 0.615426778793335,
|
2609 |
+
"eval_runtime": 18.13,
|
2610 |
+
"eval_samples_per_second": 5.516,
|
2611 |
+
"eval_wer": 0.5057096247960848,
|
2612 |
+
"step": 18600
|
2613 |
+
},
|
2614 |
+
{
|
2615 |
+
"epoch": 15.3,
|
2616 |
+
"learning_rate": 9.4924965893588e-05,
|
2617 |
+
"loss": 0.1427,
|
2618 |
+
"step": 18700
|
2619 |
+
},
|
2620 |
+
{
|
2621 |
+
"epoch": 15.3,
|
2622 |
+
"eval_loss": 0.6145089864730835,
|
2623 |
+
"eval_runtime": 18.1833,
|
2624 |
+
"eval_samples_per_second": 5.5,
|
2625 |
+
"eval_wer": 0.5106035889070146,
|
2626 |
+
"step": 18700
|
2627 |
+
},
|
2628 |
+
{
|
2629 |
+
"epoch": 15.38,
|
2630 |
+
"learning_rate": 9.489768076398363e-05,
|
2631 |
+
"loss": 0.1456,
|
2632 |
+
"step": 18800
|
2633 |
+
},
|
2634 |
+
{
|
2635 |
+
"epoch": 15.38,
|
2636 |
+
"eval_loss": 0.6026967763900757,
|
2637 |
+
"eval_runtime": 18.3448,
|
2638 |
+
"eval_samples_per_second": 5.451,
|
2639 |
+
"eval_wer": 0.5040783034257749,
|
2640 |
+
"step": 18800
|
2641 |
+
},
|
2642 |
+
{
|
2643 |
+
"epoch": 15.47,
|
2644 |
+
"learning_rate": 9.487039563437927e-05,
|
2645 |
+
"loss": 0.1463,
|
2646 |
+
"step": 18900
|
2647 |
+
},
|
2648 |
+
{
|
2649 |
+
"epoch": 15.47,
|
2650 |
+
"eval_loss": 0.6294275522232056,
|
2651 |
+
"eval_runtime": 18.1136,
|
2652 |
+
"eval_samples_per_second": 5.521,
|
2653 |
+
"eval_wer": 0.48613376835236544,
|
2654 |
+
"step": 18900
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 15.55,
|
2658 |
+
"learning_rate": 9.48431105047749e-05,
|
2659 |
+
"loss": 0.1534,
|
2660 |
+
"step": 19000
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 15.55,
|
2664 |
+
"eval_loss": 0.656607449054718,
|
2665 |
+
"eval_runtime": 18.1552,
|
2666 |
+
"eval_samples_per_second": 5.508,
|
2667 |
+
"eval_wer": 0.5171288743882545,
|
2668 |
+
"step": 19000
|
2669 |
+
},
|
2670 |
+
{
|
2671 |
+
"epoch": 15.63,
|
2672 |
+
"learning_rate": 9.481582537517055e-05,
|
2673 |
+
"loss": 0.1508,
|
2674 |
+
"step": 19100
|
2675 |
+
},
|
2676 |
+
{
|
2677 |
+
"epoch": 15.63,
|
2678 |
+
"eval_loss": 0.6075925827026367,
|
2679 |
+
"eval_runtime": 18.1522,
|
2680 |
+
"eval_samples_per_second": 5.509,
|
2681 |
+
"eval_wer": 0.49755301794453505,
|
2682 |
+
"step": 19100
|
2683 |
+
},
|
2684 |
+
{
|
2685 |
+
"epoch": 15.71,
|
2686 |
+
"learning_rate": 9.478854024556617e-05,
|
2687 |
+
"loss": 0.1403,
|
2688 |
+
"step": 19200
|
2689 |
+
},
|
2690 |
+
{
|
2691 |
+
"epoch": 15.71,
|
2692 |
+
"eval_loss": 0.5606480836868286,
|
2693 |
+
"eval_runtime": 18.178,
|
2694 |
+
"eval_samples_per_second": 5.501,
|
2695 |
+
"eval_wer": 0.4893964110929853,
|
2696 |
+
"step": 19200
|
2697 |
+
},
|
2698 |
+
{
|
2699 |
+
"epoch": 15.79,
|
2700 |
+
"learning_rate": 9.47612551159618e-05,
|
2701 |
+
"loss": 0.1483,
|
2702 |
+
"step": 19300
|
2703 |
+
},
|
2704 |
+
{
|
2705 |
+
"epoch": 15.79,
|
2706 |
+
"eval_loss": 0.5454413890838623,
|
2707 |
+
"eval_runtime": 18.1743,
|
2708 |
+
"eval_samples_per_second": 5.502,
|
2709 |
+
"eval_wer": 0.5122349102773246,
|
2710 |
+
"step": 19300
|
2711 |
+
}
|
2712 |
+
],
|
2713 |
+
"max_steps": 366600,
|
2714 |
+
"num_train_epochs": 300,
|
2715 |
+
"total_flos": 2.1672229688161493e+19,
|
2716 |
+
"trial_name": null,
|
2717 |
+
"trial_params": null
|
2718 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1119aeae94586ea94ee32b7c68aefc1b708f1c5f7f0cd5e54a3b50a04fc31d1f
|
3 |
+
size 2287
|
vocab.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"m": 0, "v": 1, "a": 2, "h": 3, "\u00ef": 4, "f": 5, "\u00e1": 6, "c": 7, "t": 8, "u": 9, "q": 10, "/": 11, "\u00f6": 13, "\u00fc": 14, "e": 15, "l": 16, "\u0165": 17, "\u00ed": 18, "i": 19, "o": 20, "d": 21, "\u010f": 22, "\u016f": 23, "\u2026": 24, "k": 25, "\u00fd": 26, "x": 27, "s": 28, "r": 29, "\u0159": 30, "y": 31, "w": 32, "\u010d": 33, "\u011b": 34, "\u2013": 35, "j": 36, "n": 37, "b": 38, "\u00e8": 39, "\u0161": 40, "g": 41, "\u017e": 42, "\u201e": 43, "\u00e9": 44, "z": 45, "p": 46, "\u2014": 47, "\u00f3": 48, "\u00e4": 49, "\u00fa": 50, "\u0148": 51, "|": 12, "[UNK]": 52, "[PAD]": 53}
|