upload files
Browse files- config.json +69 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +1716 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "elgeish/wav2vec2-large-lv60-timit-asr",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 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": "sum",
|
| 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": 0.1,
|
| 46 |
+
"final_dropout": 0.1,
|
| 47 |
+
"gradient_checkpointing": false,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_dropout_prob": 0.1,
|
| 51 |
+
"hidden_size": 1024,
|
| 52 |
+
"initializer_range": 0.02,
|
| 53 |
+
"intermediate_size": 4096,
|
| 54 |
+
"layer_norm_eps": 1e-05,
|
| 55 |
+
"layerdrop": 0.1,
|
| 56 |
+
"mask_feature_length": 10,
|
| 57 |
+
"mask_feature_prob": 0.0,
|
| 58 |
+
"mask_time_length": 10,
|
| 59 |
+
"mask_time_prob": 0.05,
|
| 60 |
+
"model_type": "wav2vec2",
|
| 61 |
+
"num_attention_heads": 16,
|
| 62 |
+
"num_conv_pos_embedding_groups": 16,
|
| 63 |
+
"num_conv_pos_embeddings": 128,
|
| 64 |
+
"num_feat_extract_layers": 7,
|
| 65 |
+
"num_hidden_layers": 24,
|
| 66 |
+
"pad_token_id": 0,
|
| 67 |
+
"transformers_version": "4.4.0.dev0",
|
| 68 |
+
"vocab_size": 32
|
| 69 |
+
}
|
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 sha256:24d33462a32e457df0bc7dbafddeec392f21c0c15eb61181f7fb38c7360ea1dc
|
| 3 |
+
size 1262059159
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "do_lower_case": true, "word_delimiter_token": "|", "return_attention_mask": false, "do_normalize": true, "special_tokens_map_file": "special_tokens_map.json", "tokenizer_file": null, "name_or_path": "facebook/wav2vec2-large-lv60"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1716 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.512554112554113,
|
| 5 |
+
"global_step": 8500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.17,
|
| 12 |
+
"learning_rate": 8.333333333333334e-06,
|
| 13 |
+
"loss": 8.4908,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.35,
|
| 18 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 19 |
+
"loss": 8.1905,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.35,
|
| 24 |
+
"eval_loss": 771.380126953125,
|
| 25 |
+
"eval_runtime": 237.9275,
|
| 26 |
+
"eval_samples_per_second": 7.061,
|
| 27 |
+
"eval_wer": 1.0,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.52,
|
| 32 |
+
"learning_rate": 2.5e-05,
|
| 33 |
+
"loss": 6.0833,
|
| 34 |
+
"step": 150
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.69,
|
| 38 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 39 |
+
"loss": 4.2899,
|
| 40 |
+
"step": 200
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.69,
|
| 44 |
+
"eval_loss": 315.0731506347656,
|
| 45 |
+
"eval_runtime": 181.0423,
|
| 46 |
+
"eval_samples_per_second": 9.28,
|
| 47 |
+
"eval_wer": 1.0,
|
| 48 |
+
"step": 200
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.87,
|
| 52 |
+
"learning_rate": 4.1666666666666665e-05,
|
| 53 |
+
"loss": 3.5966,
|
| 54 |
+
"step": 250
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.04,
|
| 58 |
+
"learning_rate": 5e-05,
|
| 59 |
+
"loss": 3.2533,
|
| 60 |
+
"step": 300
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.04,
|
| 64 |
+
"eval_loss": 284.96343994140625,
|
| 65 |
+
"eval_runtime": 179.8422,
|
| 66 |
+
"eval_samples_per_second": 9.342,
|
| 67 |
+
"eval_wer": 0.9928532160527762,
|
| 68 |
+
"step": 300
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.21,
|
| 72 |
+
"learning_rate": 5.833333333333333e-05,
|
| 73 |
+
"loss": 3.0363,
|
| 74 |
+
"step": 350
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.39,
|
| 78 |
+
"learning_rate": 6.666666666666667e-05,
|
| 79 |
+
"loss": 3.0043,
|
| 80 |
+
"step": 400
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.39,
|
| 84 |
+
"eval_loss": 280.4071044921875,
|
| 85 |
+
"eval_runtime": 180.8047,
|
| 86 |
+
"eval_samples_per_second": 9.292,
|
| 87 |
+
"eval_wer": 1.0,
|
| 88 |
+
"step": 400
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.56,
|
| 92 |
+
"learning_rate": 7.5e-05,
|
| 93 |
+
"loss": 2.9767,
|
| 94 |
+
"step": 450
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.73,
|
| 98 |
+
"learning_rate": 8.333333333333333e-05,
|
| 99 |
+
"loss": 2.9848,
|
| 100 |
+
"step": 500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.73,
|
| 104 |
+
"eval_loss": 279.59625244140625,
|
| 105 |
+
"eval_runtime": 179.9553,
|
| 106 |
+
"eval_samples_per_second": 9.336,
|
| 107 |
+
"eval_wer": 1.0,
|
| 108 |
+
"step": 500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.91,
|
| 112 |
+
"learning_rate": 9.166666666666667e-05,
|
| 113 |
+
"loss": 2.967,
|
| 114 |
+
"step": 550
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.08,
|
| 118 |
+
"learning_rate": 0.0001,
|
| 119 |
+
"loss": 3.0088,
|
| 120 |
+
"step": 600
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 2.08,
|
| 124 |
+
"eval_loss": 276.5469665527344,
|
| 125 |
+
"eval_runtime": 181.9549,
|
| 126 |
+
"eval_samples_per_second": 9.233,
|
| 127 |
+
"eval_wer": 1.0,
|
| 128 |
+
"step": 600
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.26,
|
| 132 |
+
"learning_rate": 0.00010833333333333334,
|
| 133 |
+
"loss": 2.945,
|
| 134 |
+
"step": 650
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.43,
|
| 138 |
+
"learning_rate": 0.00011666666666666667,
|
| 139 |
+
"loss": 2.9085,
|
| 140 |
+
"step": 700
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.43,
|
| 144 |
+
"eval_loss": 261.8099365234375,
|
| 145 |
+
"eval_runtime": 182.2193,
|
| 146 |
+
"eval_samples_per_second": 9.22,
|
| 147 |
+
"eval_wer": 1.0,
|
| 148 |
+
"step": 700
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 2.6,
|
| 152 |
+
"learning_rate": 0.000125,
|
| 153 |
+
"loss": 2.6998,
|
| 154 |
+
"step": 750
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.78,
|
| 158 |
+
"learning_rate": 0.00013333333333333334,
|
| 159 |
+
"loss": 2.0025,
|
| 160 |
+
"step": 800
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.78,
|
| 164 |
+
"eval_loss": 113.09307098388672,
|
| 165 |
+
"eval_runtime": 182.7306,
|
| 166 |
+
"eval_samples_per_second": 9.194,
|
| 167 |
+
"eval_wer": 0.919392523364486,
|
| 168 |
+
"step": 800
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.95,
|
| 172 |
+
"learning_rate": 0.00014166666666666668,
|
| 173 |
+
"loss": 1.3056,
|
| 174 |
+
"step": 850
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 3.12,
|
| 178 |
+
"learning_rate": 0.00015,
|
| 179 |
+
"loss": 0.9594,
|
| 180 |
+
"step": 900
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.12,
|
| 184 |
+
"eval_loss": 48.241703033447266,
|
| 185 |
+
"eval_runtime": 182.3931,
|
| 186 |
+
"eval_samples_per_second": 9.211,
|
| 187 |
+
"eval_wer": 0.44193238042880706,
|
| 188 |
+
"step": 900
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.3,
|
| 192 |
+
"learning_rate": 0.00015833333333333332,
|
| 193 |
+
"loss": 0.7557,
|
| 194 |
+
"step": 950
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.47,
|
| 198 |
+
"learning_rate": 0.00016666666666666666,
|
| 199 |
+
"loss": 0.6721,
|
| 200 |
+
"step": 1000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.47,
|
| 204 |
+
"eval_loss": 36.373294830322266,
|
| 205 |
+
"eval_runtime": 182.5274,
|
| 206 |
+
"eval_samples_per_second": 9.204,
|
| 207 |
+
"eval_wer": 0.3460005497526113,
|
| 208 |
+
"step": 1000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.64,
|
| 212 |
+
"learning_rate": 0.000175,
|
| 213 |
+
"loss": 0.5788,
|
| 214 |
+
"step": 1050
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.82,
|
| 218 |
+
"learning_rate": 0.00018333333333333334,
|
| 219 |
+
"loss": 0.5235,
|
| 220 |
+
"step": 1100
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 3.82,
|
| 224 |
+
"eval_loss": 29.015974044799805,
|
| 225 |
+
"eval_runtime": 182.315,
|
| 226 |
+
"eval_samples_per_second": 9.215,
|
| 227 |
+
"eval_wer": 0.2875893347993403,
|
| 228 |
+
"step": 1100
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.99,
|
| 232 |
+
"learning_rate": 0.00019166666666666667,
|
| 233 |
+
"loss": 0.5232,
|
| 234 |
+
"step": 1150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.17,
|
| 238 |
+
"learning_rate": 0.0002,
|
| 239 |
+
"loss": 0.4547,
|
| 240 |
+
"step": 1200
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 4.17,
|
| 244 |
+
"eval_loss": 26.849903106689453,
|
| 245 |
+
"eval_runtime": 181.8627,
|
| 246 |
+
"eval_samples_per_second": 9.238,
|
| 247 |
+
"eval_wer": 0.26580538757559097,
|
| 248 |
+
"step": 1200
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.34,
|
| 252 |
+
"learning_rate": 0.00020833333333333335,
|
| 253 |
+
"loss": 0.4091,
|
| 254 |
+
"step": 1250
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.51,
|
| 258 |
+
"learning_rate": 0.00021666666666666668,
|
| 259 |
+
"loss": 0.4003,
|
| 260 |
+
"step": 1300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.51,
|
| 264 |
+
"eval_loss": 24.326810836791992,
|
| 265 |
+
"eval_runtime": 180.904,
|
| 266 |
+
"eval_samples_per_second": 9.287,
|
| 267 |
+
"eval_wer": 0.23735568993952721,
|
| 268 |
+
"step": 1300
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 4.69,
|
| 272 |
+
"learning_rate": 0.00022500000000000002,
|
| 273 |
+
"loss": 0.3822,
|
| 274 |
+
"step": 1350
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 4.86,
|
| 278 |
+
"learning_rate": 0.00023333333333333333,
|
| 279 |
+
"loss": 0.3476,
|
| 280 |
+
"step": 1400
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 4.86,
|
| 284 |
+
"eval_loss": 22.954877853393555,
|
| 285 |
+
"eval_runtime": 182.9264,
|
| 286 |
+
"eval_samples_per_second": 9.184,
|
| 287 |
+
"eval_wer": 0.22161902144035184,
|
| 288 |
+
"step": 1400
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.03,
|
| 292 |
+
"learning_rate": 0.00024166666666666667,
|
| 293 |
+
"loss": 0.3594,
|
| 294 |
+
"step": 1450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.21,
|
| 298 |
+
"learning_rate": 0.00025,
|
| 299 |
+
"loss": 0.3125,
|
| 300 |
+
"step": 1500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.21,
|
| 304 |
+
"eval_loss": 22.298479080200195,
|
| 305 |
+
"eval_runtime": 181.7168,
|
| 306 |
+
"eval_samples_per_second": 9.245,
|
| 307 |
+
"eval_wer": 0.20794392523364486,
|
| 308 |
+
"step": 1500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 5.38,
|
| 312 |
+
"learning_rate": 0.00025833333333333334,
|
| 313 |
+
"loss": 0.2878,
|
| 314 |
+
"step": 1550
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 5.55,
|
| 318 |
+
"learning_rate": 0.0002666666666666667,
|
| 319 |
+
"loss": 0.2762,
|
| 320 |
+
"step": 1600
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.55,
|
| 324 |
+
"eval_loss": 23.098875045776367,
|
| 325 |
+
"eval_runtime": 183.2329,
|
| 326 |
+
"eval_samples_per_second": 9.169,
|
| 327 |
+
"eval_wer": 0.21247938427707533,
|
| 328 |
+
"step": 1600
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 5.73,
|
| 332 |
+
"learning_rate": 0.000275,
|
| 333 |
+
"loss": 0.2703,
|
| 334 |
+
"step": 1650
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 5.9,
|
| 338 |
+
"learning_rate": 0.00028333333333333335,
|
| 339 |
+
"loss": 0.2827,
|
| 340 |
+
"step": 1700
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 5.9,
|
| 344 |
+
"eval_loss": 21.197721481323242,
|
| 345 |
+
"eval_runtime": 182.7196,
|
| 346 |
+
"eval_samples_per_second": 9.194,
|
| 347 |
+
"eval_wer": 0.19529961517317207,
|
| 348 |
+
"step": 1700
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 6.08,
|
| 352 |
+
"learning_rate": 0.0002916666666666667,
|
| 353 |
+
"loss": 0.2624,
|
| 354 |
+
"step": 1750
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.25,
|
| 358 |
+
"learning_rate": 0.0003,
|
| 359 |
+
"loss": 0.2306,
|
| 360 |
+
"step": 1800
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 6.25,
|
| 364 |
+
"eval_loss": 22.732194900512695,
|
| 365 |
+
"eval_runtime": 181.3273,
|
| 366 |
+
"eval_samples_per_second": 9.265,
|
| 367 |
+
"eval_wer": 0.18973336998350743,
|
| 368 |
+
"step": 1800
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 6.42,
|
| 372 |
+
"learning_rate": 0.00030833333333333337,
|
| 373 |
+
"loss": 0.2384,
|
| 374 |
+
"step": 1850
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 6.6,
|
| 378 |
+
"learning_rate": 0.00031666666666666665,
|
| 379 |
+
"loss": 0.2362,
|
| 380 |
+
"step": 1900
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 6.6,
|
| 384 |
+
"eval_loss": 22.076433181762695,
|
| 385 |
+
"eval_runtime": 181.988,
|
| 386 |
+
"eval_samples_per_second": 9.231,
|
| 387 |
+
"eval_wer": 0.18629741616272677,
|
| 388 |
+
"step": 1900
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 6.77,
|
| 392 |
+
"learning_rate": 0.00032500000000000004,
|
| 393 |
+
"loss": 0.2269,
|
| 394 |
+
"step": 1950
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 6.94,
|
| 398 |
+
"learning_rate": 0.0003333333333333333,
|
| 399 |
+
"loss": 0.2342,
|
| 400 |
+
"step": 2000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 6.94,
|
| 404 |
+
"eval_loss": 22.727157592773438,
|
| 405 |
+
"eval_runtime": 182.2314,
|
| 406 |
+
"eval_samples_per_second": 9.219,
|
| 407 |
+
"eval_wer": 0.17983782297965914,
|
| 408 |
+
"step": 2000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.12,
|
| 412 |
+
"learning_rate": 0.00034166666666666666,
|
| 413 |
+
"loss": 0.2271,
|
| 414 |
+
"step": 2050
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 7.29,
|
| 418 |
+
"learning_rate": 0.00035,
|
| 419 |
+
"loss": 0.2165,
|
| 420 |
+
"step": 2100
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.29,
|
| 424 |
+
"eval_loss": 21.797693252563477,
|
| 425 |
+
"eval_runtime": 182.6296,
|
| 426 |
+
"eval_samples_per_second": 9.199,
|
| 427 |
+
"eval_wer": 0.17468389224848818,
|
| 428 |
+
"step": 2100
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 7.46,
|
| 432 |
+
"learning_rate": 0.00035833333333333333,
|
| 433 |
+
"loss": 0.2037,
|
| 434 |
+
"step": 2150
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 7.64,
|
| 438 |
+
"learning_rate": 0.00036666666666666667,
|
| 439 |
+
"loss": 0.2134,
|
| 440 |
+
"step": 2200
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 7.64,
|
| 444 |
+
"eval_loss": 22.025192260742188,
|
| 445 |
+
"eval_runtime": 182.2586,
|
| 446 |
+
"eval_samples_per_second": 9.218,
|
| 447 |
+
"eval_wer": 0.18148708081363388,
|
| 448 |
+
"step": 2200
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 7.81,
|
| 452 |
+
"learning_rate": 0.000375,
|
| 453 |
+
"loss": 0.2096,
|
| 454 |
+
"step": 2250
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 7.98,
|
| 458 |
+
"learning_rate": 0.00038333333333333334,
|
| 459 |
+
"loss": 0.2033,
|
| 460 |
+
"step": 2300
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 7.98,
|
| 464 |
+
"eval_loss": 21.954696655273438,
|
| 465 |
+
"eval_runtime": 182.9541,
|
| 466 |
+
"eval_samples_per_second": 9.183,
|
| 467 |
+
"eval_wer": 0.17990654205607476,
|
| 468 |
+
"step": 2300
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 8.16,
|
| 472 |
+
"learning_rate": 0.0003916666666666667,
|
| 473 |
+
"loss": 0.2029,
|
| 474 |
+
"step": 2350
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 8.33,
|
| 478 |
+
"learning_rate": 0.0004,
|
| 479 |
+
"loss": 0.1948,
|
| 480 |
+
"step": 2400
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 8.33,
|
| 484 |
+
"eval_loss": 20.202104568481445,
|
| 485 |
+
"eval_runtime": 181.6845,
|
| 486 |
+
"eval_samples_per_second": 9.247,
|
| 487 |
+
"eval_wer": 0.17303463441451347,
|
| 488 |
+
"step": 2400
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 8.51,
|
| 492 |
+
"learning_rate": 0.00040833333333333336,
|
| 493 |
+
"loss": 0.1828,
|
| 494 |
+
"step": 2450
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.68,
|
| 498 |
+
"learning_rate": 0.0004166666666666667,
|
| 499 |
+
"loss": 0.1876,
|
| 500 |
+
"step": 2500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.68,
|
| 504 |
+
"eval_loss": 23.213821411132812,
|
| 505 |
+
"eval_runtime": 182.8529,
|
| 506 |
+
"eval_samples_per_second": 9.188,
|
| 507 |
+
"eval_wer": 0.1705607476635514,
|
| 508 |
+
"step": 2500
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 8.85,
|
| 512 |
+
"learning_rate": 0.000425,
|
| 513 |
+
"loss": 0.1721,
|
| 514 |
+
"step": 2550
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 9.03,
|
| 518 |
+
"learning_rate": 0.00043333333333333337,
|
| 519 |
+
"loss": 0.1774,
|
| 520 |
+
"step": 2600
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 9.03,
|
| 524 |
+
"eval_loss": 24.49812126159668,
|
| 525 |
+
"eval_runtime": 181.9478,
|
| 526 |
+
"eval_samples_per_second": 9.233,
|
| 527 |
+
"eval_wer": 0.174890049477735,
|
| 528 |
+
"step": 2600
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 9.2,
|
| 532 |
+
"learning_rate": 0.00044166666666666665,
|
| 533 |
+
"loss": 0.1613,
|
| 534 |
+
"step": 2650
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 9.37,
|
| 538 |
+
"learning_rate": 0.00045000000000000004,
|
| 539 |
+
"loss": 0.183,
|
| 540 |
+
"step": 2700
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 9.37,
|
| 544 |
+
"eval_loss": 25.343666076660156,
|
| 545 |
+
"eval_runtime": 181.6008,
|
| 546 |
+
"eval_samples_per_second": 9.251,
|
| 547 |
+
"eval_wer": 0.1805937328202309,
|
| 548 |
+
"step": 2700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 9.55,
|
| 552 |
+
"learning_rate": 0.0004583333333333333,
|
| 553 |
+
"loss": 0.1738,
|
| 554 |
+
"step": 2750
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 9.72,
|
| 558 |
+
"learning_rate": 0.00046666666666666666,
|
| 559 |
+
"loss": 0.1886,
|
| 560 |
+
"step": 2800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 9.72,
|
| 564 |
+
"eval_loss": 24.328418731689453,
|
| 565 |
+
"eval_runtime": 182.0168,
|
| 566 |
+
"eval_samples_per_second": 9.23,
|
| 567 |
+
"eval_wer": 0.18519791094007695,
|
| 568 |
+
"step": 2800
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 9.89,
|
| 572 |
+
"learning_rate": 0.000475,
|
| 573 |
+
"loss": 0.1855,
|
| 574 |
+
"step": 2850
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 10.07,
|
| 578 |
+
"learning_rate": 0.00048333333333333334,
|
| 579 |
+
"loss": 0.1784,
|
| 580 |
+
"step": 2900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 10.07,
|
| 584 |
+
"eval_loss": 24.794740676879883,
|
| 585 |
+
"eval_runtime": 182.2346,
|
| 586 |
+
"eval_samples_per_second": 9.219,
|
| 587 |
+
"eval_wer": 0.1759208356239692,
|
| 588 |
+
"step": 2900
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 10.24,
|
| 592 |
+
"learning_rate": 0.0004916666666666666,
|
| 593 |
+
"loss": 0.1639,
|
| 594 |
+
"step": 2950
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 10.42,
|
| 598 |
+
"learning_rate": 0.0005,
|
| 599 |
+
"loss": 0.1951,
|
| 600 |
+
"step": 3000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 10.42,
|
| 604 |
+
"eval_loss": 23.346879959106445,
|
| 605 |
+
"eval_runtime": 183.0341,
|
| 606 |
+
"eval_samples_per_second": 9.179,
|
| 607 |
+
"eval_wer": 0.1854040681693238,
|
| 608 |
+
"step": 3000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 10.59,
|
| 612 |
+
"learning_rate": 0.0004955673758865248,
|
| 613 |
+
"loss": 0.1841,
|
| 614 |
+
"step": 3050
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.76,
|
| 618 |
+
"learning_rate": 0.0004911347517730497,
|
| 619 |
+
"loss": 0.1638,
|
| 620 |
+
"step": 3100
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 10.76,
|
| 624 |
+
"eval_loss": 26.021047592163086,
|
| 625 |
+
"eval_runtime": 182.7843,
|
| 626 |
+
"eval_samples_per_second": 9.191,
|
| 627 |
+
"eval_wer": 0.18945849367784498,
|
| 628 |
+
"step": 3100
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 10.94,
|
| 632 |
+
"learning_rate": 0.0004867021276595745,
|
| 633 |
+
"loss": 0.1699,
|
| 634 |
+
"step": 3150
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 11.11,
|
| 638 |
+
"learning_rate": 0.00048226950354609925,
|
| 639 |
+
"loss": 0.178,
|
| 640 |
+
"step": 3200
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 11.11,
|
| 644 |
+
"eval_loss": 25.51027488708496,
|
| 645 |
+
"eval_runtime": 182.9857,
|
| 646 |
+
"eval_samples_per_second": 9.181,
|
| 647 |
+
"eval_wer": 0.18499175371083013,
|
| 648 |
+
"step": 3200
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 11.28,
|
| 652 |
+
"learning_rate": 0.0004778368794326241,
|
| 653 |
+
"loss": 0.1443,
|
| 654 |
+
"step": 3250
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 11.46,
|
| 658 |
+
"learning_rate": 0.00047340425531914893,
|
| 659 |
+
"loss": 0.1448,
|
| 660 |
+
"step": 3300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 11.46,
|
| 664 |
+
"eval_loss": 23.841064453125,
|
| 665 |
+
"eval_runtime": 182.7518,
|
| 666 |
+
"eval_samples_per_second": 9.193,
|
| 667 |
+
"eval_wer": 0.17702034084661902,
|
| 668 |
+
"step": 3300
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 11.63,
|
| 672 |
+
"learning_rate": 0.00046897163120567377,
|
| 673 |
+
"loss": 0.1415,
|
| 674 |
+
"step": 3350
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 11.8,
|
| 678 |
+
"learning_rate": 0.0004645390070921986,
|
| 679 |
+
"loss": 0.1451,
|
| 680 |
+
"step": 3400
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 11.8,
|
| 684 |
+
"eval_loss": 24.386722564697266,
|
| 685 |
+
"eval_runtime": 182.0009,
|
| 686 |
+
"eval_samples_per_second": 9.231,
|
| 687 |
+
"eval_wer": 0.17440901594282573,
|
| 688 |
+
"step": 3400
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 11.98,
|
| 692 |
+
"learning_rate": 0.00046010638297872344,
|
| 693 |
+
"loss": 0.1613,
|
| 694 |
+
"step": 3450
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 12.15,
|
| 698 |
+
"learning_rate": 0.0004556737588652483,
|
| 699 |
+
"loss": 0.1479,
|
| 700 |
+
"step": 3500
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 12.15,
|
| 704 |
+
"eval_loss": 25.48659324645996,
|
| 705 |
+
"eval_runtime": 183.184,
|
| 706 |
+
"eval_samples_per_second": 9.171,
|
| 707 |
+
"eval_wer": 0.18196811434854315,
|
| 708 |
+
"step": 3500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 12.33,
|
| 712 |
+
"learning_rate": 0.00045124113475177307,
|
| 713 |
+
"loss": 0.1249,
|
| 714 |
+
"step": 3550
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 12.5,
|
| 718 |
+
"learning_rate": 0.00044680851063829785,
|
| 719 |
+
"loss": 0.1255,
|
| 720 |
+
"step": 3600
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 12.5,
|
| 724 |
+
"eval_loss": 26.741592407226562,
|
| 725 |
+
"eval_runtime": 182.0975,
|
| 726 |
+
"eval_samples_per_second": 9.226,
|
| 727 |
+
"eval_wer": 0.17447773501924135,
|
| 728 |
+
"step": 3600
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.67,
|
| 732 |
+
"learning_rate": 0.0004423758865248227,
|
| 733 |
+
"loss": 0.1236,
|
| 734 |
+
"step": 3650
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 12.85,
|
| 738 |
+
"learning_rate": 0.0004379432624113475,
|
| 739 |
+
"loss": 0.1307,
|
| 740 |
+
"step": 3700
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 12.85,
|
| 744 |
+
"eval_loss": 25.000507354736328,
|
| 745 |
+
"eval_runtime": 182.3463,
|
| 746 |
+
"eval_samples_per_second": 9.213,
|
| 747 |
+
"eval_wer": 0.1726910390324354,
|
| 748 |
+
"step": 3700
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 13.02,
|
| 752 |
+
"learning_rate": 0.00043351063829787236,
|
| 753 |
+
"loss": 0.1504,
|
| 754 |
+
"step": 3750
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 13.19,
|
| 758 |
+
"learning_rate": 0.0004290780141843972,
|
| 759 |
+
"loss": 0.1185,
|
| 760 |
+
"step": 3800
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 13.19,
|
| 764 |
+
"eval_loss": 24.76068115234375,
|
| 765 |
+
"eval_runtime": 182.3064,
|
| 766 |
+
"eval_samples_per_second": 9.215,
|
| 767 |
+
"eval_wer": 0.1630703683342496,
|
| 768 |
+
"step": 3800
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 13.37,
|
| 772 |
+
"learning_rate": 0.000424645390070922,
|
| 773 |
+
"loss": 0.1151,
|
| 774 |
+
"step": 3850
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 13.54,
|
| 778 |
+
"learning_rate": 0.0004202127659574468,
|
| 779 |
+
"loss": 0.1296,
|
| 780 |
+
"step": 3900
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 13.54,
|
| 784 |
+
"eval_loss": 24.10735511779785,
|
| 785 |
+
"eval_runtime": 183.409,
|
| 786 |
+
"eval_samples_per_second": 9.16,
|
| 787 |
+
"eval_wer": 0.16197086311159978,
|
| 788 |
+
"step": 3900
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 13.71,
|
| 792 |
+
"learning_rate": 0.0004157801418439716,
|
| 793 |
+
"loss": 0.1099,
|
| 794 |
+
"step": 3950
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 13.89,
|
| 798 |
+
"learning_rate": 0.00041134751773049644,
|
| 799 |
+
"loss": 0.1141,
|
| 800 |
+
"step": 4000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 13.89,
|
| 804 |
+
"eval_loss": 27.41958236694336,
|
| 805 |
+
"eval_runtime": 182.5069,
|
| 806 |
+
"eval_samples_per_second": 9.205,
|
| 807 |
+
"eval_wer": 0.17647058823529413,
|
| 808 |
+
"step": 4000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 14.06,
|
| 812 |
+
"learning_rate": 0.0004069148936170213,
|
| 813 |
+
"loss": 0.1016,
|
| 814 |
+
"step": 4050
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 14.24,
|
| 818 |
+
"learning_rate": 0.0004024822695035461,
|
| 819 |
+
"loss": 0.1112,
|
| 820 |
+
"step": 4100
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 14.24,
|
| 824 |
+
"eval_loss": 26.094676971435547,
|
| 825 |
+
"eval_runtime": 182.6788,
|
| 826 |
+
"eval_samples_per_second": 9.196,
|
| 827 |
+
"eval_wer": 0.16760582737768004,
|
| 828 |
+
"step": 4100
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 14.41,
|
| 832 |
+
"learning_rate": 0.00039804964539007096,
|
| 833 |
+
"loss": 0.0982,
|
| 834 |
+
"step": 4150
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 14.58,
|
| 838 |
+
"learning_rate": 0.00039361702127659574,
|
| 839 |
+
"loss": 0.1021,
|
| 840 |
+
"step": 4200
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 14.58,
|
| 844 |
+
"eval_loss": 27.608373641967773,
|
| 845 |
+
"eval_runtime": 182.4822,
|
| 846 |
+
"eval_samples_per_second": 9.206,
|
| 847 |
+
"eval_wer": 0.1689802089059923,
|
| 848 |
+
"step": 4200
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 14.75,
|
| 852 |
+
"learning_rate": 0.0003891843971631206,
|
| 853 |
+
"loss": 0.1064,
|
| 854 |
+
"step": 4250
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 14.93,
|
| 858 |
+
"learning_rate": 0.00038475177304964536,
|
| 859 |
+
"loss": 0.112,
|
| 860 |
+
"step": 4300
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 14.93,
|
| 864 |
+
"eval_loss": 27.2677059173584,
|
| 865 |
+
"eval_runtime": 182.8419,
|
| 866 |
+
"eval_samples_per_second": 9.188,
|
| 867 |
+
"eval_wer": 0.17550852116547552,
|
| 868 |
+
"step": 4300
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 15.1,
|
| 872 |
+
"learning_rate": 0.0003803191489361702,
|
| 873 |
+
"loss": 0.1037,
|
| 874 |
+
"step": 4350
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 15.28,
|
| 878 |
+
"learning_rate": 0.00037588652482269504,
|
| 879 |
+
"loss": 0.1002,
|
| 880 |
+
"step": 4400
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 15.28,
|
| 884 |
+
"eval_loss": 25.244844436645508,
|
| 885 |
+
"eval_runtime": 182.1487,
|
| 886 |
+
"eval_samples_per_second": 9.223,
|
| 887 |
+
"eval_wer": 0.17021715228147333,
|
| 888 |
+
"step": 4400
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 15.45,
|
| 892 |
+
"learning_rate": 0.0003714539007092199,
|
| 893 |
+
"loss": 0.0889,
|
| 894 |
+
"step": 4450
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 15.62,
|
| 898 |
+
"learning_rate": 0.0003670212765957447,
|
| 899 |
+
"loss": 0.0893,
|
| 900 |
+
"step": 4500
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 15.62,
|
| 904 |
+
"eval_loss": 31.379478454589844,
|
| 905 |
+
"eval_runtime": 181.9808,
|
| 906 |
+
"eval_samples_per_second": 9.232,
|
| 907 |
+
"eval_wer": 0.17145409565695438,
|
| 908 |
+
"step": 4500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 15.8,
|
| 912 |
+
"learning_rate": 0.0003625886524822695,
|
| 913 |
+
"loss": 0.0925,
|
| 914 |
+
"step": 4550
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 15.97,
|
| 918 |
+
"learning_rate": 0.00035815602836879434,
|
| 919 |
+
"loss": 0.0911,
|
| 920 |
+
"step": 4600
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 15.97,
|
| 924 |
+
"eval_loss": 29.340713500976562,
|
| 925 |
+
"eval_runtime": 182.5822,
|
| 926 |
+
"eval_samples_per_second": 9.201,
|
| 927 |
+
"eval_wer": 0.1657504123144585,
|
| 928 |
+
"step": 4600
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 16.15,
|
| 932 |
+
"learning_rate": 0.0003537234042553192,
|
| 933 |
+
"loss": 0.0945,
|
| 934 |
+
"step": 4650
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 16.32,
|
| 938 |
+
"learning_rate": 0.00034929078014184396,
|
| 939 |
+
"loss": 0.0798,
|
| 940 |
+
"step": 4700
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 16.32,
|
| 944 |
+
"eval_loss": 27.59283447265625,
|
| 945 |
+
"eval_runtime": 183.4998,
|
| 946 |
+
"eval_samples_per_second": 9.155,
|
| 947 |
+
"eval_wer": 0.16877405167674547,
|
| 948 |
+
"step": 4700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 16.49,
|
| 952 |
+
"learning_rate": 0.0003448581560283688,
|
| 953 |
+
"loss": 0.0868,
|
| 954 |
+
"step": 4750
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 16.66,
|
| 958 |
+
"learning_rate": 0.00034042553191489364,
|
| 959 |
+
"loss": 0.0893,
|
| 960 |
+
"step": 4800
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 16.66,
|
| 964 |
+
"eval_loss": 26.155086517333984,
|
| 965 |
+
"eval_runtime": 183.3578,
|
| 966 |
+
"eval_samples_per_second": 9.162,
|
| 967 |
+
"eval_wer": 0.16877405167674547,
|
| 968 |
+
"step": 4800
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 16.84,
|
| 972 |
+
"learning_rate": 0.0003359929078014184,
|
| 973 |
+
"loss": 0.0844,
|
| 974 |
+
"step": 4850
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 17.01,
|
| 978 |
+
"learning_rate": 0.00033156028368794326,
|
| 979 |
+
"loss": 0.088,
|
| 980 |
+
"step": 4900
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 17.01,
|
| 984 |
+
"eval_loss": 27.282428741455078,
|
| 985 |
+
"eval_runtime": 183.8154,
|
| 986 |
+
"eval_samples_per_second": 9.14,
|
| 987 |
+
"eval_wer": 0.16870533260032985,
|
| 988 |
+
"step": 4900
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 17.19,
|
| 992 |
+
"learning_rate": 0.0003271276595744681,
|
| 993 |
+
"loss": 0.0844,
|
| 994 |
+
"step": 4950
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 17.36,
|
| 998 |
+
"learning_rate": 0.00032269503546099293,
|
| 999 |
+
"loss": 0.0647,
|
| 1000 |
+
"step": 5000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 17.36,
|
| 1004 |
+
"eval_loss": 27.251529693603516,
|
| 1005 |
+
"eval_runtime": 183.0422,
|
| 1006 |
+
"eval_samples_per_second": 9.178,
|
| 1007 |
+
"eval_wer": 0.1663001649257834,
|
| 1008 |
+
"step": 5000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 17.53,
|
| 1012 |
+
"learning_rate": 0.0003182624113475177,
|
| 1013 |
+
"loss": 0.0811,
|
| 1014 |
+
"step": 5050
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 17.71,
|
| 1018 |
+
"learning_rate": 0.00031382978723404256,
|
| 1019 |
+
"loss": 0.0727,
|
| 1020 |
+
"step": 5100
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 17.71,
|
| 1024 |
+
"eval_loss": 24.84876823425293,
|
| 1025 |
+
"eval_runtime": 182.8979,
|
| 1026 |
+
"eval_samples_per_second": 9.185,
|
| 1027 |
+
"eval_wer": 0.165131940626718,
|
| 1028 |
+
"step": 5100
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 17.88,
|
| 1032 |
+
"learning_rate": 0.0003093971631205674,
|
| 1033 |
+
"loss": 0.0744,
|
| 1034 |
+
"step": 5150
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 18.06,
|
| 1038 |
+
"learning_rate": 0.0003049645390070922,
|
| 1039 |
+
"loss": 0.0801,
|
| 1040 |
+
"step": 5200
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 18.06,
|
| 1044 |
+
"eval_loss": 26.26492691040039,
|
| 1045 |
+
"eval_runtime": 182.8513,
|
| 1046 |
+
"eval_samples_per_second": 9.188,
|
| 1047 |
+
"eval_wer": 0.1648570643210555,
|
| 1048 |
+
"step": 5200
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 18.23,
|
| 1052 |
+
"learning_rate": 0.000300531914893617,
|
| 1053 |
+
"loss": 0.0669,
|
| 1054 |
+
"step": 5250
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 18.4,
|
| 1058 |
+
"learning_rate": 0.00029609929078014185,
|
| 1059 |
+
"loss": 0.0613,
|
| 1060 |
+
"step": 5300
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 18.4,
|
| 1064 |
+
"eval_loss": 25.977203369140625,
|
| 1065 |
+
"eval_runtime": 182.5431,
|
| 1066 |
+
"eval_samples_per_second": 9.203,
|
| 1067 |
+
"eval_wer": 0.16114623419461244,
|
| 1068 |
+
"step": 5300
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 18.57,
|
| 1072 |
+
"learning_rate": 0.0002916666666666667,
|
| 1073 |
+
"loss": 0.0641,
|
| 1074 |
+
"step": 5350
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 18.75,
|
| 1078 |
+
"learning_rate": 0.00028723404255319153,
|
| 1079 |
+
"loss": 0.0588,
|
| 1080 |
+
"step": 5400
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 18.75,
|
| 1084 |
+
"eval_loss": 26.410762786865234,
|
| 1085 |
+
"eval_runtime": 182.6733,
|
| 1086 |
+
"eval_samples_per_second": 9.197,
|
| 1087 |
+
"eval_wer": 0.1667811984606927,
|
| 1088 |
+
"step": 5400
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 18.92,
|
| 1092 |
+
"learning_rate": 0.0002828014184397163,
|
| 1093 |
+
"loss": 0.0667,
|
| 1094 |
+
"step": 5450
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 19.1,
|
| 1098 |
+
"learning_rate": 0.00027836879432624115,
|
| 1099 |
+
"loss": 0.0768,
|
| 1100 |
+
"step": 5500
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 19.1,
|
| 1104 |
+
"eval_loss": 26.26220703125,
|
| 1105 |
+
"eval_runtime": 182.9701,
|
| 1106 |
+
"eval_samples_per_second": 9.182,
|
| 1107 |
+
"eval_wer": 0.15874106652006598,
|
| 1108 |
+
"step": 5500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 19.27,
|
| 1112 |
+
"learning_rate": 0.00027393617021276593,
|
| 1113 |
+
"loss": 0.0746,
|
| 1114 |
+
"step": 5550
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 19.44,
|
| 1118 |
+
"learning_rate": 0.00026950354609929077,
|
| 1119 |
+
"loss": 0.0616,
|
| 1120 |
+
"step": 5600
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 19.44,
|
| 1124 |
+
"eval_loss": 23.946300506591797,
|
| 1125 |
+
"eval_runtime": 183.2503,
|
| 1126 |
+
"eval_samples_per_second": 9.168,
|
| 1127 |
+
"eval_wer": 0.15681693238042882,
|
| 1128 |
+
"step": 5600
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 19.62,
|
| 1132 |
+
"learning_rate": 0.0002650709219858156,
|
| 1133 |
+
"loss": 0.061,
|
| 1134 |
+
"step": 5650
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 19.79,
|
| 1138 |
+
"learning_rate": 0.00026063829787234045,
|
| 1139 |
+
"loss": 0.066,
|
| 1140 |
+
"step": 5700
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 19.79,
|
| 1144 |
+
"eval_loss": 27.404529571533203,
|
| 1145 |
+
"eval_runtime": 183.1378,
|
| 1146 |
+
"eval_samples_per_second": 9.173,
|
| 1147 |
+
"eval_wer": 0.16004672897196262,
|
| 1148 |
+
"step": 5700
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 19.96,
|
| 1152 |
+
"learning_rate": 0.0002562056737588653,
|
| 1153 |
+
"loss": 0.0606,
|
| 1154 |
+
"step": 5750
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 20.14,
|
| 1158 |
+
"learning_rate": 0.00025177304964539007,
|
| 1159 |
+
"loss": 0.0499,
|
| 1160 |
+
"step": 5800
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 20.14,
|
| 1164 |
+
"eval_loss": 26.04939079284668,
|
| 1165 |
+
"eval_runtime": 182.3343,
|
| 1166 |
+
"eval_samples_per_second": 9.214,
|
| 1167 |
+
"eval_wer": 0.15805387575590985,
|
| 1168 |
+
"step": 5800
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 20.31,
|
| 1172 |
+
"learning_rate": 0.0002473404255319149,
|
| 1173 |
+
"loss": 0.0635,
|
| 1174 |
+
"step": 5850
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 20.48,
|
| 1178 |
+
"learning_rate": 0.00024290780141843972,
|
| 1179 |
+
"loss": 0.0576,
|
| 1180 |
+
"step": 5900
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 20.48,
|
| 1184 |
+
"eval_loss": 25.570714950561523,
|
| 1185 |
+
"eval_runtime": 182.4858,
|
| 1186 |
+
"eval_samples_per_second": 9.206,
|
| 1187 |
+
"eval_wer": 0.1499450247388675,
|
| 1188 |
+
"step": 5900
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 20.66,
|
| 1192 |
+
"learning_rate": 0.00023847517730496453,
|
| 1193 |
+
"loss": 0.0535,
|
| 1194 |
+
"step": 5950
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 20.83,
|
| 1198 |
+
"learning_rate": 0.00023404255319148937,
|
| 1199 |
+
"loss": 0.0538,
|
| 1200 |
+
"step": 6000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 20.83,
|
| 1204 |
+
"eval_loss": 26.44669532775879,
|
| 1205 |
+
"eval_runtime": 182.7081,
|
| 1206 |
+
"eval_samples_per_second": 9.195,
|
| 1207 |
+
"eval_wer": 0.15276250687190765,
|
| 1208 |
+
"step": 6000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 21.01,
|
| 1212 |
+
"learning_rate": 0.0002296099290780142,
|
| 1213 |
+
"loss": 0.06,
|
| 1214 |
+
"step": 6050
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 21.18,
|
| 1218 |
+
"learning_rate": 0.00022517730496453902,
|
| 1219 |
+
"loss": 0.0547,
|
| 1220 |
+
"step": 6100
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 21.18,
|
| 1224 |
+
"eval_loss": 26.02959442138672,
|
| 1225 |
+
"eval_runtime": 183.3837,
|
| 1226 |
+
"eval_samples_per_second": 9.161,
|
| 1227 |
+
"eval_wer": 0.15083837273227046,
|
| 1228 |
+
"step": 6100
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 21.35,
|
| 1232 |
+
"learning_rate": 0.00022074468085106383,
|
| 1233 |
+
"loss": 0.0506,
|
| 1234 |
+
"step": 6150
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 21.53,
|
| 1238 |
+
"learning_rate": 0.00021631205673758867,
|
| 1239 |
+
"loss": 0.0578,
|
| 1240 |
+
"step": 6200
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 21.53,
|
| 1244 |
+
"eval_loss": 25.79215431213379,
|
| 1245 |
+
"eval_runtime": 182.8615,
|
| 1246 |
+
"eval_samples_per_second": 9.187,
|
| 1247 |
+
"eval_wer": 0.1515255634964266,
|
| 1248 |
+
"step": 6200
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 21.7,
|
| 1252 |
+
"learning_rate": 0.00021187943262411348,
|
| 1253 |
+
"loss": 0.0504,
|
| 1254 |
+
"step": 6250
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 21.87,
|
| 1258 |
+
"learning_rate": 0.0002074468085106383,
|
| 1259 |
+
"loss": 0.0425,
|
| 1260 |
+
"step": 6300
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 21.87,
|
| 1264 |
+
"eval_loss": 26.764015197753906,
|
| 1265 |
+
"eval_runtime": 182.7224,
|
| 1266 |
+
"eval_samples_per_second": 9.194,
|
| 1267 |
+
"eval_wer": 0.15592358438702583,
|
| 1268 |
+
"step": 6300
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 22.05,
|
| 1272 |
+
"learning_rate": 0.00020301418439716313,
|
| 1273 |
+
"loss": 0.0509,
|
| 1274 |
+
"step": 6350
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 22.22,
|
| 1278 |
+
"learning_rate": 0.00019858156028368796,
|
| 1279 |
+
"loss": 0.0473,
|
| 1280 |
+
"step": 6400
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 22.22,
|
| 1284 |
+
"eval_loss": 26.02473258972168,
|
| 1285 |
+
"eval_runtime": 183.8649,
|
| 1286 |
+
"eval_samples_per_second": 9.137,
|
| 1287 |
+
"eval_wer": 0.15028862012094557,
|
| 1288 |
+
"step": 6400
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 22.39,
|
| 1292 |
+
"learning_rate": 0.00019414893617021275,
|
| 1293 |
+
"loss": 0.039,
|
| 1294 |
+
"step": 6450
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 22.57,
|
| 1298 |
+
"learning_rate": 0.00018971631205673758,
|
| 1299 |
+
"loss": 0.0519,
|
| 1300 |
+
"step": 6500
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 22.57,
|
| 1304 |
+
"eval_loss": 26.268949508666992,
|
| 1305 |
+
"eval_runtime": 183.1205,
|
| 1306 |
+
"eval_samples_per_second": 9.174,
|
| 1307 |
+
"eval_wer": 0.15166300164925783,
|
| 1308 |
+
"step": 6500
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 22.74,
|
| 1312 |
+
"learning_rate": 0.00018528368794326242,
|
| 1313 |
+
"loss": 0.0533,
|
| 1314 |
+
"step": 6550
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 22.91,
|
| 1318 |
+
"learning_rate": 0.00018085106382978726,
|
| 1319 |
+
"loss": 0.0412,
|
| 1320 |
+
"step": 6600
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 22.91,
|
| 1324 |
+
"eval_loss": 26.845340728759766,
|
| 1325 |
+
"eval_runtime": 183.0902,
|
| 1326 |
+
"eval_samples_per_second": 9.176,
|
| 1327 |
+
"eval_wer": 0.15063221550302364,
|
| 1328 |
+
"step": 6600
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 23.09,
|
| 1332 |
+
"learning_rate": 0.00017641843971631204,
|
| 1333 |
+
"loss": 0.0535,
|
| 1334 |
+
"step": 6650
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 23.26,
|
| 1338 |
+
"learning_rate": 0.00017198581560283688,
|
| 1339 |
+
"loss": 0.0449,
|
| 1340 |
+
"step": 6700
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 23.26,
|
| 1344 |
+
"eval_loss": 24.5848445892334,
|
| 1345 |
+
"eval_runtime": 183.5855,
|
| 1346 |
+
"eval_samples_per_second": 9.151,
|
| 1347 |
+
"eval_wer": 0.14898295766904893,
|
| 1348 |
+
"step": 6700
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 23.44,
|
| 1352 |
+
"learning_rate": 0.00016755319148936172,
|
| 1353 |
+
"loss": 0.0366,
|
| 1354 |
+
"step": 6750
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 23.61,
|
| 1358 |
+
"learning_rate": 0.00016312056737588653,
|
| 1359 |
+
"loss": 0.0419,
|
| 1360 |
+
"step": 6800
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 23.61,
|
| 1364 |
+
"eval_loss": 26.725446701049805,
|
| 1365 |
+
"eval_runtime": 182.6943,
|
| 1366 |
+
"eval_samples_per_second": 9.196,
|
| 1367 |
+
"eval_wer": 0.15063221550302364,
|
| 1368 |
+
"step": 6800
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 23.78,
|
| 1372 |
+
"learning_rate": 0.00015868794326241134,
|
| 1373 |
+
"loss": 0.0404,
|
| 1374 |
+
"step": 6850
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 23.96,
|
| 1378 |
+
"learning_rate": 0.00015425531914893618,
|
| 1379 |
+
"loss": 0.0417,
|
| 1380 |
+
"step": 6900
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 23.96,
|
| 1384 |
+
"eval_loss": 26.662160873413086,
|
| 1385 |
+
"eval_runtime": 182.9944,
|
| 1386 |
+
"eval_samples_per_second": 9.181,
|
| 1387 |
+
"eval_wer": 0.14444749862561848,
|
| 1388 |
+
"step": 6900
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 24.13,
|
| 1392 |
+
"learning_rate": 0.000149822695035461,
|
| 1393 |
+
"loss": 0.0341,
|
| 1394 |
+
"step": 6950
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 24.3,
|
| 1398 |
+
"learning_rate": 0.0001453900709219858,
|
| 1399 |
+
"loss": 0.0383,
|
| 1400 |
+
"step": 7000
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 24.3,
|
| 1404 |
+
"eval_loss": 25.917531967163086,
|
| 1405 |
+
"eval_runtime": 182.9271,
|
| 1406 |
+
"eval_samples_per_second": 9.184,
|
| 1407 |
+
"eval_wer": 0.14699010445299615,
|
| 1408 |
+
"step": 7000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 24.48,
|
| 1412 |
+
"learning_rate": 0.00014095744680851064,
|
| 1413 |
+
"loss": 0.0361,
|
| 1414 |
+
"step": 7050
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 24.65,
|
| 1418 |
+
"learning_rate": 0.00013652482269503548,
|
| 1419 |
+
"loss": 0.037,
|
| 1420 |
+
"step": 7100
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 24.65,
|
| 1424 |
+
"eval_loss": 26.451175689697266,
|
| 1425 |
+
"eval_runtime": 182.8419,
|
| 1426 |
+
"eval_samples_per_second": 9.188,
|
| 1427 |
+
"eval_wer": 0.14314183617372184,
|
| 1428 |
+
"step": 7100
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 24.82,
|
| 1432 |
+
"learning_rate": 0.0001320921985815603,
|
| 1433 |
+
"loss": 0.0358,
|
| 1434 |
+
"step": 7150
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 25.0,
|
| 1438 |
+
"learning_rate": 0.0001276595744680851,
|
| 1439 |
+
"loss": 0.0421,
|
| 1440 |
+
"step": 7200
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 25.0,
|
| 1444 |
+
"eval_loss": 24.637380599975586,
|
| 1445 |
+
"eval_runtime": 182.468,
|
| 1446 |
+
"eval_samples_per_second": 9.207,
|
| 1447 |
+
"eval_wer": 0.14307311709730622,
|
| 1448 |
+
"step": 7200
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 25.17,
|
| 1452 |
+
"learning_rate": 0.00012322695035460994,
|
| 1453 |
+
"loss": 0.034,
|
| 1454 |
+
"step": 7250
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 25.35,
|
| 1458 |
+
"learning_rate": 0.00011879432624113476,
|
| 1459 |
+
"loss": 0.0366,
|
| 1460 |
+
"step": 7300
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 25.35,
|
| 1464 |
+
"eval_loss": 25.25677490234375,
|
| 1465 |
+
"eval_runtime": 183.0484,
|
| 1466 |
+
"eval_samples_per_second": 9.178,
|
| 1467 |
+
"eval_wer": 0.14369158878504673,
|
| 1468 |
+
"step": 7300
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 25.52,
|
| 1472 |
+
"learning_rate": 0.00011436170212765957,
|
| 1473 |
+
"loss": 0.0358,
|
| 1474 |
+
"step": 7350
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 25.69,
|
| 1478 |
+
"learning_rate": 0.00010992907801418441,
|
| 1479 |
+
"loss": 0.0411,
|
| 1480 |
+
"step": 7400
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 25.69,
|
| 1484 |
+
"eval_loss": 23.710235595703125,
|
| 1485 |
+
"eval_runtime": 183.8494,
|
| 1486 |
+
"eval_samples_per_second": 9.138,
|
| 1487 |
+
"eval_wer": 0.1435541506322155,
|
| 1488 |
+
"step": 7400
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 25.87,
|
| 1492 |
+
"learning_rate": 0.00010549645390070922,
|
| 1493 |
+
"loss": 0.0324,
|
| 1494 |
+
"step": 7450
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 26.04,
|
| 1498 |
+
"learning_rate": 0.00010106382978723403,
|
| 1499 |
+
"loss": 0.0395,
|
| 1500 |
+
"step": 7500
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 26.04,
|
| 1504 |
+
"eval_loss": 25.433012008666992,
|
| 1505 |
+
"eval_runtime": 183.0989,
|
| 1506 |
+
"eval_samples_per_second": 9.175,
|
| 1507 |
+
"eval_wer": 0.14108026388125344,
|
| 1508 |
+
"step": 7500
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 26.21,
|
| 1512 |
+
"learning_rate": 9.663120567375887e-05,
|
| 1513 |
+
"loss": 0.0373,
|
| 1514 |
+
"step": 7550
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 26.39,
|
| 1518 |
+
"learning_rate": 9.219858156028368e-05,
|
| 1519 |
+
"loss": 0.0378,
|
| 1520 |
+
"step": 7600
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 26.39,
|
| 1524 |
+
"eval_loss": 24.688125610351562,
|
| 1525 |
+
"eval_runtime": 187.9819,
|
| 1526 |
+
"eval_samples_per_second": 8.937,
|
| 1527 |
+
"eval_wer": 0.13963716327652556,
|
| 1528 |
+
"step": 7600
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 26.56,
|
| 1532 |
+
"learning_rate": 8.776595744680852e-05,
|
| 1533 |
+
"loss": 0.0305,
|
| 1534 |
+
"step": 7650
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 26.73,
|
| 1538 |
+
"learning_rate": 8.333333333333333e-05,
|
| 1539 |
+
"loss": 0.0295,
|
| 1540 |
+
"step": 7700
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 26.73,
|
| 1544 |
+
"eval_loss": 24.53731918334961,
|
| 1545 |
+
"eval_runtime": 184.8939,
|
| 1546 |
+
"eval_samples_per_second": 9.086,
|
| 1547 |
+
"eval_wer": 0.14114898295766906,
|
| 1548 |
+
"step": 7700
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 26.91,
|
| 1552 |
+
"learning_rate": 7.890070921985815e-05,
|
| 1553 |
+
"loss": 0.0391,
|
| 1554 |
+
"step": 7750
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 27.08,
|
| 1558 |
+
"learning_rate": 7.446808510638298e-05,
|
| 1559 |
+
"loss": 0.0295,
|
| 1560 |
+
"step": 7800
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 27.08,
|
| 1564 |
+
"eval_loss": 24.03723907470703,
|
| 1565 |
+
"eval_runtime": 183.5479,
|
| 1566 |
+
"eval_samples_per_second": 9.153,
|
| 1567 |
+
"eval_wer": 0.14183617372182517,
|
| 1568 |
+
"step": 7800
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 27.26,
|
| 1572 |
+
"learning_rate": 7.00354609929078e-05,
|
| 1573 |
+
"loss": 0.0367,
|
| 1574 |
+
"step": 7850
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 27.43,
|
| 1578 |
+
"learning_rate": 6.560283687943263e-05,
|
| 1579 |
+
"loss": 0.0357,
|
| 1580 |
+
"step": 7900
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 27.43,
|
| 1584 |
+
"eval_loss": 23.551311492919922,
|
| 1585 |
+
"eval_runtime": 186.1508,
|
| 1586 |
+
"eval_samples_per_second": 9.025,
|
| 1587 |
+
"eval_wer": 0.1392935678944475,
|
| 1588 |
+
"step": 7900
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 27.6,
|
| 1592 |
+
"learning_rate": 6.117021276595745e-05,
|
| 1593 |
+
"loss": 0.0278,
|
| 1594 |
+
"step": 7950
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 27.78,
|
| 1598 |
+
"learning_rate": 5.673758865248227e-05,
|
| 1599 |
+
"loss": 0.0292,
|
| 1600 |
+
"step": 8000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 27.78,
|
| 1604 |
+
"eval_loss": 26.10912322998047,
|
| 1605 |
+
"eval_runtime": 183.6318,
|
| 1606 |
+
"eval_samples_per_second": 9.149,
|
| 1607 |
+
"eval_wer": 0.1403243540406817,
|
| 1608 |
+
"step": 8000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 27.95,
|
| 1612 |
+
"learning_rate": 5.2304964539007095e-05,
|
| 1613 |
+
"loss": 0.0293,
|
| 1614 |
+
"step": 8050
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 28.12,
|
| 1618 |
+
"learning_rate": 4.787234042553191e-05,
|
| 1619 |
+
"loss": 0.0247,
|
| 1620 |
+
"step": 8100
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 28.12,
|
| 1624 |
+
"eval_loss": 25.034645080566406,
|
| 1625 |
+
"eval_runtime": 183.9776,
|
| 1626 |
+
"eval_samples_per_second": 9.132,
|
| 1627 |
+
"eval_wer": 0.13688840021990104,
|
| 1628 |
+
"step": 8100
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 28.3,
|
| 1632 |
+
"learning_rate": 4.343971631205674e-05,
|
| 1633 |
+
"loss": 0.0334,
|
| 1634 |
+
"step": 8150
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 28.47,
|
| 1638 |
+
"learning_rate": 3.900709219858156e-05,
|
| 1639 |
+
"loss": 0.0238,
|
| 1640 |
+
"step": 8200
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 28.47,
|
| 1644 |
+
"eval_loss": 24.014360427856445,
|
| 1645 |
+
"eval_runtime": 183.9625,
|
| 1646 |
+
"eval_samples_per_second": 9.132,
|
| 1647 |
+
"eval_wer": 0.1355827377680044,
|
| 1648 |
+
"step": 8200
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 28.64,
|
| 1652 |
+
"learning_rate": 3.4574468085106386e-05,
|
| 1653 |
+
"loss": 0.0342,
|
| 1654 |
+
"step": 8250
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 28.82,
|
| 1658 |
+
"learning_rate": 3.0141843971631203e-05,
|
| 1659 |
+
"loss": 0.0278,
|
| 1660 |
+
"step": 8300
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 28.82,
|
| 1664 |
+
"eval_loss": 24.382017135620117,
|
| 1665 |
+
"eval_runtime": 183.4893,
|
| 1666 |
+
"eval_samples_per_second": 9.156,
|
| 1667 |
+
"eval_wer": 0.13585761407366684,
|
| 1668 |
+
"step": 8300
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 28.99,
|
| 1672 |
+
"learning_rate": 2.5709219858156028e-05,
|
| 1673 |
+
"loss": 0.0209,
|
| 1674 |
+
"step": 8350
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 29.17,
|
| 1678 |
+
"learning_rate": 2.1276595744680852e-05,
|
| 1679 |
+
"loss": 0.0289,
|
| 1680 |
+
"step": 8400
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 29.17,
|
| 1684 |
+
"eval_loss": 25.17869758605957,
|
| 1685 |
+
"eval_runtime": 184.0892,
|
| 1686 |
+
"eval_samples_per_second": 9.126,
|
| 1687 |
+
"eval_wer": 0.13551401869158877,
|
| 1688 |
+
"step": 8400
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 29.34,
|
| 1692 |
+
"learning_rate": 1.6843971631205673e-05,
|
| 1693 |
+
"loss": 0.0306,
|
| 1694 |
+
"step": 8450
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 29.51,
|
| 1698 |
+
"learning_rate": 1.2411347517730498e-05,
|
| 1699 |
+
"loss": 0.0238,
|
| 1700 |
+
"step": 8500
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 29.51,
|
| 1704 |
+
"eval_loss": 25.424150466918945,
|
| 1705 |
+
"eval_runtime": 182.9499,
|
| 1706 |
+
"eval_samples_per_second": 9.183,
|
| 1707 |
+
"eval_wer": 0.1351704233095107,
|
| 1708 |
+
"step": 8500
|
| 1709 |
+
}
|
| 1710 |
+
],
|
| 1711 |
+
"max_steps": 8640,
|
| 1712 |
+
"num_train_epochs": 30,
|
| 1713 |
+
"total_flos": 1.2862009633623437e+19,
|
| 1714 |
+
"trial_name": null,
|
| 1715 |
+
"trial_params": null
|
| 1716 |
+
}
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "E": 5, "T": 6, "A": 7, "O": 8, "N": 9, "I": 10, "H": 11, "S": 12, "R": 13, "D": 14, "L": 15, "U": 16, "M": 17, "W": 18, "C": 19, "F": 20, "G": 21, "Y": 22, "P": 23, "B": 24, "V": 25, "K": 26, "'": 27, "X": 28, "J": 29, "Q": 30, "Z": 31}
|