Commit
•
025ea45
1
Parent(s):
c10dec3
Upload 9 files
Browse files- config.json +115 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +874 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 768,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": true,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.0,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.0,
|
| 55 |
+
"hidden_act": "gelu",
|
| 56 |
+
"hidden_dropout": 0.1,
|
| 57 |
+
"hidden_size": 1024,
|
| 58 |
+
"initializer_range": 0.02,
|
| 59 |
+
"intermediate_size": 4096,
|
| 60 |
+
"layer_norm_eps": 1e-05,
|
| 61 |
+
"layerdrop": 0.1,
|
| 62 |
+
"mask_channel_length": 10,
|
| 63 |
+
"mask_channel_min_space": 1,
|
| 64 |
+
"mask_channel_other": 0.0,
|
| 65 |
+
"mask_channel_prob": 0.0,
|
| 66 |
+
"mask_channel_selection": "static",
|
| 67 |
+
"mask_feature_length": 10,
|
| 68 |
+
"mask_feature_min_masks": 0,
|
| 69 |
+
"mask_feature_prob": 0.0,
|
| 70 |
+
"mask_time_length": 10,
|
| 71 |
+
"mask_time_min_masks": 2,
|
| 72 |
+
"mask_time_min_space": 1,
|
| 73 |
+
"mask_time_other": 0.0,
|
| 74 |
+
"mask_time_prob": 0.05,
|
| 75 |
+
"mask_time_selection": "static",
|
| 76 |
+
"model_type": "wav2vec2",
|
| 77 |
+
"num_adapter_layers": 3,
|
| 78 |
+
"num_attention_heads": 16,
|
| 79 |
+
"num_codevector_groups": 2,
|
| 80 |
+
"num_codevectors_per_group": 320,
|
| 81 |
+
"num_conv_pos_embedding_groups": 16,
|
| 82 |
+
"num_conv_pos_embeddings": 128,
|
| 83 |
+
"num_feat_extract_layers": 7,
|
| 84 |
+
"num_hidden_layers": 24,
|
| 85 |
+
"num_negatives": 100,
|
| 86 |
+
"output_hidden_size": 1024,
|
| 87 |
+
"pad_token_id": 55,
|
| 88 |
+
"proj_codevector_dim": 768,
|
| 89 |
+
"tdnn_dilation": [
|
| 90 |
+
1,
|
| 91 |
+
2,
|
| 92 |
+
3,
|
| 93 |
+
1,
|
| 94 |
+
1
|
| 95 |
+
],
|
| 96 |
+
"tdnn_dim": [
|
| 97 |
+
512,
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
1500
|
| 102 |
+
],
|
| 103 |
+
"tdnn_kernel": [
|
| 104 |
+
5,
|
| 105 |
+
3,
|
| 106 |
+
3,
|
| 107 |
+
1,
|
| 108 |
+
1
|
| 109 |
+
],
|
| 110 |
+
"torch_dtype": "float32",
|
| 111 |
+
"transformers_version": "4.25.1",
|
| 112 |
+
"use_weighted_layer_sum": false,
|
| 113 |
+
"vocab_size": 56,
|
| 114 |
+
"xvector_output_dim": 512
|
| 115 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa5c5f97d0530f6d8c3d62759c5c5f42af2bf711cb916015e659a6485993b9e8
|
| 3 |
+
size 2524217989
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": false,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93088ae299a94f7670a2ddcbcb818bd0340fc8ca561cbb60a9cbb2c060fb15ab
|
| 3 |
+
size 1262131437
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5eb2bfa65ad7fb9b3570fc335191a88983bcec06eae6e52cb5bd0e69a30fa7dd
|
| 3 |
+
size 15661
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f81667d8de4cb859a61ad387e6225578dead4ba12c9a17895b1d7a4398329342
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e2cb7e0d02020c1dd18a390e7b645de0e8a7602c78cebee2fbe05af498d0851
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,874 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4851953222194576,
|
| 5 |
+
"global_step": 3900,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 2.6399999999999995e-05,
|
| 13 |
+
"loss": 12.4264,
|
| 14 |
+
"step": 50
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 5.6399999999999995e-05,
|
| 19 |
+
"loss": 7.1545,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"eval_cer": 0.990295846526883,
|
| 25 |
+
"eval_loss": 4.435502529144287,
|
| 26 |
+
"eval_runtime": 2212.4333,
|
| 27 |
+
"eval_samples_per_second": 21.798,
|
| 28 |
+
"eval_steps_per_second": 1.363,
|
| 29 |
+
"eval_wer": 0.9803344576185385,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02,
|
| 34 |
+
"learning_rate": 8.639999999999999e-05,
|
| 35 |
+
"loss": 3.2621,
|
| 36 |
+
"step": 150
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.02,
|
| 40 |
+
"learning_rate": 0.0001164,
|
| 41 |
+
"loss": 2.9431,
|
| 42 |
+
"step": 200
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.02,
|
| 46 |
+
"eval_cer": 0.990295846526883,
|
| 47 |
+
"eval_loss": 3.032039165496826,
|
| 48 |
+
"eval_runtime": 2093.0103,
|
| 49 |
+
"eval_samples_per_second": 23.041,
|
| 50 |
+
"eval_steps_per_second": 1.441,
|
| 51 |
+
"eval_wer": 0.9803344576185385,
|
| 52 |
+
"step": 200
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.03,
|
| 56 |
+
"learning_rate": 0.00014639999999999998,
|
| 57 |
+
"loss": 2.9236,
|
| 58 |
+
"step": 250
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.04,
|
| 62 |
+
"learning_rate": 0.00017639999999999998,
|
| 63 |
+
"loss": 2.9188,
|
| 64 |
+
"step": 300
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.04,
|
| 68 |
+
"eval_cer": 0.990295846526883,
|
| 69 |
+
"eval_loss": 2.976602077484131,
|
| 70 |
+
"eval_runtime": 2082.4254,
|
| 71 |
+
"eval_samples_per_second": 23.159,
|
| 72 |
+
"eval_steps_per_second": 1.448,
|
| 73 |
+
"eval_wer": 0.9803344576185385,
|
| 74 |
+
"step": 300
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 0.00020639999999999998,
|
| 79 |
+
"loss": 2.8775,
|
| 80 |
+
"step": 350
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.05,
|
| 84 |
+
"learning_rate": 0.0002364,
|
| 85 |
+
"loss": 2.7809,
|
| 86 |
+
"step": 400
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05,
|
| 90 |
+
"eval_cer": 0.9696764691640024,
|
| 91 |
+
"eval_loss": 2.7085533142089844,
|
| 92 |
+
"eval_runtime": 2082.5137,
|
| 93 |
+
"eval_samples_per_second": 23.158,
|
| 94 |
+
"eval_steps_per_second": 1.448,
|
| 95 |
+
"eval_wer": 0.9589848218162464,
|
| 96 |
+
"step": 400
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.06,
|
| 100 |
+
"learning_rate": 0.00026639999999999997,
|
| 101 |
+
"loss": 2.5919,
|
| 102 |
+
"step": 450
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.06,
|
| 106 |
+
"learning_rate": 0.0002964,
|
| 107 |
+
"loss": 2.3631,
|
| 108 |
+
"step": 500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.06,
|
| 112 |
+
"eval_cer": 0.862518823322377,
|
| 113 |
+
"eval_loss": 1.9892584085464478,
|
| 114 |
+
"eval_runtime": 2074.1857,
|
| 115 |
+
"eval_samples_per_second": 23.251,
|
| 116 |
+
"eval_steps_per_second": 1.454,
|
| 117 |
+
"eval_wer": 0.9145354269893812,
|
| 118 |
+
"step": 500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.07,
|
| 122 |
+
"learning_rate": 0.0002994410095705937,
|
| 123 |
+
"loss": 1.5803,
|
| 124 |
+
"step": 550
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.07,
|
| 128 |
+
"learning_rate": 0.0002988057931735411,
|
| 129 |
+
"loss": 0.9821,
|
| 130 |
+
"step": 600
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.07,
|
| 134 |
+
"eval_cer": 0.1523255882620419,
|
| 135 |
+
"eval_loss": 0.5998358726501465,
|
| 136 |
+
"eval_runtime": 2075.5254,
|
| 137 |
+
"eval_samples_per_second": 23.236,
|
| 138 |
+
"eval_steps_per_second": 1.453,
|
| 139 |
+
"eval_wer": 0.3576434163952664,
|
| 140 |
+
"step": 600
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.08,
|
| 144 |
+
"learning_rate": 0.0002981705767764885,
|
| 145 |
+
"loss": 0.7758,
|
| 146 |
+
"step": 650
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.09,
|
| 150 |
+
"learning_rate": 0.0002975353603794359,
|
| 151 |
+
"loss": 0.6585,
|
| 152 |
+
"step": 700
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.09,
|
| 156 |
+
"eval_cer": 0.10982663055738294,
|
| 157 |
+
"eval_loss": 0.4177716076374054,
|
| 158 |
+
"eval_runtime": 2075.6462,
|
| 159 |
+
"eval_samples_per_second": 23.234,
|
| 160 |
+
"eval_steps_per_second": 1.453,
|
| 161 |
+
"eval_wer": 0.25202041187922813,
|
| 162 |
+
"step": 700
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.09,
|
| 166 |
+
"learning_rate": 0.0002969001439823833,
|
| 167 |
+
"loss": 0.6077,
|
| 168 |
+
"step": 750
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.1,
|
| 172 |
+
"learning_rate": 0.0002962649275853307,
|
| 173 |
+
"loss": 0.5657,
|
| 174 |
+
"step": 800
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.1,
|
| 178 |
+
"eval_cer": 0.09587676230587619,
|
| 179 |
+
"eval_loss": 0.36425185203552246,
|
| 180 |
+
"eval_runtime": 2094.5001,
|
| 181 |
+
"eval_samples_per_second": 23.025,
|
| 182 |
+
"eval_steps_per_second": 1.439,
|
| 183 |
+
"eval_wer": 0.21289942193910186,
|
| 184 |
+
"step": 800
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.11,
|
| 188 |
+
"learning_rate": 0.00029562971118827813,
|
| 189 |
+
"loss": 0.5295,
|
| 190 |
+
"step": 850
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.11,
|
| 194 |
+
"learning_rate": 0.0002949944947912255,
|
| 195 |
+
"loss": 0.4998,
|
| 196 |
+
"step": 900
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.11,
|
| 200 |
+
"eval_cer": 0.09022533657576749,
|
| 201 |
+
"eval_loss": 0.3022037148475647,
|
| 202 |
+
"eval_runtime": 2108.0878,
|
| 203 |
+
"eval_samples_per_second": 22.877,
|
| 204 |
+
"eval_steps_per_second": 1.43,
|
| 205 |
+
"eval_wer": 0.20626376037718358,
|
| 206 |
+
"step": 900
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.12,
|
| 210 |
+
"learning_rate": 0.00029435927839417295,
|
| 211 |
+
"loss": 0.4887,
|
| 212 |
+
"step": 950
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.12,
|
| 216 |
+
"learning_rate": 0.0002937240619971203,
|
| 217 |
+
"loss": 0.4902,
|
| 218 |
+
"step": 1000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.12,
|
| 222 |
+
"eval_cer": 0.0833139440810026,
|
| 223 |
+
"eval_loss": 0.2917855978012085,
|
| 224 |
+
"eval_runtime": 2122.6451,
|
| 225 |
+
"eval_samples_per_second": 22.72,
|
| 226 |
+
"eval_steps_per_second": 1.42,
|
| 227 |
+
"eval_wer": 0.1899083786482299,
|
| 228 |
+
"step": 1000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.13,
|
| 232 |
+
"learning_rate": 0.0002930888456000677,
|
| 233 |
+
"loss": 0.4776,
|
| 234 |
+
"step": 1050
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.14,
|
| 238 |
+
"learning_rate": 0.00029245362920301513,
|
| 239 |
+
"loss": 0.468,
|
| 240 |
+
"step": 1100
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.14,
|
| 244 |
+
"eval_cer": 0.08389881054318646,
|
| 245 |
+
"eval_loss": 0.2726401686668396,
|
| 246 |
+
"eval_runtime": 2129.6342,
|
| 247 |
+
"eval_samples_per_second": 22.645,
|
| 248 |
+
"eval_steps_per_second": 1.416,
|
| 249 |
+
"eval_wer": 0.1888892520157431,
|
| 250 |
+
"step": 1100
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.14,
|
| 254 |
+
"learning_rate": 0.00029181841280596254,
|
| 255 |
+
"loss": 0.44,
|
| 256 |
+
"step": 1150
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.15,
|
| 260 |
+
"learning_rate": 0.00029118319640890995,
|
| 261 |
+
"loss": 0.4332,
|
| 262 |
+
"step": 1200
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15,
|
| 266 |
+
"eval_cer": 0.07579453612837772,
|
| 267 |
+
"eval_loss": 0.2764703929424286,
|
| 268 |
+
"eval_runtime": 2387.2613,
|
| 269 |
+
"eval_samples_per_second": 20.201,
|
| 270 |
+
"eval_steps_per_second": 1.263,
|
| 271 |
+
"eval_wer": 0.1726946095004212,
|
| 272 |
+
"step": 1200
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.16,
|
| 276 |
+
"learning_rate": 0.00029054798001185736,
|
| 277 |
+
"loss": 0.4363,
|
| 278 |
+
"step": 1250
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.16,
|
| 282 |
+
"learning_rate": 0.00028991276361480477,
|
| 283 |
+
"loss": 0.4178,
|
| 284 |
+
"step": 1300
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.16,
|
| 288 |
+
"eval_cer": 0.07956404299571626,
|
| 289 |
+
"eval_loss": 0.2525430917739868,
|
| 290 |
+
"eval_runtime": 2685.8951,
|
| 291 |
+
"eval_samples_per_second": 17.955,
|
| 292 |
+
"eval_steps_per_second": 1.123,
|
| 293 |
+
"eval_wer": 0.18290121608219176,
|
| 294 |
+
"step": 1300
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.17,
|
| 298 |
+
"learning_rate": 0.0002892775472177521,
|
| 299 |
+
"loss": 0.4072,
|
| 300 |
+
"step": 1350
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.17,
|
| 304 |
+
"learning_rate": 0.0002886423308206996,
|
| 305 |
+
"loss": 0.4171,
|
| 306 |
+
"step": 1400
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.17,
|
| 310 |
+
"eval_cer": 0.07121410576965095,
|
| 311 |
+
"eval_loss": 0.24207398295402527,
|
| 312 |
+
"eval_runtime": 2387.6421,
|
| 313 |
+
"eval_samples_per_second": 20.198,
|
| 314 |
+
"eval_steps_per_second": 1.263,
|
| 315 |
+
"eval_wer": 0.16114939876197457,
|
| 316 |
+
"step": 1400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.18,
|
| 320 |
+
"learning_rate": 0.00028800711442364694,
|
| 321 |
+
"loss": 0.4074,
|
| 322 |
+
"step": 1450
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.19,
|
| 326 |
+
"learning_rate": 0.00028737189802659435,
|
| 327 |
+
"loss": 0.4049,
|
| 328 |
+
"step": 1500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.19,
|
| 332 |
+
"eval_cer": 0.07119875184394742,
|
| 333 |
+
"eval_loss": 0.23904423415660858,
|
| 334 |
+
"eval_runtime": 2682.696,
|
| 335 |
+
"eval_samples_per_second": 17.977,
|
| 336 |
+
"eval_steps_per_second": 1.124,
|
| 337 |
+
"eval_wer": 0.1614502011908041,
|
| 338 |
+
"step": 1500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19,
|
| 342 |
+
"learning_rate": 0.00028673668162954176,
|
| 343 |
+
"loss": 0.3901,
|
| 344 |
+
"step": 1550
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.2,
|
| 348 |
+
"learning_rate": 0.0002861014652324892,
|
| 349 |
+
"loss": 0.3963,
|
| 350 |
+
"step": 1600
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.2,
|
| 354 |
+
"eval_cer": 0.07037200199837249,
|
| 355 |
+
"eval_loss": 0.22700898349285126,
|
| 356 |
+
"eval_runtime": 2633.1938,
|
| 357 |
+
"eval_samples_per_second": 18.315,
|
| 358 |
+
"eval_steps_per_second": 1.145,
|
| 359 |
+
"eval_wer": 0.16208515443303412,
|
| 360 |
+
"step": 1600
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.21,
|
| 364 |
+
"learning_rate": 0.0002854662488354366,
|
| 365 |
+
"loss": 0.3813,
|
| 366 |
+
"step": 1650
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.21,
|
| 370 |
+
"learning_rate": 0.000284831032438384,
|
| 371 |
+
"loss": 0.3924,
|
| 372 |
+
"step": 1700
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.21,
|
| 376 |
+
"eval_cer": 0.07119875184394742,
|
| 377 |
+
"eval_loss": 0.2278161644935608,
|
| 378 |
+
"eval_runtime": 2675.9933,
|
| 379 |
+
"eval_samples_per_second": 18.022,
|
| 380 |
+
"eval_steps_per_second": 1.127,
|
| 381 |
+
"eval_wer": 0.16218386565136397,
|
| 382 |
+
"step": 1700
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.22,
|
| 386 |
+
"learning_rate": 0.0002841958160413314,
|
| 387 |
+
"loss": 0.3938,
|
| 388 |
+
"step": 1750
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.22,
|
| 392 |
+
"learning_rate": 0.00028356059964427876,
|
| 393 |
+
"loss": 0.3942,
|
| 394 |
+
"step": 1800
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.22,
|
| 398 |
+
"eval_cer": 0.06874165130289871,
|
| 399 |
+
"eval_loss": 0.22094863653182983,
|
| 400 |
+
"eval_runtime": 2685.0595,
|
| 401 |
+
"eval_samples_per_second": 17.961,
|
| 402 |
+
"eval_steps_per_second": 1.123,
|
| 403 |
+
"eval_wer": 0.1562058342331839,
|
| 404 |
+
"step": 1800
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.23,
|
| 408 |
+
"learning_rate": 0.0002829253832472262,
|
| 409 |
+
"loss": 0.3674,
|
| 410 |
+
"step": 1850
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.24,
|
| 414 |
+
"learning_rate": 0.0002822901668501736,
|
| 415 |
+
"loss": 0.3745,
|
| 416 |
+
"step": 1900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.24,
|
| 420 |
+
"eval_cer": 0.0659505438242377,
|
| 421 |
+
"eval_loss": 0.21558238565921783,
|
| 422 |
+
"eval_runtime": 2439.0418,
|
| 423 |
+
"eval_samples_per_second": 19.773,
|
| 424 |
+
"eval_steps_per_second": 1.236,
|
| 425 |
+
"eval_wer": 0.1506740042005625,
|
| 426 |
+
"step": 1900
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.24,
|
| 430 |
+
"learning_rate": 0.000281654950453121,
|
| 431 |
+
"loss": 0.3711,
|
| 432 |
+
"step": 1950
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.25,
|
| 436 |
+
"learning_rate": 0.0002810197340560684,
|
| 437 |
+
"loss": 0.3598,
|
| 438 |
+
"step": 2000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.25,
|
| 442 |
+
"eval_cer": 0.06898353468628977,
|
| 443 |
+
"eval_loss": 0.221131831407547,
|
| 444 |
+
"eval_runtime": 2663.9985,
|
| 445 |
+
"eval_samples_per_second": 18.103,
|
| 446 |
+
"eval_steps_per_second": 1.132,
|
| 447 |
+
"eval_wer": 0.15953867178719996,
|
| 448 |
+
"step": 2000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.26,
|
| 452 |
+
"learning_rate": 0.0002803845176590158,
|
| 453 |
+
"loss": 0.3756,
|
| 454 |
+
"step": 2050
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.26,
|
| 458 |
+
"learning_rate": 0.0002797493012619632,
|
| 459 |
+
"loss": 0.3653,
|
| 460 |
+
"step": 2100
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.26,
|
| 464 |
+
"eval_cer": 0.06659281034712863,
|
| 465 |
+
"eval_loss": 0.2138465791940689,
|
| 466 |
+
"eval_runtime": 2389.088,
|
| 467 |
+
"eval_samples_per_second": 20.186,
|
| 468 |
+
"eval_steps_per_second": 1.262,
|
| 469 |
+
"eval_wer": 0.1510541757846708,
|
| 470 |
+
"step": 2100
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.27,
|
| 474 |
+
"learning_rate": 0.00027911408486491063,
|
| 475 |
+
"loss": 0.3751,
|
| 476 |
+
"step": 2150
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.27,
|
| 480 |
+
"learning_rate": 0.00027847886846785804,
|
| 481 |
+
"loss": 0.3576,
|
| 482 |
+
"step": 2200
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.27,
|
| 486 |
+
"eval_cer": 0.06642438959287295,
|
| 487 |
+
"eval_loss": 0.2111993283033371,
|
| 488 |
+
"eval_runtime": 2676.0178,
|
| 489 |
+
"eval_samples_per_second": 18.022,
|
| 490 |
+
"eval_steps_per_second": 1.127,
|
| 491 |
+
"eval_wer": 0.1520546273217979,
|
| 492 |
+
"step": 2200
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.28,
|
| 496 |
+
"learning_rate": 0.00027784365207080545,
|
| 497 |
+
"loss": 0.353,
|
| 498 |
+
"step": 2250
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.29,
|
| 502 |
+
"learning_rate": 0.00027720843567375286,
|
| 503 |
+
"loss": 0.3569,
|
| 504 |
+
"step": 2300
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.29,
|
| 508 |
+
"eval_cer": 0.06564322909592542,
|
| 509 |
+
"eval_loss": 0.21856307983398438,
|
| 510 |
+
"eval_runtime": 2642.3302,
|
| 511 |
+
"eval_samples_per_second": 18.251,
|
| 512 |
+
"eval_steps_per_second": 1.141,
|
| 513 |
+
"eval_wer": 0.14776135629213985,
|
| 514 |
+
"step": 2300
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.29,
|
| 518 |
+
"learning_rate": 0.00027657321927670027,
|
| 519 |
+
"loss": 0.371,
|
| 520 |
+
"step": 2350
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3,
|
| 524 |
+
"learning_rate": 0.0002759380028796476,
|
| 525 |
+
"loss": 0.3473,
|
| 526 |
+
"step": 2400
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.3,
|
| 530 |
+
"eval_cer": 0.0664718686554331,
|
| 531 |
+
"eval_loss": 0.2098075896501541,
|
| 532 |
+
"eval_runtime": 2685.1293,
|
| 533 |
+
"eval_samples_per_second": 17.96,
|
| 534 |
+
"eval_steps_per_second": 1.123,
|
| 535 |
+
"eval_wer": 0.15094546005096968,
|
| 536 |
+
"step": 2400
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.3,
|
| 540 |
+
"learning_rate": 0.0002753027864825951,
|
| 541 |
+
"loss": 0.3574,
|
| 542 |
+
"step": 2450
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.31,
|
| 546 |
+
"learning_rate": 0.00027466757008554244,
|
| 547 |
+
"loss": 0.3735,
|
| 548 |
+
"step": 2500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.31,
|
| 552 |
+
"eval_cer": 0.06522796445920521,
|
| 553 |
+
"eval_loss": 0.20497867465019226,
|
| 554 |
+
"eval_runtime": 2686.3997,
|
| 555 |
+
"eval_samples_per_second": 17.952,
|
| 556 |
+
"eval_steps_per_second": 1.122,
|
| 557 |
+
"eval_wer": 0.1481148491685914,
|
| 558 |
+
"step": 2500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 0.32,
|
| 562 |
+
"learning_rate": 0.00027403235368848985,
|
| 563 |
+
"loss": 0.3617,
|
| 564 |
+
"step": 2550
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.32,
|
| 568 |
+
"learning_rate": 0.00027339713729143726,
|
| 569 |
+
"loss": 0.3523,
|
| 570 |
+
"step": 2600
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.32,
|
| 574 |
+
"eval_cer": 0.06236906349320707,
|
| 575 |
+
"eval_loss": 0.20248664915561676,
|
| 576 |
+
"eval_runtime": 2418.5335,
|
| 577 |
+
"eval_samples_per_second": 19.94,
|
| 578 |
+
"eval_steps_per_second": 1.247,
|
| 579 |
+
"eval_wer": 0.14132578503764698,
|
| 580 |
+
"step": 2600
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 0.33,
|
| 584 |
+
"learning_rate": 0.0002727619208943847,
|
| 585 |
+
"loss": 0.3466,
|
| 586 |
+
"step": 2650
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.34,
|
| 590 |
+
"learning_rate": 0.0002721267044973321,
|
| 591 |
+
"loss": 0.3443,
|
| 592 |
+
"step": 2700
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.34,
|
| 596 |
+
"eval_cer": 0.0631530585610537,
|
| 597 |
+
"eval_loss": 0.19711410999298096,
|
| 598 |
+
"eval_runtime": 2678.8408,
|
| 599 |
+
"eval_samples_per_second": 18.003,
|
| 600 |
+
"eval_steps_per_second": 1.125,
|
| 601 |
+
"eval_wer": 0.14343206900714522,
|
| 602 |
+
"step": 2700
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.34,
|
| 606 |
+
"learning_rate": 0.0002714914881002795,
|
| 607 |
+
"loss": 0.3353,
|
| 608 |
+
"step": 2750
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.35,
|
| 612 |
+
"learning_rate": 0.0002708562717032269,
|
| 613 |
+
"loss": 0.3271,
|
| 614 |
+
"step": 2800
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.35,
|
| 618 |
+
"eval_cer": 0.06349556921136332,
|
| 619 |
+
"eval_loss": 0.1974596232175827,
|
| 620 |
+
"eval_runtime": 2413.2013,
|
| 621 |
+
"eval_samples_per_second": 19.984,
|
| 622 |
+
"eval_steps_per_second": 1.249,
|
| 623 |
+
"eval_wer": 0.14444185809195217,
|
| 624 |
+
"step": 2800
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.35,
|
| 628 |
+
"learning_rate": 0.00027022105530617426,
|
| 629 |
+
"loss": 0.3411,
|
| 630 |
+
"step": 2850
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 0.36,
|
| 634 |
+
"learning_rate": 0.0002695858389091217,
|
| 635 |
+
"loss": 0.347,
|
| 636 |
+
"step": 2900
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.36,
|
| 640 |
+
"eval_cer": 0.06194080707319925,
|
| 641 |
+
"eval_loss": 0.19428882002830505,
|
| 642 |
+
"eval_runtime": 2689.3153,
|
| 643 |
+
"eval_samples_per_second": 17.932,
|
| 644 |
+
"eval_steps_per_second": 1.121,
|
| 645 |
+
"eval_wer": 0.14129443755615034,
|
| 646 |
+
"step": 2900
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.37,
|
| 650 |
+
"learning_rate": 0.0002689506225120691,
|
| 651 |
+
"loss": 0.3414,
|
| 652 |
+
"step": 2950
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 0.37,
|
| 656 |
+
"learning_rate": 0.0002683154061150165,
|
| 657 |
+
"loss": 0.3291,
|
| 658 |
+
"step": 3000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.37,
|
| 662 |
+
"eval_cer": 0.06220513080954164,
|
| 663 |
+
"eval_loss": 0.19397367537021637,
|
| 664 |
+
"eval_runtime": 2678.8937,
|
| 665 |
+
"eval_samples_per_second": 18.002,
|
| 666 |
+
"eval_steps_per_second": 1.125,
|
| 667 |
+
"eval_wer": 0.14132445110226416,
|
| 668 |
+
"step": 3000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.38,
|
| 672 |
+
"learning_rate": 0.0002676801897179639,
|
| 673 |
+
"loss": 0.3472,
|
| 674 |
+
"step": 3050
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.39,
|
| 678 |
+
"learning_rate": 0.0002670449733209113,
|
| 679 |
+
"loss": 0.3386,
|
| 680 |
+
"step": 3100
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.39,
|
| 684 |
+
"eval_cer": 0.06227859343867701,
|
| 685 |
+
"eval_loss": 0.18810917437076569,
|
| 686 |
+
"eval_runtime": 2096.7687,
|
| 687 |
+
"eval_samples_per_second": 23.0,
|
| 688 |
+
"eval_steps_per_second": 1.438,
|
| 689 |
+
"eval_wer": 0.143275331599662,
|
| 690 |
+
"step": 3100
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.39,
|
| 694 |
+
"learning_rate": 0.0002664097569238587,
|
| 695 |
+
"loss": 0.3561,
|
| 696 |
+
"step": 3150
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.4,
|
| 700 |
+
"learning_rate": 0.00026577454052680613,
|
| 701 |
+
"loss": 0.3346,
|
| 702 |
+
"step": 3200
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.4,
|
| 706 |
+
"eval_cer": 0.062040017054668245,
|
| 707 |
+
"eval_loss": 0.19153311848640442,
|
| 708 |
+
"eval_runtime": 2107.1655,
|
| 709 |
+
"eval_samples_per_second": 22.887,
|
| 710 |
+
"eval_steps_per_second": 1.431,
|
| 711 |
+
"eval_wer": 0.141297105426916,
|
| 712 |
+
"step": 3200
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.4,
|
| 716 |
+
"learning_rate": 0.00026513932412975354,
|
| 717 |
+
"loss": 0.3362,
|
| 718 |
+
"step": 3250
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.41,
|
| 722 |
+
"learning_rate": 0.0002645041077327009,
|
| 723 |
+
"loss": 0.3286,
|
| 724 |
+
"step": 3300
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.41,
|
| 728 |
+
"eval_cer": 0.06275858077759366,
|
| 729 |
+
"eval_loss": 0.18800656497478485,
|
| 730 |
+
"eval_runtime": 2108.9053,
|
| 731 |
+
"eval_samples_per_second": 22.868,
|
| 732 |
+
"eval_steps_per_second": 1.43,
|
| 733 |
+
"eval_wer": 0.14638073317090447,
|
| 734 |
+
"step": 3300
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.42,
|
| 738 |
+
"learning_rate": 0.00026386889133564836,
|
| 739 |
+
"loss": 0.3088,
|
| 740 |
+
"step": 3350
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.42,
|
| 744 |
+
"learning_rate": 0.0002632336749385957,
|
| 745 |
+
"loss": 0.3211,
|
| 746 |
+
"step": 3400
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.42,
|
| 750 |
+
"eval_cer": 0.06097185625418542,
|
| 751 |
+
"eval_loss": 0.1875077337026596,
|
| 752 |
+
"eval_runtime": 2192.9622,
|
| 753 |
+
"eval_samples_per_second": 21.991,
|
| 754 |
+
"eval_steps_per_second": 1.375,
|
| 755 |
+
"eval_wer": 0.1381210052803832,
|
| 756 |
+
"step": 3400
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 0.43,
|
| 760 |
+
"learning_rate": 0.0002625984585415431,
|
| 761 |
+
"loss": 0.3185,
|
| 762 |
+
"step": 3450
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.44,
|
| 766 |
+
"learning_rate": 0.00026196324214449053,
|
| 767 |
+
"loss": 0.3272,
|
| 768 |
+
"step": 3500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.44,
|
| 772 |
+
"eval_cer": 0.060900047124741195,
|
| 773 |
+
"eval_loss": 0.18167421221733093,
|
| 774 |
+
"eval_runtime": 2110.7367,
|
| 775 |
+
"eval_samples_per_second": 22.848,
|
| 776 |
+
"eval_steps_per_second": 1.428,
|
| 777 |
+
"eval_wer": 0.14127176065464214,
|
| 778 |
+
"step": 3500
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.44,
|
| 782 |
+
"learning_rate": 0.00026132802574743794,
|
| 783 |
+
"loss": 0.321,
|
| 784 |
+
"step": 3550
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.45,
|
| 788 |
+
"learning_rate": 0.00026069280935038535,
|
| 789 |
+
"loss": 0.3246,
|
| 790 |
+
"step": 3600
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.45,
|
| 794 |
+
"eval_cer": 0.05953118559471069,
|
| 795 |
+
"eval_loss": 0.18253999948501587,
|
| 796 |
+
"eval_runtime": 2122.2554,
|
| 797 |
+
"eval_samples_per_second": 22.724,
|
| 798 |
+
"eval_steps_per_second": 1.421,
|
| 799 |
+
"eval_wer": 0.13588532957874988,
|
| 800 |
+
"step": 3600
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.45,
|
| 804 |
+
"learning_rate": 0.00026005759295333276,
|
| 805 |
+
"loss": 0.3266,
|
| 806 |
+
"step": 3650
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.46,
|
| 810 |
+
"learning_rate": 0.0002594223765562802,
|
| 811 |
+
"loss": 0.3239,
|
| 812 |
+
"step": 3700
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.46,
|
| 816 |
+
"eval_cer": 0.06006573842343529,
|
| 817 |
+
"eval_loss": 0.18680287897586823,
|
| 818 |
+
"eval_runtime": 2677.6172,
|
| 819 |
+
"eval_samples_per_second": 18.011,
|
| 820 |
+
"eval_steps_per_second": 1.126,
|
| 821 |
+
"eval_wer": 0.13703584884644604,
|
| 822 |
+
"step": 3700
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.47,
|
| 826 |
+
"learning_rate": 0.00025878716015922753,
|
| 827 |
+
"loss": 0.3272,
|
| 828 |
+
"step": 3750
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.47,
|
| 832 |
+
"learning_rate": 0.000258151943762175,
|
| 833 |
+
"loss": 0.32,
|
| 834 |
+
"step": 3800
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.47,
|
| 838 |
+
"eval_cer": 0.05884947129347375,
|
| 839 |
+
"eval_loss": 0.18254593014717102,
|
| 840 |
+
"eval_runtime": 2588.5152,
|
| 841 |
+
"eval_samples_per_second": 18.631,
|
| 842 |
+
"eval_steps_per_second": 1.165,
|
| 843 |
+
"eval_wer": 0.13350625582346165,
|
| 844 |
+
"step": 3800
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.48,
|
| 848 |
+
"learning_rate": 0.00025751672736512235,
|
| 849 |
+
"loss": 0.3148,
|
| 850 |
+
"step": 3850
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.49,
|
| 854 |
+
"learning_rate": 0.00025688151096806976,
|
| 855 |
+
"loss": 0.3185,
|
| 856 |
+
"step": 3900
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.49,
|
| 860 |
+
"eval_cer": 0.06071367408812445,
|
| 861 |
+
"eval_loss": 0.1804201900959015,
|
| 862 |
+
"eval_runtime": 2677.674,
|
| 863 |
+
"eval_samples_per_second": 18.01,
|
| 864 |
+
"eval_steps_per_second": 1.126,
|
| 865 |
+
"eval_wer": 0.14048273787569457,
|
| 866 |
+
"step": 3900
|
| 867 |
+
}
|
| 868 |
+
],
|
| 869 |
+
"max_steps": 24114,
|
| 870 |
+
"num_train_epochs": 3,
|
| 871 |
+
"total_flos": 3.282673713375086e+19,
|
| 872 |
+
"trial_name": null,
|
| 873 |
+
"trial_params": null
|
| 874 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4cb666f5f8e6ca5ae9d87c4a2f91beb06a5571cc4fc75c7b5cb2d039650ca38d
|
| 3 |
+
size 3451
|