Commit
•
9db48fb
1
Parent(s):
0508f39
Add model files
Browse files- config.json +76 -0
- optimizer.pt +3 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1570 -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": 37,
|
| 74 |
+
"transformers_version": "4.4.0",
|
| 75 |
+
"vocab_size": 38
|
| 76 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1885ba54b1f88b8a70532ae0d8c228ba0a0c5cbbbe7733a2afb92970b5ea592
|
| 3 |
+
size 2490380826
|
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:d4b38a6977828d09107e72b8a2ddc93fbce4b2a12a03d9f8d3feb03d011c581c
|
| 3 |
+
size 1262085388
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60196e9b4bbb3376d3b5154ad301a76c83588fe7cd19de675a5e110e2ca178b1
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1570 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 112.12090680100755,
|
| 5 |
+
"global_step": 22200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.01,
|
| 12 |
+
"learning_rate": 0.0003,
|
| 13 |
+
"loss": 6.4459,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.01,
|
| 18 |
+
"eval_loss": 3.1488852500915527,
|
| 19 |
+
"eval_runtime": 134.1012,
|
| 20 |
+
"eval_samples_per_second": 4.892,
|
| 21 |
+
"eval_wer": 1.0,
|
| 22 |
+
"step": 200
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.02,
|
| 26 |
+
"learning_rate": 0.00029898648648648646,
|
| 27 |
+
"loss": 3.0913,
|
| 28 |
+
"step": 400
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.02,
|
| 32 |
+
"eval_loss": 3.1127500534057617,
|
| 33 |
+
"eval_runtime": 138.7081,
|
| 34 |
+
"eval_samples_per_second": 4.729,
|
| 35 |
+
"eval_wer": 1.0,
|
| 36 |
+
"step": 400
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.03,
|
| 40 |
+
"learning_rate": 0.00029797297297297294,
|
| 41 |
+
"loss": 3.0716,
|
| 42 |
+
"step": 600
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.03,
|
| 46 |
+
"eval_loss": 3.0870425701141357,
|
| 47 |
+
"eval_runtime": 138.315,
|
| 48 |
+
"eval_samples_per_second": 4.743,
|
| 49 |
+
"eval_wer": 1.0,
|
| 50 |
+
"step": 600
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.04,
|
| 54 |
+
"learning_rate": 0.0002969594594594594,
|
| 55 |
+
"loss": 3.0298,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.04,
|
| 60 |
+
"eval_loss": 3.0003437995910645,
|
| 61 |
+
"eval_runtime": 134.8731,
|
| 62 |
+
"eval_samples_per_second": 4.864,
|
| 63 |
+
"eval_wer": 1.0,
|
| 64 |
+
"step": 800
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 5.05,
|
| 68 |
+
"learning_rate": 0.0002959459459459459,
|
| 69 |
+
"loss": 1.728,
|
| 70 |
+
"step": 1000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 5.05,
|
| 74 |
+
"eval_loss": 0.805480420589447,
|
| 75 |
+
"eval_runtime": 135.0581,
|
| 76 |
+
"eval_samples_per_second": 4.857,
|
| 77 |
+
"eval_wer": 0.8985599322321051,
|
| 78 |
+
"step": 1000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 6.06,
|
| 82 |
+
"learning_rate": 0.0002949324324324324,
|
| 83 |
+
"loss": 0.6228,
|
| 84 |
+
"step": 1200
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.06,
|
| 88 |
+
"eval_loss": 0.6153338551521301,
|
| 89 |
+
"eval_runtime": 138.3824,
|
| 90 |
+
"eval_samples_per_second": 4.74,
|
| 91 |
+
"eval_wer": 0.8263447691656078,
|
| 92 |
+
"step": 1200
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 7.07,
|
| 96 |
+
"learning_rate": 0.0002939189189189189,
|
| 97 |
+
"loss": 0.4109,
|
| 98 |
+
"step": 1400
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.07,
|
| 102 |
+
"eval_loss": 0.5761246085166931,
|
| 103 |
+
"eval_runtime": 136.2401,
|
| 104 |
+
"eval_samples_per_second": 4.815,
|
| 105 |
+
"eval_wer": 0.8009318085556968,
|
| 106 |
+
"step": 1400
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 8.08,
|
| 110 |
+
"learning_rate": 0.00029290540540540536,
|
| 111 |
+
"loss": 0.3384,
|
| 112 |
+
"step": 1600
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.08,
|
| 116 |
+
"eval_loss": 0.6021555662155151,
|
| 117 |
+
"eval_runtime": 135.414,
|
| 118 |
+
"eval_samples_per_second": 4.844,
|
| 119 |
+
"eval_wer": 0.7981787378229563,
|
| 120 |
+
"step": 1600
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 9.09,
|
| 124 |
+
"learning_rate": 0.0002918918918918919,
|
| 125 |
+
"loss": 0.2676,
|
| 126 |
+
"step": 1800
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.09,
|
| 130 |
+
"eval_loss": 0.556929349899292,
|
| 131 |
+
"eval_runtime": 136.0159,
|
| 132 |
+
"eval_samples_per_second": 4.823,
|
| 133 |
+
"eval_wer": 0.7505294366793731,
|
| 134 |
+
"step": 1800
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.1,
|
| 138 |
+
"learning_rate": 0.0002908783783783784,
|
| 139 |
+
"loss": 0.2321,
|
| 140 |
+
"step": 2000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 10.1,
|
| 144 |
+
"eval_loss": 0.6468214392662048,
|
| 145 |
+
"eval_runtime": 135.5988,
|
| 146 |
+
"eval_samples_per_second": 4.838,
|
| 147 |
+
"eval_wer": 0.7736128759000424,
|
| 148 |
+
"step": 2000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 11.11,
|
| 152 |
+
"learning_rate": 0.00028986486486486487,
|
| 153 |
+
"loss": 0.2031,
|
| 154 |
+
"step": 2200
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 11.11,
|
| 158 |
+
"eval_loss": 0.6612684726715088,
|
| 159 |
+
"eval_runtime": 135.8609,
|
| 160 |
+
"eval_samples_per_second": 4.828,
|
| 161 |
+
"eval_wer": 0.7884371029224905,
|
| 162 |
+
"step": 2200
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.12,
|
| 166 |
+
"learning_rate": 0.00028885135135135135,
|
| 167 |
+
"loss": 0.1853,
|
| 168 |
+
"step": 2400
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 12.12,
|
| 172 |
+
"eval_loss": 0.6352970600128174,
|
| 173 |
+
"eval_runtime": 135.9654,
|
| 174 |
+
"eval_samples_per_second": 4.825,
|
| 175 |
+
"eval_wer": 0.736552308343922,
|
| 176 |
+
"step": 2400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 13.13,
|
| 180 |
+
"learning_rate": 0.0002878378378378378,
|
| 181 |
+
"loss": 0.1826,
|
| 182 |
+
"step": 2600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.13,
|
| 186 |
+
"eval_loss": 0.6177716851234436,
|
| 187 |
+
"eval_runtime": 136.1311,
|
| 188 |
+
"eval_samples_per_second": 4.819,
|
| 189 |
+
"eval_wer": 0.7238458280389666,
|
| 190 |
+
"step": 2600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 14.14,
|
| 194 |
+
"learning_rate": 0.00028682432432432426,
|
| 195 |
+
"loss": 0.1627,
|
| 196 |
+
"step": 2800
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 14.14,
|
| 200 |
+
"eval_loss": 0.6629394888877869,
|
| 201 |
+
"eval_runtime": 135.7236,
|
| 202 |
+
"eval_samples_per_second": 4.833,
|
| 203 |
+
"eval_wer": 0.734011012282931,
|
| 204 |
+
"step": 2800
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 15.15,
|
| 208 |
+
"learning_rate": 0.0002858108108108108,
|
| 209 |
+
"loss": 0.1506,
|
| 210 |
+
"step": 3000
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 15.15,
|
| 214 |
+
"eval_loss": 0.6606569886207581,
|
| 215 |
+
"eval_runtime": 137.5748,
|
| 216 |
+
"eval_samples_per_second": 4.768,
|
| 217 |
+
"eval_wer": 0.7272342227869547,
|
| 218 |
+
"step": 3000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 16.16,
|
| 222 |
+
"learning_rate": 0.0002847972972972973,
|
| 223 |
+
"loss": 0.1432,
|
| 224 |
+
"step": 3200
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.16,
|
| 228 |
+
"eval_loss": 0.6748045086860657,
|
| 229 |
+
"eval_runtime": 135.8657,
|
| 230 |
+
"eval_samples_per_second": 4.828,
|
| 231 |
+
"eval_wer": 0.7335874629394324,
|
| 232 |
+
"step": 3200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 17.17,
|
| 236 |
+
"learning_rate": 0.00028378378378378377,
|
| 237 |
+
"loss": 0.1396,
|
| 238 |
+
"step": 3400
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 17.17,
|
| 242 |
+
"eval_loss": 0.6931908130645752,
|
| 243 |
+
"eval_runtime": 136.0012,
|
| 244 |
+
"eval_samples_per_second": 4.823,
|
| 245 |
+
"eval_wer": 0.7236340533672173,
|
| 246 |
+
"step": 3400
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 18.18,
|
| 250 |
+
"learning_rate": 0.00028277027027027025,
|
| 251 |
+
"loss": 0.1397,
|
| 252 |
+
"step": 3600
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 18.18,
|
| 256 |
+
"eval_loss": 0.6838018894195557,
|
| 257 |
+
"eval_runtime": 135.7614,
|
| 258 |
+
"eval_samples_per_second": 4.832,
|
| 259 |
+
"eval_wer": 0.7316814908936892,
|
| 260 |
+
"step": 3600
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 19.19,
|
| 264 |
+
"learning_rate": 0.00028175675675675674,
|
| 265 |
+
"loss": 0.1317,
|
| 266 |
+
"step": 3800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 19.19,
|
| 270 |
+
"eval_loss": 0.689399242401123,
|
| 271 |
+
"eval_runtime": 136.2101,
|
| 272 |
+
"eval_samples_per_second": 4.816,
|
| 273 |
+
"eval_wer": 0.7242693773824651,
|
| 274 |
+
"step": 3800
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 20.2,
|
| 278 |
+
"learning_rate": 0.0002807432432432432,
|
| 279 |
+
"loss": 0.1225,
|
| 280 |
+
"step": 4000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.2,
|
| 284 |
+
"eval_loss": 0.7114871144294739,
|
| 285 |
+
"eval_runtime": 135.7901,
|
| 286 |
+
"eval_samples_per_second": 4.831,
|
| 287 |
+
"eval_wer": 0.745446844557391,
|
| 288 |
+
"step": 4000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 21.21,
|
| 292 |
+
"learning_rate": 0.0002797297297297297,
|
| 293 |
+
"loss": 0.121,
|
| 294 |
+
"step": 4200
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 21.21,
|
| 298 |
+
"eval_loss": 0.7714667916297913,
|
| 299 |
+
"eval_runtime": 144.4676,
|
| 300 |
+
"eval_samples_per_second": 4.541,
|
| 301 |
+
"eval_wer": 0.7439644218551461,
|
| 302 |
+
"step": 4200
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 22.22,
|
| 306 |
+
"learning_rate": 0.0002787162162162162,
|
| 307 |
+
"loss": 0.1214,
|
| 308 |
+
"step": 4400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 22.22,
|
| 312 |
+
"eval_loss": 0.6892368197441101,
|
| 313 |
+
"eval_runtime": 141.5565,
|
| 314 |
+
"eval_samples_per_second": 4.634,
|
| 315 |
+
"eval_wer": 0.7246929267259635,
|
| 316 |
+
"step": 4400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 23.23,
|
| 320 |
+
"learning_rate": 0.00027770270270270267,
|
| 321 |
+
"loss": 0.1201,
|
| 322 |
+
"step": 4600
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 23.23,
|
| 326 |
+
"eval_loss": 0.7176805734634399,
|
| 327 |
+
"eval_runtime": 136.4081,
|
| 328 |
+
"eval_samples_per_second": 4.809,
|
| 329 |
+
"eval_wer": 0.7329521389241846,
|
| 330 |
+
"step": 4600
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.24,
|
| 334 |
+
"learning_rate": 0.00027668918918918916,
|
| 335 |
+
"loss": 0.1144,
|
| 336 |
+
"step": 4800
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 24.24,
|
| 340 |
+
"eval_loss": 0.7144946455955505,
|
| 341 |
+
"eval_runtime": 135.6564,
|
| 342 |
+
"eval_samples_per_second": 4.836,
|
| 343 |
+
"eval_wer": 0.7143159678102499,
|
| 344 |
+
"step": 4800
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 25.25,
|
| 348 |
+
"learning_rate": 0.00027567567567567564,
|
| 349 |
+
"loss": 0.1087,
|
| 350 |
+
"step": 5000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 25.25,
|
| 354 |
+
"eval_loss": 0.6824408769607544,
|
| 355 |
+
"eval_runtime": 146.0202,
|
| 356 |
+
"eval_samples_per_second": 4.493,
|
| 357 |
+
"eval_wer": 0.7551884794578568,
|
| 358 |
+
"step": 5000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 26.26,
|
| 362 |
+
"learning_rate": 0.0002746621621621621,
|
| 363 |
+
"loss": 0.1088,
|
| 364 |
+
"step": 5200
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 26.26,
|
| 368 |
+
"eval_loss": 0.7299026846885681,
|
| 369 |
+
"eval_runtime": 136.6707,
|
| 370 |
+
"eval_samples_per_second": 4.8,
|
| 371 |
+
"eval_wer": 0.7293519695044473,
|
| 372 |
+
"step": 5200
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 27.27,
|
| 376 |
+
"learning_rate": 0.0002736486486486486,
|
| 377 |
+
"loss": 0.1033,
|
| 378 |
+
"step": 5400
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 27.27,
|
| 382 |
+
"eval_loss": 0.7187995910644531,
|
| 383 |
+
"eval_runtime": 135.6478,
|
| 384 |
+
"eval_samples_per_second": 4.836,
|
| 385 |
+
"eval_wer": 0.7187632359169843,
|
| 386 |
+
"step": 5400
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 28.28,
|
| 390 |
+
"learning_rate": 0.0002726351351351351,
|
| 391 |
+
"loss": 0.1035,
|
| 392 |
+
"step": 5600
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 28.28,
|
| 396 |
+
"eval_loss": 0.7574965953826904,
|
| 397 |
+
"eval_runtime": 135.5201,
|
| 398 |
+
"eval_samples_per_second": 4.841,
|
| 399 |
+
"eval_wer": 0.743117323168149,
|
| 400 |
+
"step": 5600
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 29.29,
|
| 404 |
+
"learning_rate": 0.0002716216216216216,
|
| 405 |
+
"loss": 0.1052,
|
| 406 |
+
"step": 5800
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 29.29,
|
| 410 |
+
"eval_loss": 0.7272396683692932,
|
| 411 |
+
"eval_runtime": 135.608,
|
| 412 |
+
"eval_samples_per_second": 4.837,
|
| 413 |
+
"eval_wer": 0.7181279119017365,
|
| 414 |
+
"step": 5800
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 30.3,
|
| 418 |
+
"learning_rate": 0.00027060810810810806,
|
| 419 |
+
"loss": 0.1074,
|
| 420 |
+
"step": 6000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 30.3,
|
| 424 |
+
"eval_loss": 0.7088936567306519,
|
| 425 |
+
"eval_runtime": 135.3489,
|
| 426 |
+
"eval_samples_per_second": 4.847,
|
| 427 |
+
"eval_wer": 0.7011859381617959,
|
| 428 |
+
"step": 6000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 31.31,
|
| 432 |
+
"learning_rate": 0.0002695945945945946,
|
| 433 |
+
"loss": 0.105,
|
| 434 |
+
"step": 6200
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 31.31,
|
| 438 |
+
"eval_loss": 0.7318005561828613,
|
| 439 |
+
"eval_runtime": 136.2308,
|
| 440 |
+
"eval_samples_per_second": 4.815,
|
| 441 |
+
"eval_wer": 0.7157983905124947,
|
| 442 |
+
"step": 6200
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 32.32,
|
| 446 |
+
"learning_rate": 0.0002685810810810811,
|
| 447 |
+
"loss": 0.0937,
|
| 448 |
+
"step": 6400
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 32.32,
|
| 452 |
+
"eval_loss": 0.743655264377594,
|
| 453 |
+
"eval_runtime": 135.8626,
|
| 454 |
+
"eval_samples_per_second": 4.828,
|
| 455 |
+
"eval_wer": 0.7115628970775095,
|
| 456 |
+
"step": 6400
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 33.33,
|
| 460 |
+
"learning_rate": 0.00026756756756756756,
|
| 461 |
+
"loss": 0.0976,
|
| 462 |
+
"step": 6600
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 33.33,
|
| 466 |
+
"eval_loss": 0.7406659126281738,
|
| 467 |
+
"eval_runtime": 136.2332,
|
| 468 |
+
"eval_samples_per_second": 4.815,
|
| 469 |
+
"eval_wer": 0.7257518000847099,
|
| 470 |
+
"step": 6600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 34.34,
|
| 474 |
+
"learning_rate": 0.00026655405405405405,
|
| 475 |
+
"loss": 0.0961,
|
| 476 |
+
"step": 6800
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 34.34,
|
| 480 |
+
"eval_loss": 0.7474074959754944,
|
| 481 |
+
"eval_runtime": 136.361,
|
| 482 |
+
"eval_samples_per_second": 4.811,
|
| 483 |
+
"eval_wer": 0.729987293519695,
|
| 484 |
+
"step": 6800
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 35.35,
|
| 488 |
+
"learning_rate": 0.0002655405405405405,
|
| 489 |
+
"loss": 0.0922,
|
| 490 |
+
"step": 7000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 35.35,
|
| 494 |
+
"eval_loss": 0.7413117289543152,
|
| 495 |
+
"eval_runtime": 137.8262,
|
| 496 |
+
"eval_samples_per_second": 4.76,
|
| 497 |
+
"eval_wer": 0.7399407030919102,
|
| 498 |
+
"step": 7000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 36.36,
|
| 502 |
+
"learning_rate": 0.000264527027027027,
|
| 503 |
+
"loss": 0.0897,
|
| 504 |
+
"step": 7200
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 36.36,
|
| 508 |
+
"eval_loss": 0.7571277618408203,
|
| 509 |
+
"eval_runtime": 135.8962,
|
| 510 |
+
"eval_samples_per_second": 4.827,
|
| 511 |
+
"eval_wer": 0.7467174925878864,
|
| 512 |
+
"step": 7200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 37.37,
|
| 516 |
+
"learning_rate": 0.0002635135135135135,
|
| 517 |
+
"loss": 0.0936,
|
| 518 |
+
"step": 7400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 37.37,
|
| 522 |
+
"eval_loss": 0.7212100028991699,
|
| 523 |
+
"eval_runtime": 136.3693,
|
| 524 |
+
"eval_samples_per_second": 4.81,
|
| 525 |
+
"eval_wer": 0.7162219398559933,
|
| 526 |
+
"step": 7400
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 38.38,
|
| 530 |
+
"learning_rate": 0.0002625,
|
| 531 |
+
"loss": 0.0898,
|
| 532 |
+
"step": 7600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 38.38,
|
| 536 |
+
"eval_loss": 0.751598060131073,
|
| 537 |
+
"eval_runtime": 138.668,
|
| 538 |
+
"eval_samples_per_second": 4.731,
|
| 539 |
+
"eval_wer": 0.7210927573062261,
|
| 540 |
+
"step": 7600
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 39.39,
|
| 544 |
+
"learning_rate": 0.00026148648648648647,
|
| 545 |
+
"loss": 0.0902,
|
| 546 |
+
"step": 7800
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 39.39,
|
| 550 |
+
"eval_loss": 0.7493733167648315,
|
| 551 |
+
"eval_runtime": 136.1591,
|
| 552 |
+
"eval_samples_per_second": 4.818,
|
| 553 |
+
"eval_wer": 0.7477763659466328,
|
| 554 |
+
"step": 7800
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 40.4,
|
| 558 |
+
"learning_rate": 0.00026047297297297295,
|
| 559 |
+
"loss": 0.0847,
|
| 560 |
+
"step": 8000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 40.4,
|
| 564 |
+
"eval_loss": 0.6728424429893494,
|
| 565 |
+
"eval_runtime": 140.5502,
|
| 566 |
+
"eval_samples_per_second": 4.667,
|
| 567 |
+
"eval_wer": 0.7306226175349428,
|
| 568 |
+
"step": 8000
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 41.41,
|
| 572 |
+
"learning_rate": 0.00025945945945945944,
|
| 573 |
+
"loss": 0.0866,
|
| 574 |
+
"step": 8200
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 41.41,
|
| 578 |
+
"eval_loss": 0.749858558177948,
|
| 579 |
+
"eval_runtime": 146.7628,
|
| 580 |
+
"eval_samples_per_second": 4.47,
|
| 581 |
+
"eval_wer": 0.7151630664972469,
|
| 582 |
+
"step": 8200
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 42.42,
|
| 586 |
+
"learning_rate": 0.0002584459459459459,
|
| 587 |
+
"loss": 0.0809,
|
| 588 |
+
"step": 8400
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 42.42,
|
| 592 |
+
"eval_loss": 0.7739278078079224,
|
| 593 |
+
"eval_runtime": 136.5403,
|
| 594 |
+
"eval_samples_per_second": 4.804,
|
| 595 |
+
"eval_wer": 0.7429055484963998,
|
| 596 |
+
"step": 8400
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 43.43,
|
| 600 |
+
"learning_rate": 0.0002574324324324324,
|
| 601 |
+
"loss": 0.0847,
|
| 602 |
+
"step": 8600
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 43.43,
|
| 606 |
+
"eval_loss": 0.7759637236595154,
|
| 607 |
+
"eval_runtime": 136.7555,
|
| 608 |
+
"eval_samples_per_second": 4.797,
|
| 609 |
+
"eval_wer": 0.7244811520542143,
|
| 610 |
+
"step": 8600
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 44.44,
|
| 614 |
+
"learning_rate": 0.0002564189189189189,
|
| 615 |
+
"loss": 0.0852,
|
| 616 |
+
"step": 8800
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 44.44,
|
| 620 |
+
"eval_loss": 0.7829259634017944,
|
| 621 |
+
"eval_runtime": 139.6603,
|
| 622 |
+
"eval_samples_per_second": 4.697,
|
| 623 |
+
"eval_wer": 0.7227869546802203,
|
| 624 |
+
"step": 8800
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 45.45,
|
| 628 |
+
"learning_rate": 0.00025540540540540537,
|
| 629 |
+
"loss": 0.0802,
|
| 630 |
+
"step": 9000
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 45.45,
|
| 634 |
+
"eval_loss": 0.7056646347045898,
|
| 635 |
+
"eval_runtime": 135.8647,
|
| 636 |
+
"eval_samples_per_second": 4.828,
|
| 637 |
+
"eval_wer": 0.7100804743752647,
|
| 638 |
+
"step": 9000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 46.46,
|
| 642 |
+
"learning_rate": 0.0002543918918918919,
|
| 643 |
+
"loss": 0.0794,
|
| 644 |
+
"step": 9200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 46.46,
|
| 648 |
+
"eval_loss": 0.7538515329360962,
|
| 649 |
+
"eval_runtime": 140.8763,
|
| 650 |
+
"eval_samples_per_second": 4.657,
|
| 651 |
+
"eval_wer": 0.7098686997035154,
|
| 652 |
+
"step": 9200
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 47.47,
|
| 656 |
+
"learning_rate": 0.0002533783783783784,
|
| 657 |
+
"loss": 0.0827,
|
| 658 |
+
"step": 9400
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 47.47,
|
| 662 |
+
"eval_loss": 0.7439035773277283,
|
| 663 |
+
"eval_runtime": 143.0424,
|
| 664 |
+
"eval_samples_per_second": 4.586,
|
| 665 |
+
"eval_wer": 0.7151630664972469,
|
| 666 |
+
"step": 9400
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 48.48,
|
| 670 |
+
"learning_rate": 0.0002523648648648648,
|
| 671 |
+
"loss": 0.0803,
|
| 672 |
+
"step": 9600
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 48.48,
|
| 676 |
+
"eval_loss": 0.7053723335266113,
|
| 677 |
+
"eval_runtime": 143.445,
|
| 678 |
+
"eval_samples_per_second": 4.573,
|
| 679 |
+
"eval_wer": 0.7075391783142736,
|
| 680 |
+
"step": 9600
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 49.49,
|
| 684 |
+
"learning_rate": 0.0002513513513513513,
|
| 685 |
+
"loss": 0.0781,
|
| 686 |
+
"step": 9800
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 49.49,
|
| 690 |
+
"eval_loss": 0.8703842163085938,
|
| 691 |
+
"eval_runtime": 143.62,
|
| 692 |
+
"eval_samples_per_second": 4.568,
|
| 693 |
+
"eval_wer": 0.7630241423125794,
|
| 694 |
+
"step": 9800
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 50.5,
|
| 698 |
+
"learning_rate": 0.0002503378378378378,
|
| 699 |
+
"loss": 0.0766,
|
| 700 |
+
"step": 10000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 50.5,
|
| 704 |
+
"eval_loss": 0.7902593612670898,
|
| 705 |
+
"eval_runtime": 139.0704,
|
| 706 |
+
"eval_samples_per_second": 4.717,
|
| 707 |
+
"eval_wer": 0.7130453197797544,
|
| 708 |
+
"step": 10000
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 51.51,
|
| 712 |
+
"learning_rate": 0.0002493243243243243,
|
| 713 |
+
"loss": 0.0771,
|
| 714 |
+
"step": 10200
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 51.51,
|
| 718 |
+
"eval_loss": 0.7181321978569031,
|
| 719 |
+
"eval_runtime": 138.4984,
|
| 720 |
+
"eval_samples_per_second": 4.737,
|
| 721 |
+
"eval_wer": 0.7149512918254977,
|
| 722 |
+
"step": 10200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 52.52,
|
| 726 |
+
"learning_rate": 0.0002483108108108108,
|
| 727 |
+
"loss": 0.0713,
|
| 728 |
+
"step": 10400
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 52.52,
|
| 732 |
+
"eval_loss": 0.7941905856132507,
|
| 733 |
+
"eval_runtime": 137.9511,
|
| 734 |
+
"eval_samples_per_second": 4.755,
|
| 735 |
+
"eval_wer": 0.7164337145277425,
|
| 736 |
+
"step": 10400
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 53.53,
|
| 740 |
+
"learning_rate": 0.0002472972972972973,
|
| 741 |
+
"loss": 0.0761,
|
| 742 |
+
"step": 10600
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 53.53,
|
| 746 |
+
"eval_loss": 0.8347806334495544,
|
| 747 |
+
"eval_runtime": 137.6637,
|
| 748 |
+
"eval_samples_per_second": 4.765,
|
| 749 |
+
"eval_wer": 0.7100804743752647,
|
| 750 |
+
"step": 10600
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 54.54,
|
| 754 |
+
"learning_rate": 0.0002462837837837838,
|
| 755 |
+
"loss": 0.0721,
|
| 756 |
+
"step": 10800
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 54.54,
|
| 760 |
+
"eval_loss": 0.7271831035614014,
|
| 761 |
+
"eval_runtime": 137.7418,
|
| 762 |
+
"eval_samples_per_second": 4.763,
|
| 763 |
+
"eval_wer": 0.7024565861922915,
|
| 764 |
+
"step": 10800
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 55.55,
|
| 768 |
+
"learning_rate": 0.00024527027027027026,
|
| 769 |
+
"loss": 0.0684,
|
| 770 |
+
"step": 11000
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 55.55,
|
| 774 |
+
"eval_loss": 0.7921620011329651,
|
| 775 |
+
"eval_runtime": 137.9904,
|
| 776 |
+
"eval_samples_per_second": 4.754,
|
| 777 |
+
"eval_wer": 0.6977975434138077,
|
| 778 |
+
"step": 11000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 56.56,
|
| 782 |
+
"learning_rate": 0.00024425675675675675,
|
| 783 |
+
"loss": 0.0733,
|
| 784 |
+
"step": 11200
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 56.56,
|
| 788 |
+
"eval_loss": 0.782407820224762,
|
| 789 |
+
"eval_runtime": 138.0874,
|
| 790 |
+
"eval_samples_per_second": 4.751,
|
| 791 |
+
"eval_wer": 0.7052096569250318,
|
| 792 |
+
"step": 11200
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 57.57,
|
| 796 |
+
"learning_rate": 0.00024324324324324323,
|
| 797 |
+
"loss": 0.0723,
|
| 798 |
+
"step": 11400
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 57.57,
|
| 802 |
+
"eval_loss": 0.7807161211967468,
|
| 803 |
+
"eval_runtime": 137.5671,
|
| 804 |
+
"eval_samples_per_second": 4.769,
|
| 805 |
+
"eval_wer": 0.6994917407878017,
|
| 806 |
+
"step": 11400
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 58.58,
|
| 810 |
+
"learning_rate": 0.00024222972972972971,
|
| 811 |
+
"loss": 0.0701,
|
| 812 |
+
"step": 11600
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 58.58,
|
| 816 |
+
"eval_loss": 0.7515387535095215,
|
| 817 |
+
"eval_runtime": 138.0585,
|
| 818 |
+
"eval_samples_per_second": 4.752,
|
| 819 |
+
"eval_wer": 0.7085980516730199,
|
| 820 |
+
"step": 11600
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 59.59,
|
| 824 |
+
"learning_rate": 0.0002412162162162162,
|
| 825 |
+
"loss": 0.0687,
|
| 826 |
+
"step": 11800
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 59.59,
|
| 830 |
+
"eval_loss": 0.7535402774810791,
|
| 831 |
+
"eval_runtime": 137.596,
|
| 832 |
+
"eval_samples_per_second": 4.768,
|
| 833 |
+
"eval_wer": 0.688267683185091,
|
| 834 |
+
"step": 11800
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 60.6,
|
| 838 |
+
"learning_rate": 0.00024020270270270268,
|
| 839 |
+
"loss": 0.0676,
|
| 840 |
+
"step": 12000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 60.6,
|
| 844 |
+
"eval_loss": 0.7971649169921875,
|
| 845 |
+
"eval_runtime": 138.6186,
|
| 846 |
+
"eval_samples_per_second": 4.732,
|
| 847 |
+
"eval_wer": 0.7132570944515036,
|
| 848 |
+
"step": 12000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 61.61,
|
| 852 |
+
"learning_rate": 0.00023918918918918917,
|
| 853 |
+
"loss": 0.0733,
|
| 854 |
+
"step": 12200
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 61.61,
|
| 858 |
+
"eval_loss": 0.7672501802444458,
|
| 859 |
+
"eval_runtime": 138.3057,
|
| 860 |
+
"eval_samples_per_second": 4.743,
|
| 861 |
+
"eval_wer": 0.711139347734011,
|
| 862 |
+
"step": 12200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 62.62,
|
| 866 |
+
"learning_rate": 0.00023817567567567568,
|
| 867 |
+
"loss": 0.0661,
|
| 868 |
+
"step": 12400
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 62.62,
|
| 872 |
+
"eval_loss": 0.7474766373634338,
|
| 873 |
+
"eval_runtime": 138.3209,
|
| 874 |
+
"eval_samples_per_second": 4.743,
|
| 875 |
+
"eval_wer": 0.6975857687420585,
|
| 876 |
+
"step": 12400
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 63.63,
|
| 880 |
+
"learning_rate": 0.00023716216216216216,
|
| 881 |
+
"loss": 0.0666,
|
| 882 |
+
"step": 12600
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 63.63,
|
| 886 |
+
"eval_loss": 0.7444884181022644,
|
| 887 |
+
"eval_runtime": 137.4553,
|
| 888 |
+
"eval_samples_per_second": 4.772,
|
| 889 |
+
"eval_wer": 0.7083862770012707,
|
| 890 |
+
"step": 12600
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 64.64,
|
| 894 |
+
"learning_rate": 0.00023614864864864862,
|
| 895 |
+
"loss": 0.0712,
|
| 896 |
+
"step": 12800
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 64.64,
|
| 900 |
+
"eval_loss": 0.7816994190216064,
|
| 901 |
+
"eval_runtime": 138.1127,
|
| 902 |
+
"eval_samples_per_second": 4.75,
|
| 903 |
+
"eval_wer": 0.7210927573062261,
|
| 904 |
+
"step": 12800
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 65.65,
|
| 908 |
+
"learning_rate": 0.0002351351351351351,
|
| 909 |
+
"loss": 0.063,
|
| 910 |
+
"step": 13000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 65.65,
|
| 914 |
+
"eval_loss": 0.8192810416221619,
|
| 915 |
+
"eval_runtime": 137.4083,
|
| 916 |
+
"eval_samples_per_second": 4.774,
|
| 917 |
+
"eval_wer": 0.7092333756882677,
|
| 918 |
+
"step": 13000
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 66.66,
|
| 922 |
+
"learning_rate": 0.00023412162162162159,
|
| 923 |
+
"loss": 0.0681,
|
| 924 |
+
"step": 13200
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 66.66,
|
| 928 |
+
"eval_loss": 0.7333322763442993,
|
| 929 |
+
"eval_runtime": 137.9838,
|
| 930 |
+
"eval_samples_per_second": 4.754,
|
| 931 |
+
"eval_wer": 0.698009318085557,
|
| 932 |
+
"step": 13200
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 67.68,
|
| 936 |
+
"learning_rate": 0.00023310810810810807,
|
| 937 |
+
"loss": 0.0642,
|
| 938 |
+
"step": 13400
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 67.68,
|
| 942 |
+
"eval_loss": 0.7417888045310974,
|
| 943 |
+
"eval_runtime": 137.6106,
|
| 944 |
+
"eval_samples_per_second": 4.767,
|
| 945 |
+
"eval_wer": 0.7083862770012707,
|
| 946 |
+
"step": 13400
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 68.69,
|
| 950 |
+
"learning_rate": 0.00023209459459459458,
|
| 951 |
+
"loss": 0.0622,
|
| 952 |
+
"step": 13600
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 68.69,
|
| 956 |
+
"eval_loss": 0.7969706654548645,
|
| 957 |
+
"eval_runtime": 137.6918,
|
| 958 |
+
"eval_samples_per_second": 4.764,
|
| 959 |
+
"eval_wer": 0.7329521389241846,
|
| 960 |
+
"step": 13600
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 69.7,
|
| 964 |
+
"learning_rate": 0.00023108108108108106,
|
| 965 |
+
"loss": 0.0639,
|
| 966 |
+
"step": 13800
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 69.7,
|
| 970 |
+
"eval_loss": 0.8013876080513,
|
| 971 |
+
"eval_runtime": 137.5846,
|
| 972 |
+
"eval_samples_per_second": 4.768,
|
| 973 |
+
"eval_wer": 0.6988564167725541,
|
| 974 |
+
"step": 13800
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 70.71,
|
| 978 |
+
"learning_rate": 0.00023006756756756755,
|
| 979 |
+
"loss": 0.065,
|
| 980 |
+
"step": 14000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 70.71,
|
| 984 |
+
"eval_loss": 0.7792071104049683,
|
| 985 |
+
"eval_runtime": 137.6266,
|
| 986 |
+
"eval_samples_per_second": 4.767,
|
| 987 |
+
"eval_wer": 0.7043625582380347,
|
| 988 |
+
"step": 14000
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 71.72,
|
| 992 |
+
"learning_rate": 0.00022905405405405403,
|
| 993 |
+
"loss": 0.0608,
|
| 994 |
+
"step": 14200
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 71.72,
|
| 998 |
+
"eval_loss": 0.7951736450195312,
|
| 999 |
+
"eval_runtime": 138.4177,
|
| 1000 |
+
"eval_samples_per_second": 4.739,
|
| 1001 |
+
"eval_wer": 0.6944091486658196,
|
| 1002 |
+
"step": 14200
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 72.73,
|
| 1006 |
+
"learning_rate": 0.00022804054054054052,
|
| 1007 |
+
"loss": 0.0557,
|
| 1008 |
+
"step": 14400
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 72.73,
|
| 1012 |
+
"eval_loss": 0.8266920447349548,
|
| 1013 |
+
"eval_runtime": 138.5687,
|
| 1014 |
+
"eval_samples_per_second": 4.734,
|
| 1015 |
+
"eval_wer": 0.6977975434138077,
|
| 1016 |
+
"step": 14400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 73.74,
|
| 1020 |
+
"learning_rate": 0.00022702702702702703,
|
| 1021 |
+
"loss": 0.0592,
|
| 1022 |
+
"step": 14600
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 73.74,
|
| 1026 |
+
"eval_loss": 0.8056704998016357,
|
| 1027 |
+
"eval_runtime": 137.1892,
|
| 1028 |
+
"eval_samples_per_second": 4.782,
|
| 1029 |
+
"eval_wer": 0.7075391783142736,
|
| 1030 |
+
"step": 14600
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 74.75,
|
| 1034 |
+
"learning_rate": 0.0002260135135135135,
|
| 1035 |
+
"loss": 0.064,
|
| 1036 |
+
"step": 14800
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 74.75,
|
| 1040 |
+
"eval_loss": 0.815148115158081,
|
| 1041 |
+
"eval_runtime": 137.9946,
|
| 1042 |
+
"eval_samples_per_second": 4.754,
|
| 1043 |
+
"eval_wer": 0.7009741634900466,
|
| 1044 |
+
"step": 14800
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 75.76,
|
| 1048 |
+
"learning_rate": 0.000225,
|
| 1049 |
+
"loss": 0.0603,
|
| 1050 |
+
"step": 15000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 75.76,
|
| 1054 |
+
"eval_loss": 0.7697699069976807,
|
| 1055 |
+
"eval_runtime": 138.1754,
|
| 1056 |
+
"eval_samples_per_second": 4.748,
|
| 1057 |
+
"eval_wer": 0.7124099957645066,
|
| 1058 |
+
"step": 15000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 76.77,
|
| 1062 |
+
"learning_rate": 0.00022398648648648645,
|
| 1063 |
+
"loss": 0.0613,
|
| 1064 |
+
"step": 15200
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 76.77,
|
| 1068 |
+
"eval_loss": 0.7657320499420166,
|
| 1069 |
+
"eval_runtime": 137.8551,
|
| 1070 |
+
"eval_samples_per_second": 4.759,
|
| 1071 |
+
"eval_wer": 0.7170690385429903,
|
| 1072 |
+
"step": 15200
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 77.78,
|
| 1076 |
+
"learning_rate": 0.00022297297297297293,
|
| 1077 |
+
"loss": 0.0602,
|
| 1078 |
+
"step": 15400
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 77.78,
|
| 1082 |
+
"eval_loss": 0.7715324759483337,
|
| 1083 |
+
"eval_runtime": 137.9821,
|
| 1084 |
+
"eval_samples_per_second": 4.754,
|
| 1085 |
+
"eval_wer": 0.716857263871241,
|
| 1086 |
+
"step": 15400
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 78.79,
|
| 1090 |
+
"learning_rate": 0.00022195945945945945,
|
| 1091 |
+
"loss": 0.0597,
|
| 1092 |
+
"step": 15600
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 78.79,
|
| 1096 |
+
"eval_loss": 0.7833596467971802,
|
| 1097 |
+
"eval_runtime": 137.7858,
|
| 1098 |
+
"eval_samples_per_second": 4.761,
|
| 1099 |
+
"eval_wer": 0.7132570944515036,
|
| 1100 |
+
"step": 15600
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 79.8,
|
| 1104 |
+
"learning_rate": 0.00022094594594594593,
|
| 1105 |
+
"loss": 0.0585,
|
| 1106 |
+
"step": 15800
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 79.8,
|
| 1110 |
+
"eval_loss": 0.8398387432098389,
|
| 1111 |
+
"eval_runtime": 137.7412,
|
| 1112 |
+
"eval_samples_per_second": 4.763,
|
| 1113 |
+
"eval_wer": 0.7088098263447692,
|
| 1114 |
+
"step": 15800
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 80.81,
|
| 1118 |
+
"learning_rate": 0.0002199324324324324,
|
| 1119 |
+
"loss": 0.056,
|
| 1120 |
+
"step": 16000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 80.81,
|
| 1124 |
+
"eval_loss": 0.8182487487792969,
|
| 1125 |
+
"eval_runtime": 137.8628,
|
| 1126 |
+
"eval_samples_per_second": 4.758,
|
| 1127 |
+
"eval_wer": 0.7128335451080051,
|
| 1128 |
+
"step": 16000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 81.82,
|
| 1132 |
+
"learning_rate": 0.0002189189189189189,
|
| 1133 |
+
"loss": 0.06,
|
| 1134 |
+
"step": 16200
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 81.82,
|
| 1138 |
+
"eval_loss": 0.8023802638053894,
|
| 1139 |
+
"eval_runtime": 137.792,
|
| 1140 |
+
"eval_samples_per_second": 4.761,
|
| 1141 |
+
"eval_wer": 0.6918678526048284,
|
| 1142 |
+
"step": 16200
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 82.83,
|
| 1146 |
+
"learning_rate": 0.00021790540540540538,
|
| 1147 |
+
"loss": 0.0542,
|
| 1148 |
+
"step": 16400
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 82.83,
|
| 1152 |
+
"eval_loss": 0.7613797187805176,
|
| 1153 |
+
"eval_runtime": 138.0605,
|
| 1154 |
+
"eval_samples_per_second": 4.752,
|
| 1155 |
+
"eval_wer": 0.6916560779330791,
|
| 1156 |
+
"step": 16400
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 83.84,
|
| 1160 |
+
"learning_rate": 0.0002168918918918919,
|
| 1161 |
+
"loss": 0.0565,
|
| 1162 |
+
"step": 16600
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 83.84,
|
| 1166 |
+
"eval_loss": 0.7780935764312744,
|
| 1167 |
+
"eval_runtime": 138.2761,
|
| 1168 |
+
"eval_samples_per_second": 4.744,
|
| 1169 |
+
"eval_wer": 0.7153748411689962,
|
| 1170 |
+
"step": 16600
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 84.85,
|
| 1174 |
+
"learning_rate": 0.00021587837837837838,
|
| 1175 |
+
"loss": 0.0518,
|
| 1176 |
+
"step": 16800
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 84.85,
|
| 1180 |
+
"eval_loss": 0.8466140031814575,
|
| 1181 |
+
"eval_runtime": 138.8619,
|
| 1182 |
+
"eval_samples_per_second": 4.724,
|
| 1183 |
+
"eval_wer": 0.7475645912748835,
|
| 1184 |
+
"step": 16800
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 85.86,
|
| 1188 |
+
"learning_rate": 0.00021486486486486486,
|
| 1189 |
+
"loss": 0.0566,
|
| 1190 |
+
"step": 17000
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 85.86,
|
| 1194 |
+
"eval_loss": 0.8004465699195862,
|
| 1195 |
+
"eval_runtime": 137.4262,
|
| 1196 |
+
"eval_samples_per_second": 4.773,
|
| 1197 |
+
"eval_wer": 0.7064803049555273,
|
| 1198 |
+
"step": 17000
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 86.87,
|
| 1202 |
+
"learning_rate": 0.00021385135135135134,
|
| 1203 |
+
"loss": 0.0535,
|
| 1204 |
+
"step": 17200
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 86.87,
|
| 1208 |
+
"eval_loss": 0.8326082825660706,
|
| 1209 |
+
"eval_runtime": 137.6615,
|
| 1210 |
+
"eval_samples_per_second": 4.765,
|
| 1211 |
+
"eval_wer": 0.7043625582380347,
|
| 1212 |
+
"step": 17200
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 87.88,
|
| 1216 |
+
"learning_rate": 0.0002128378378378378,
|
| 1217 |
+
"loss": 0.0483,
|
| 1218 |
+
"step": 17400
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 87.88,
|
| 1222 |
+
"eval_loss": 0.7926554083824158,
|
| 1223 |
+
"eval_runtime": 138.0063,
|
| 1224 |
+
"eval_samples_per_second": 4.753,
|
| 1225 |
+
"eval_wer": 0.7251164760694621,
|
| 1226 |
+
"step": 17400
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 88.89,
|
| 1230 |
+
"learning_rate": 0.00021182432432432428,
|
| 1231 |
+
"loss": 0.0486,
|
| 1232 |
+
"step": 17600
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 88.89,
|
| 1236 |
+
"eval_loss": 0.7882806062698364,
|
| 1237 |
+
"eval_runtime": 138.7006,
|
| 1238 |
+
"eval_samples_per_second": 4.73,
|
| 1239 |
+
"eval_wer": 0.6899618805590851,
|
| 1240 |
+
"step": 17600
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 89.9,
|
| 1244 |
+
"learning_rate": 0.0002108108108108108,
|
| 1245 |
+
"loss": 0.0504,
|
| 1246 |
+
"step": 17800
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 89.9,
|
| 1250 |
+
"eval_loss": 0.7685484290122986,
|
| 1251 |
+
"eval_runtime": 138.3804,
|
| 1252 |
+
"eval_samples_per_second": 4.741,
|
| 1253 |
+
"eval_wer": 0.6939855993223211,
|
| 1254 |
+
"step": 17800
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 90.91,
|
| 1258 |
+
"learning_rate": 0.00020979729729729728,
|
| 1259 |
+
"loss": 0.056,
|
| 1260 |
+
"step": 18000
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 90.91,
|
| 1264 |
+
"eval_loss": 0.848064124584198,
|
| 1265 |
+
"eval_runtime": 137.1191,
|
| 1266 |
+
"eval_samples_per_second": 4.784,
|
| 1267 |
+
"eval_wer": 0.7007623888182973,
|
| 1268 |
+
"step": 18000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 91.92,
|
| 1272 |
+
"learning_rate": 0.00020878378378378376,
|
| 1273 |
+
"loss": 0.049,
|
| 1274 |
+
"step": 18200
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 91.92,
|
| 1278 |
+
"eval_loss": 0.845516562461853,
|
| 1279 |
+
"eval_runtime": 137.1172,
|
| 1280 |
+
"eval_samples_per_second": 4.784,
|
| 1281 |
+
"eval_wer": 0.7119864464210081,
|
| 1282 |
+
"step": 18200
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 92.93,
|
| 1286 |
+
"learning_rate": 0.00020777027027027025,
|
| 1287 |
+
"loss": 0.0536,
|
| 1288 |
+
"step": 18400
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 92.93,
|
| 1292 |
+
"eval_loss": 0.7710561752319336,
|
| 1293 |
+
"eval_runtime": 138.0433,
|
| 1294 |
+
"eval_samples_per_second": 4.752,
|
| 1295 |
+
"eval_wer": 0.6925031766200762,
|
| 1296 |
+
"step": 18400
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 93.94,
|
| 1300 |
+
"learning_rate": 0.00020675675675675673,
|
| 1301 |
+
"loss": 0.0503,
|
| 1302 |
+
"step": 18600
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 93.94,
|
| 1306 |
+
"eval_loss": 0.8255069851875305,
|
| 1307 |
+
"eval_runtime": 137.6117,
|
| 1308 |
+
"eval_samples_per_second": 4.767,
|
| 1309 |
+
"eval_wer": 0.7174925878864887,
|
| 1310 |
+
"step": 18600
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 94.95,
|
| 1314 |
+
"learning_rate": 0.00020574324324324324,
|
| 1315 |
+
"loss": 0.0488,
|
| 1316 |
+
"step": 18800
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 94.95,
|
| 1320 |
+
"eval_loss": 0.8600437641143799,
|
| 1321 |
+
"eval_runtime": 137.4326,
|
| 1322 |
+
"eval_samples_per_second": 4.773,
|
| 1323 |
+
"eval_wer": 0.7246929267259635,
|
| 1324 |
+
"step": 18800
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 95.96,
|
| 1328 |
+
"learning_rate": 0.00020472972972972972,
|
| 1329 |
+
"loss": 0.0468,
|
| 1330 |
+
"step": 19000
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 95.96,
|
| 1334 |
+
"eval_loss": 0.7781614661216736,
|
| 1335 |
+
"eval_runtime": 137.8454,
|
| 1336 |
+
"eval_samples_per_second": 4.759,
|
| 1337 |
+
"eval_wer": 0.7001270648030495,
|
| 1338 |
+
"step": 19000
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 96.97,
|
| 1342 |
+
"learning_rate": 0.0002037162162162162,
|
| 1343 |
+
"loss": 0.0486,
|
| 1344 |
+
"step": 19200
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 96.97,
|
| 1348 |
+
"eval_loss": 0.7687584161758423,
|
| 1349 |
+
"eval_runtime": 137.2673,
|
| 1350 |
+
"eval_samples_per_second": 4.779,
|
| 1351 |
+
"eval_wer": 0.7054214315967811,
|
| 1352 |
+
"step": 19200
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 97.98,
|
| 1356 |
+
"learning_rate": 0.0002027027027027027,
|
| 1357 |
+
"loss": 0.0523,
|
| 1358 |
+
"step": 19400
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 97.98,
|
| 1362 |
+
"eval_loss": 0.8229833245277405,
|
| 1363 |
+
"eval_runtime": 137.1013,
|
| 1364 |
+
"eval_samples_per_second": 4.785,
|
| 1365 |
+
"eval_wer": 0.7066920796272765,
|
| 1366 |
+
"step": 19400
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 98.99,
|
| 1370 |
+
"learning_rate": 0.00020168918918918915,
|
| 1371 |
+
"loss": 0.0507,
|
| 1372 |
+
"step": 19600
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 98.99,
|
| 1376 |
+
"eval_loss": 0.761595606803894,
|
| 1377 |
+
"eval_runtime": 137.3405,
|
| 1378 |
+
"eval_samples_per_second": 4.776,
|
| 1379 |
+
"eval_wer": 0.6910207539178315,
|
| 1380 |
+
"step": 19600
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 100.0,
|
| 1384 |
+
"learning_rate": 0.00020067567567567566,
|
| 1385 |
+
"loss": 0.049,
|
| 1386 |
+
"step": 19800
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 100.0,
|
| 1390 |
+
"eval_loss": 0.8465907573699951,
|
| 1391 |
+
"eval_runtime": 136.8552,
|
| 1392 |
+
"eval_samples_per_second": 4.793,
|
| 1393 |
+
"eval_wer": 0.704574332909784,
|
| 1394 |
+
"step": 19800
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 101.01,
|
| 1398 |
+
"learning_rate": 0.00019966216216216214,
|
| 1399 |
+
"loss": 0.0502,
|
| 1400 |
+
"step": 20000
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 101.01,
|
| 1404 |
+
"eval_loss": 0.8005170226097107,
|
| 1405 |
+
"eval_runtime": 137.3151,
|
| 1406 |
+
"eval_samples_per_second": 4.777,
|
| 1407 |
+
"eval_wer": 0.6925031766200762,
|
| 1408 |
+
"step": 20000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 102.02,
|
| 1412 |
+
"learning_rate": 0.00019864864864864863,
|
| 1413 |
+
"loss": 0.0443,
|
| 1414 |
+
"step": 20200
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 102.02,
|
| 1418 |
+
"eval_loss": 0.8749315142631531,
|
| 1419 |
+
"eval_runtime": 137.9488,
|
| 1420 |
+
"eval_samples_per_second": 4.755,
|
| 1421 |
+
"eval_wer": 0.6927149512918255,
|
| 1422 |
+
"step": 20200
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 103.03,
|
| 1426 |
+
"learning_rate": 0.0001976351351351351,
|
| 1427 |
+
"loss": 0.0486,
|
| 1428 |
+
"step": 20400
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 103.03,
|
| 1432 |
+
"eval_loss": 0.7846581339836121,
|
| 1433 |
+
"eval_runtime": 136.7448,
|
| 1434 |
+
"eval_samples_per_second": 4.797,
|
| 1435 |
+
"eval_wer": 0.7124099957645066,
|
| 1436 |
+
"step": 20400
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 104.04,
|
| 1440 |
+
"learning_rate": 0.0001966216216216216,
|
| 1441 |
+
"loss": 0.0455,
|
| 1442 |
+
"step": 20600
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 104.04,
|
| 1446 |
+
"eval_loss": 0.842466413974762,
|
| 1447 |
+
"eval_runtime": 137.1975,
|
| 1448 |
+
"eval_samples_per_second": 4.781,
|
| 1449 |
+
"eval_wer": 0.7189750105887336,
|
| 1450 |
+
"step": 20600
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 105.05,
|
| 1454 |
+
"learning_rate": 0.00019560810810810808,
|
| 1455 |
+
"loss": 0.0435,
|
| 1456 |
+
"step": 20800
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 105.05,
|
| 1460 |
+
"eval_loss": 0.8235080242156982,
|
| 1461 |
+
"eval_runtime": 137.1509,
|
| 1462 |
+
"eval_samples_per_second": 4.783,
|
| 1463 |
+
"eval_wer": 0.6927149512918255,
|
| 1464 |
+
"step": 20800
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 106.06,
|
| 1468 |
+
"learning_rate": 0.0001945945945945946,
|
| 1469 |
+
"loss": 0.0452,
|
| 1470 |
+
"step": 21000
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 106.06,
|
| 1474 |
+
"eval_loss": 0.7949621677398682,
|
| 1475 |
+
"eval_runtime": 137.8819,
|
| 1476 |
+
"eval_samples_per_second": 4.758,
|
| 1477 |
+
"eval_wer": 0.6891147818720881,
|
| 1478 |
+
"step": 21000
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 107.07,
|
| 1482 |
+
"learning_rate": 0.00019358108108108107,
|
| 1483 |
+
"loss": 0.0441,
|
| 1484 |
+
"step": 21200
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 107.07,
|
| 1488 |
+
"eval_loss": 0.8163775205612183,
|
| 1489 |
+
"eval_runtime": 136.9828,
|
| 1490 |
+
"eval_samples_per_second": 4.789,
|
| 1491 |
+
"eval_wer": 0.6992799661160525,
|
| 1492 |
+
"step": 21200
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 108.08,
|
| 1496 |
+
"learning_rate": 0.00019256756756756756,
|
| 1497 |
+
"loss": 0.0433,
|
| 1498 |
+
"step": 21400
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 108.08,
|
| 1502 |
+
"eval_loss": 0.9103622436523438,
|
| 1503 |
+
"eval_runtime": 138.0294,
|
| 1504 |
+
"eval_samples_per_second": 4.753,
|
| 1505 |
+
"eval_wer": 0.7009741634900466,
|
| 1506 |
+
"step": 21400
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 109.09,
|
| 1510 |
+
"learning_rate": 0.00019155405405405404,
|
| 1511 |
+
"loss": 0.0423,
|
| 1512 |
+
"step": 21600
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 109.09,
|
| 1516 |
+
"eval_loss": 0.8365850448608398,
|
| 1517 |
+
"eval_runtime": 137.9595,
|
| 1518 |
+
"eval_samples_per_second": 4.755,
|
| 1519 |
+
"eval_wer": 0.6954680220245658,
|
| 1520 |
+
"step": 21600
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 110.1,
|
| 1524 |
+
"learning_rate": 0.0001905405405405405,
|
| 1525 |
+
"loss": 0.0428,
|
| 1526 |
+
"step": 21800
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 110.1,
|
| 1530 |
+
"eval_loss": 0.9046308994293213,
|
| 1531 |
+
"eval_runtime": 137.2986,
|
| 1532 |
+
"eval_samples_per_second": 4.778,
|
| 1533 |
+
"eval_wer": 0.6863617111393477,
|
| 1534 |
+
"step": 21800
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 111.11,
|
| 1538 |
+
"learning_rate": 0.00018952702702702704,
|
| 1539 |
+
"loss": 0.0389,
|
| 1540 |
+
"step": 22000
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 111.11,
|
| 1544 |
+
"eval_loss": 0.8430789709091187,
|
| 1545 |
+
"eval_runtime": 137.4909,
|
| 1546 |
+
"eval_samples_per_second": 4.771,
|
| 1547 |
+
"eval_wer": 0.6869970351545955,
|
| 1548 |
+
"step": 22000
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 112.12,
|
| 1552 |
+
"learning_rate": 0.0001885135135135135,
|
| 1553 |
+
"loss": 0.0388,
|
| 1554 |
+
"step": 22200
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 112.12,
|
| 1558 |
+
"eval_loss": 0.8418474793434143,
|
| 1559 |
+
"eval_runtime": 138.0347,
|
| 1560 |
+
"eval_samples_per_second": 4.752,
|
| 1561 |
+
"eval_wer": 0.6990681914443032,
|
| 1562 |
+
"step": 22200
|
| 1563 |
+
}
|
| 1564 |
+
],
|
| 1565 |
+
"max_steps": 59400,
|
| 1566 |
+
"num_train_epochs": 300,
|
| 1567 |
+
"total_flos": 3.071129167748675e+19,
|
| 1568 |
+
"trial_name": null,
|
| 1569 |
+
"trial_params": null
|
| 1570 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:911e066f501a17bc00de61c981dcf019d6204340e764be0f0b60a7c80bf67155
|
| 3 |
+
size 2287
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"\u201e": 0, "\u10d7": 1, "\u10e7": 2, "\u10ed": 3, "\u10d5": 4, "\u10dd": 5, "\u10d9": 6, "\u10d1": 7, "\u10de": 8, "\u10e6": 9, "\u10e5": 10, "\u10d0": 11, "\u10e2": 12, "\u10d6": 13, "\u10ec": 14, "\u10e4": 15, "\u10e0": 16, "\u10da": 17, "\u10ef": 18, "\u10ea": 19, "\u10e1": 20, "\u10e3": 21, "\u10f0": 22, "\u10d2": 23, "\u10dc": 24, "\u10eb": 25, "\u10e9": 26, "\u2013": 27, "\u10ee": 28, "\u10d8": 29, "\u10db": 30, "\u10d3": 31, "\u10e8": 32, "\u10d4": 33, "\u10df": 34, "|": 35, "[UNK]": 36, "[PAD]": 37}
|