Training in progress, epoch 53, checkpoint
Browse files- checkpoint-13674/config.json +82 -0
- checkpoint-13674/model.safetensors +3 -0
- checkpoint-13674/optimizer.pt +3 -0
- checkpoint-13674/preprocessor_config.json +11 -0
- checkpoint-13674/rng_state.pth +3 -0
- checkpoint-13674/scheduler.pt +3 -0
- checkpoint-13674/trainer_state.json +943 -0
- checkpoint-13674/training_args.bin +3 -0
checkpoint-13674/config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/w2v-bert-2.0",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_act": "relu",
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": true,
|
| 8 |
+
"apply_spec_augment": false,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2BertForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 768,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"conformer_conv_dropout": 0.1,
|
| 17 |
+
"contrastive_logits_temperature": 0.1,
|
| 18 |
+
"conv_depthwise_kernel_size": 31,
|
| 19 |
+
"ctc_loss_reduction": "mean",
|
| 20 |
+
"ctc_zero_infinity": true,
|
| 21 |
+
"diversity_loss_weight": 0.1,
|
| 22 |
+
"eos_token_id": 2,
|
| 23 |
+
"feat_proj_dropout": 0.0,
|
| 24 |
+
"feat_quantizer_dropout": 0.0,
|
| 25 |
+
"feature_projection_input_dim": 160,
|
| 26 |
+
"final_dropout": 0.1,
|
| 27 |
+
"hidden_act": "swish",
|
| 28 |
+
"hidden_dropout": 0.0,
|
| 29 |
+
"hidden_size": 1024,
|
| 30 |
+
"initializer_range": 0.02,
|
| 31 |
+
"intermediate_size": 4096,
|
| 32 |
+
"layer_norm_eps": 1e-05,
|
| 33 |
+
"layerdrop": 0.0,
|
| 34 |
+
"left_max_position_embeddings": 64,
|
| 35 |
+
"mask_feature_length": 10,
|
| 36 |
+
"mask_feature_min_masks": 0,
|
| 37 |
+
"mask_feature_prob": 0.0,
|
| 38 |
+
"mask_time_length": 10,
|
| 39 |
+
"mask_time_min_masks": 2,
|
| 40 |
+
"mask_time_prob": 0.0,
|
| 41 |
+
"max_source_positions": 5000,
|
| 42 |
+
"model_type": "wav2vec2-bert",
|
| 43 |
+
"num_adapter_layers": 1,
|
| 44 |
+
"num_attention_heads": 16,
|
| 45 |
+
"num_codevector_groups": 2,
|
| 46 |
+
"num_codevectors_per_group": 320,
|
| 47 |
+
"num_hidden_layers": 24,
|
| 48 |
+
"num_negatives": 100,
|
| 49 |
+
"output_hidden_size": 1024,
|
| 50 |
+
"pad_token_id": 43,
|
| 51 |
+
"position_embeddings_type": "relative_key",
|
| 52 |
+
"proj_codevector_dim": 768,
|
| 53 |
+
"right_max_position_embeddings": 8,
|
| 54 |
+
"rotary_embedding_base": 10000,
|
| 55 |
+
"tdnn_dilation": [
|
| 56 |
+
1,
|
| 57 |
+
2,
|
| 58 |
+
3,
|
| 59 |
+
1,
|
| 60 |
+
1
|
| 61 |
+
],
|
| 62 |
+
"tdnn_dim": [
|
| 63 |
+
512,
|
| 64 |
+
512,
|
| 65 |
+
512,
|
| 66 |
+
512,
|
| 67 |
+
1500
|
| 68 |
+
],
|
| 69 |
+
"tdnn_kernel": [
|
| 70 |
+
5,
|
| 71 |
+
3,
|
| 72 |
+
3,
|
| 73 |
+
1,
|
| 74 |
+
1
|
| 75 |
+
],
|
| 76 |
+
"torch_dtype": "float32",
|
| 77 |
+
"transformers_version": "4.46.0",
|
| 78 |
+
"use_intermediate_ffn_before_adapter": false,
|
| 79 |
+
"use_weighted_layer_sum": false,
|
| 80 |
+
"vocab_size": 46,
|
| 81 |
+
"xvector_output_dim": 512
|
| 82 |
+
}
|
checkpoint-13674/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9abf3eebb0cda617444d203c4e05b582f3cd5424b4d149692e3b2aa659330e34
|
| 3 |
+
size 2423003160
|
checkpoint-13674/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af153f37c835637b578442503d557c29c36175295ed22edbb77b34652c4d1cd7
|
| 3 |
+
size 4846467050
|
checkpoint-13674/preprocessor_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"feature_extractor_type": "SeamlessM4TFeatureExtractor",
|
| 3 |
+
"feature_size": 80,
|
| 4 |
+
"num_mel_bins": 80,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 1,
|
| 7 |
+
"processor_class": "Wav2Vec2BertProcessor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000,
|
| 10 |
+
"stride": 2
|
| 11 |
+
}
|
checkpoint-13674/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0f9bf88414f264c3d47cac7765c092aaf2ed35288979219e64aba7734488424
|
| 3 |
+
size 14244
|
checkpoint-13674/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:759cd4edd9102f2c597823853a87e08fcc521b9e289585d1825e135b280ff2e9
|
| 3 |
+
size 1064
|
checkpoint-13674/trainer_state.json
ADDED
|
@@ -0,0 +1,943 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.39242170690891703,
|
| 3 |
+
"best_model_checkpoint": "w2v-bert-2.0-CV_Fleurs-lg-5hrs-v4/checkpoint-11868",
|
| 4 |
+
"epoch": 53.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 13674,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 11.246596336364746,
|
| 14 |
+
"learning_rate": 5.1200000000000004e-05,
|
| 15 |
+
"loss": 1.9959,
|
| 16 |
+
"step": 258
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_cer": 0.13379445701894616,
|
| 21 |
+
"eval_loss": 0.5475470423698425,
|
| 22 |
+
"eval_runtime": 29.6601,
|
| 23 |
+
"eval_samples_per_second": 27.849,
|
| 24 |
+
"eval_steps_per_second": 6.979,
|
| 25 |
+
"eval_wer": 0.6078173559646187,
|
| 26 |
+
"step": 258
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.0,
|
| 30 |
+
"grad_norm": 1.1338785886764526,
|
| 31 |
+
"learning_rate": 9.994466403162056e-05,
|
| 32 |
+
"loss": 0.5447,
|
| 33 |
+
"step": 516
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"eval_cer": 0.1309612371199291,
|
| 38 |
+
"eval_loss": 0.5649892687797546,
|
| 39 |
+
"eval_runtime": 29.1014,
|
| 40 |
+
"eval_samples_per_second": 28.384,
|
| 41 |
+
"eval_steps_per_second": 7.113,
|
| 42 |
+
"eval_wer": 0.5978962467128854,
|
| 43 |
+
"step": 516
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"grad_norm": 5.585188388824463,
|
| 48 |
+
"learning_rate": 9.892490118577076e-05,
|
| 49 |
+
"loss": 0.4574,
|
| 50 |
+
"step": 774
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"eval_cer": 0.11218917678342487,
|
| 55 |
+
"eval_loss": 0.45441317558288574,
|
| 56 |
+
"eval_runtime": 29.5398,
|
| 57 |
+
"eval_samples_per_second": 27.962,
|
| 58 |
+
"eval_steps_per_second": 7.007,
|
| 59 |
+
"eval_wer": 0.5168539325842697,
|
| 60 |
+
"step": 774
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"grad_norm": 3.5034706592559814,
|
| 65 |
+
"learning_rate": 9.790513833992096e-05,
|
| 66 |
+
"loss": 0.3712,
|
| 67 |
+
"step": 1032
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 4.0,
|
| 71 |
+
"eval_cer": 0.10859621076623562,
|
| 72 |
+
"eval_loss": 0.4465671181678772,
|
| 73 |
+
"eval_runtime": 29.2068,
|
| 74 |
+
"eval_samples_per_second": 28.281,
|
| 75 |
+
"eval_steps_per_second": 7.087,
|
| 76 |
+
"eval_wer": 0.5049007889074827,
|
| 77 |
+
"step": 1032
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.0,
|
| 81 |
+
"grad_norm": 17.74321174621582,
|
| 82 |
+
"learning_rate": 9.688932806324111e-05,
|
| 83 |
+
"loss": 0.3187,
|
| 84 |
+
"step": 1290
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.0,
|
| 88 |
+
"eval_cer": 0.10036562781937036,
|
| 89 |
+
"eval_loss": 0.44379866123199463,
|
| 90 |
+
"eval_runtime": 29.231,
|
| 91 |
+
"eval_samples_per_second": 28.258,
|
| 92 |
+
"eval_steps_per_second": 7.082,
|
| 93 |
+
"eval_wer": 0.4769304327038011,
|
| 94 |
+
"step": 1290
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"grad_norm": 10.85033893585205,
|
| 99 |
+
"learning_rate": 9.586956521739131e-05,
|
| 100 |
+
"loss": 0.276,
|
| 101 |
+
"step": 1548
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 6.0,
|
| 105 |
+
"eval_cer": 0.1056996786907042,
|
| 106 |
+
"eval_loss": 0.4188626706600189,
|
| 107 |
+
"eval_runtime": 29.2483,
|
| 108 |
+
"eval_samples_per_second": 28.241,
|
| 109 |
+
"eval_steps_per_second": 7.077,
|
| 110 |
+
"eval_wer": 0.4794405928759264,
|
| 111 |
+
"step": 1548
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 7.0,
|
| 115 |
+
"grad_norm": 1.633737564086914,
|
| 116 |
+
"learning_rate": 9.484980237154151e-05,
|
| 117 |
+
"loss": 0.235,
|
| 118 |
+
"step": 1806
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 7.0,
|
| 122 |
+
"eval_cer": 0.1004447680400133,
|
| 123 |
+
"eval_loss": 0.4695667326450348,
|
| 124 |
+
"eval_runtime": 29.6156,
|
| 125 |
+
"eval_samples_per_second": 27.891,
|
| 126 |
+
"eval_steps_per_second": 6.99,
|
| 127 |
+
"eval_wer": 0.4796796557494621,
|
| 128 |
+
"step": 1806
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"grad_norm": 7.645702838897705,
|
| 133 |
+
"learning_rate": 9.383003952569171e-05,
|
| 134 |
+
"loss": 0.2086,
|
| 135 |
+
"step": 2064
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.0,
|
| 139 |
+
"eval_cer": 0.09632947656658067,
|
| 140 |
+
"eval_loss": 0.4218059182167053,
|
| 141 |
+
"eval_runtime": 29.7338,
|
| 142 |
+
"eval_samples_per_second": 27.78,
|
| 143 |
+
"eval_steps_per_second": 6.962,
|
| 144 |
+
"eval_wer": 0.4603155629930672,
|
| 145 |
+
"step": 2064
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 9.0,
|
| 149 |
+
"grad_norm": 5.886650562286377,
|
| 150 |
+
"learning_rate": 9.281027667984191e-05,
|
| 151 |
+
"loss": 0.1883,
|
| 152 |
+
"step": 2322
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.0,
|
| 156 |
+
"eval_cer": 0.09757989205273904,
|
| 157 |
+
"eval_loss": 0.4250072240829468,
|
| 158 |
+
"eval_runtime": 29.3849,
|
| 159 |
+
"eval_samples_per_second": 28.11,
|
| 160 |
+
"eval_steps_per_second": 7.044,
|
| 161 |
+
"eval_wer": 0.45206789385608415,
|
| 162 |
+
"step": 2322
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 10.0,
|
| 166 |
+
"grad_norm": 1.1357783079147339,
|
| 167 |
+
"learning_rate": 9.179051383399209e-05,
|
| 168 |
+
"loss": 0.1671,
|
| 169 |
+
"step": 2580
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 10.0,
|
| 173 |
+
"eval_cer": 0.0972633311701673,
|
| 174 |
+
"eval_loss": 0.45072105526924133,
|
| 175 |
+
"eval_runtime": 29.0789,
|
| 176 |
+
"eval_samples_per_second": 28.405,
|
| 177 |
+
"eval_steps_per_second": 7.119,
|
| 178 |
+
"eval_wer": 0.458642122878317,
|
| 179 |
+
"step": 2580
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 11.0,
|
| 183 |
+
"grad_norm": 0.13285361230373383,
|
| 184 |
+
"learning_rate": 9.077075098814229e-05,
|
| 185 |
+
"loss": 0.1466,
|
| 186 |
+
"step": 2838
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 11.0,
|
| 190 |
+
"eval_cer": 0.09696259833172415,
|
| 191 |
+
"eval_loss": 0.4737579822540283,
|
| 192 |
+
"eval_runtime": 29.2438,
|
| 193 |
+
"eval_samples_per_second": 28.245,
|
| 194 |
+
"eval_steps_per_second": 7.078,
|
| 195 |
+
"eval_wer": 0.4634233803490318,
|
| 196 |
+
"step": 2838
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 12.0,
|
| 200 |
+
"grad_norm": 18.049510955810547,
|
| 201 |
+
"learning_rate": 8.975098814229249e-05,
|
| 202 |
+
"loss": 0.1313,
|
| 203 |
+
"step": 3096
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 12.0,
|
| 207 |
+
"eval_cer": 0.09357539688820653,
|
| 208 |
+
"eval_loss": 0.46005183458328247,
|
| 209 |
+
"eval_runtime": 29.8062,
|
| 210 |
+
"eval_samples_per_second": 27.712,
|
| 211 |
+
"eval_steps_per_second": 6.945,
|
| 212 |
+
"eval_wer": 0.4410710016734401,
|
| 213 |
+
"step": 3096
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 13.0,
|
| 217 |
+
"grad_norm": 1.1974396705627441,
|
| 218 |
+
"learning_rate": 8.873517786561266e-05,
|
| 219 |
+
"loss": 0.1181,
|
| 220 |
+
"step": 3354
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 13.0,
|
| 224 |
+
"eval_cer": 0.08908023235568781,
|
| 225 |
+
"eval_loss": 0.47361627221107483,
|
| 226 |
+
"eval_runtime": 29.1477,
|
| 227 |
+
"eval_samples_per_second": 28.338,
|
| 228 |
+
"eval_steps_per_second": 7.102,
|
| 229 |
+
"eval_wer": 0.4217069089170452,
|
| 230 |
+
"step": 3354
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 14.0,
|
| 234 |
+
"grad_norm": 1.3642749786376953,
|
| 235 |
+
"learning_rate": 8.771541501976284e-05,
|
| 236 |
+
"loss": 0.1026,
|
| 237 |
+
"step": 3612
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 14.0,
|
| 241 |
+
"eval_cer": 0.09121701831304706,
|
| 242 |
+
"eval_loss": 0.4470472037792206,
|
| 243 |
+
"eval_runtime": 29.2097,
|
| 244 |
+
"eval_samples_per_second": 28.278,
|
| 245 |
+
"eval_steps_per_second": 7.087,
|
| 246 |
+
"eval_wer": 0.42780301219220657,
|
| 247 |
+
"step": 3612
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 15.0,
|
| 251 |
+
"grad_norm": 50.79623031616211,
|
| 252 |
+
"learning_rate": 8.669565217391304e-05,
|
| 253 |
+
"loss": 0.092,
|
| 254 |
+
"step": 3870
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.0,
|
| 258 |
+
"eval_cer": 0.09560138653666567,
|
| 259 |
+
"eval_loss": 0.4730050265789032,
|
| 260 |
+
"eval_runtime": 29.5306,
|
| 261 |
+
"eval_samples_per_second": 27.971,
|
| 262 |
+
"eval_steps_per_second": 7.01,
|
| 263 |
+
"eval_wer": 0.4619890031078174,
|
| 264 |
+
"step": 3870
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.0,
|
| 268 |
+
"grad_norm": 0.18782581388950348,
|
| 269 |
+
"learning_rate": 8.567588932806324e-05,
|
| 270 |
+
"loss": 0.083,
|
| 271 |
+
"step": 4128
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 16.0,
|
| 275 |
+
"eval_cer": 0.09370202124123522,
|
| 276 |
+
"eval_loss": 0.533858597278595,
|
| 277 |
+
"eval_runtime": 30.1467,
|
| 278 |
+
"eval_samples_per_second": 27.399,
|
| 279 |
+
"eval_steps_per_second": 6.866,
|
| 280 |
+
"eval_wer": 0.44824288787951233,
|
| 281 |
+
"step": 4128
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 17.0,
|
| 285 |
+
"grad_norm": 0.19636400043964386,
|
| 286 |
+
"learning_rate": 8.465612648221344e-05,
|
| 287 |
+
"loss": 0.0744,
|
| 288 |
+
"step": 4386
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 17.0,
|
| 292 |
+
"eval_cer": 0.09436679909463587,
|
| 293 |
+
"eval_loss": 0.485524982213974,
|
| 294 |
+
"eval_runtime": 30.0117,
|
| 295 |
+
"eval_samples_per_second": 27.523,
|
| 296 |
+
"eval_steps_per_second": 6.897,
|
| 297 |
+
"eval_wer": 0.4508725794884055,
|
| 298 |
+
"step": 4386
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 18.0,
|
| 302 |
+
"grad_norm": 3.5025405883789062,
|
| 303 |
+
"learning_rate": 8.363636363636364e-05,
|
| 304 |
+
"loss": 0.0697,
|
| 305 |
+
"step": 4644
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 18.0,
|
| 309 |
+
"eval_cer": 0.09033064784184619,
|
| 310 |
+
"eval_loss": 0.5221356153488159,
|
| 311 |
+
"eval_runtime": 29.9898,
|
| 312 |
+
"eval_samples_per_second": 27.543,
|
| 313 |
+
"eval_steps_per_second": 6.902,
|
| 314 |
+
"eval_wer": 0.43748505857040404,
|
| 315 |
+
"step": 4644
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 19.0,
|
| 319 |
+
"grad_norm": 1.851936936378479,
|
| 320 |
+
"learning_rate": 8.261660079051384e-05,
|
| 321 |
+
"loss": 0.0602,
|
| 322 |
+
"step": 4902
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 19.0,
|
| 326 |
+
"eval_cer": 0.08944427737064531,
|
| 327 |
+
"eval_loss": 0.51482093334198,
|
| 328 |
+
"eval_runtime": 29.3263,
|
| 329 |
+
"eval_samples_per_second": 28.166,
|
| 330 |
+
"eval_steps_per_second": 7.059,
|
| 331 |
+
"eval_wer": 0.42708582357159935,
|
| 332 |
+
"step": 4902
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 20.0,
|
| 336 |
+
"grad_norm": 0.04486146196722984,
|
| 337 |
+
"learning_rate": 8.159683794466404e-05,
|
| 338 |
+
"loss": 0.056,
|
| 339 |
+
"step": 5160
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 20.0,
|
| 343 |
+
"eval_cer": 0.08983997847385998,
|
| 344 |
+
"eval_loss": 0.5517579317092896,
|
| 345 |
+
"eval_runtime": 29.8422,
|
| 346 |
+
"eval_samples_per_second": 27.679,
|
| 347 |
+
"eval_steps_per_second": 6.936,
|
| 348 |
+
"eval_wer": 0.4312694238584748,
|
| 349 |
+
"step": 5160
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 21.0,
|
| 353 |
+
"grad_norm": 0.09858173131942749,
|
| 354 |
+
"learning_rate": 8.057707509881424e-05,
|
| 355 |
+
"loss": 0.05,
|
| 356 |
+
"step": 5418
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 21.0,
|
| 360 |
+
"eval_cer": 0.09120119026891847,
|
| 361 |
+
"eval_loss": 0.5373868346214294,
|
| 362 |
+
"eval_runtime": 29.6954,
|
| 363 |
+
"eval_samples_per_second": 27.816,
|
| 364 |
+
"eval_steps_per_second": 6.971,
|
| 365 |
+
"eval_wer": 0.431030360984939,
|
| 366 |
+
"step": 5418
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 22.0,
|
| 370 |
+
"grad_norm": 0.7042592167854309,
|
| 371 |
+
"learning_rate": 7.955731225296444e-05,
|
| 372 |
+
"loss": 0.0464,
|
| 373 |
+
"step": 5676
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 22.0,
|
| 377 |
+
"eval_cer": 0.08990329065037433,
|
| 378 |
+
"eval_loss": 0.5166798233985901,
|
| 379 |
+
"eval_runtime": 29.6186,
|
| 380 |
+
"eval_samples_per_second": 27.888,
|
| 381 |
+
"eval_steps_per_second": 6.989,
|
| 382 |
+
"eval_wer": 0.42648816638776,
|
| 383 |
+
"step": 5676
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 23.0,
|
| 387 |
+
"grad_norm": 8.724380493164062,
|
| 388 |
+
"learning_rate": 7.853754940711463e-05,
|
| 389 |
+
"loss": 0.0453,
|
| 390 |
+
"step": 5934
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 23.0,
|
| 394 |
+
"eval_cer": 0.08813054970797259,
|
| 395 |
+
"eval_loss": 0.5781718492507935,
|
| 396 |
+
"eval_runtime": 29.9397,
|
| 397 |
+
"eval_samples_per_second": 27.589,
|
| 398 |
+
"eval_steps_per_second": 6.914,
|
| 399 |
+
"eval_wer": 0.4226631604111881,
|
| 400 |
+
"step": 5934
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 24.0,
|
| 404 |
+
"grad_norm": 0.4805084764957428,
|
| 405 |
+
"learning_rate": 7.751778656126482e-05,
|
| 406 |
+
"loss": 0.0412,
|
| 407 |
+
"step": 6192
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 24.0,
|
| 411 |
+
"eval_cer": 0.09294227512306304,
|
| 412 |
+
"eval_loss": 0.5274640321731567,
|
| 413 |
+
"eval_runtime": 29.0914,
|
| 414 |
+
"eval_samples_per_second": 28.393,
|
| 415 |
+
"eval_steps_per_second": 7.116,
|
| 416 |
+
"eval_wer": 0.43533349270858235,
|
| 417 |
+
"step": 6192
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 25.0,
|
| 421 |
+
"grad_norm": 0.1090458482503891,
|
| 422 |
+
"learning_rate": 7.649802371541502e-05,
|
| 423 |
+
"loss": 0.0369,
|
| 424 |
+
"step": 6450
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 25.0,
|
| 428 |
+
"eval_cer": 0.09186596812231912,
|
| 429 |
+
"eval_loss": 0.6112269759178162,
|
| 430 |
+
"eval_runtime": 29.3026,
|
| 431 |
+
"eval_samples_per_second": 28.189,
|
| 432 |
+
"eval_steps_per_second": 7.064,
|
| 433 |
+
"eval_wer": 0.4233803490317954,
|
| 434 |
+
"step": 6450
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 26.0,
|
| 438 |
+
"grad_norm": 0.0822739452123642,
|
| 439 |
+
"learning_rate": 7.547826086956522e-05,
|
| 440 |
+
"loss": 0.0339,
|
| 441 |
+
"step": 6708
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 26.0,
|
| 445 |
+
"eval_cer": 0.09094794156286108,
|
| 446 |
+
"eval_loss": 0.6159156560897827,
|
| 447 |
+
"eval_runtime": 29.9499,
|
| 448 |
+
"eval_samples_per_second": 27.579,
|
| 449 |
+
"eval_steps_per_second": 6.912,
|
| 450 |
+
"eval_wer": 0.4164475256992589,
|
| 451 |
+
"step": 6708
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 27.0,
|
| 455 |
+
"grad_norm": 0.11275064945220947,
|
| 456 |
+
"learning_rate": 7.445849802371542e-05,
|
| 457 |
+
"loss": 0.0316,
|
| 458 |
+
"step": 6966
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 27.0,
|
| 462 |
+
"eval_cer": 0.08453758369078333,
|
| 463 |
+
"eval_loss": 0.5938315987586975,
|
| 464 |
+
"eval_runtime": 29.2749,
|
| 465 |
+
"eval_samples_per_second": 28.215,
|
| 466 |
+
"eval_steps_per_second": 7.071,
|
| 467 |
+
"eval_wer": 0.4031795362180253,
|
| 468 |
+
"step": 6966
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 28.0,
|
| 472 |
+
"grad_norm": 0.01683381013572216,
|
| 473 |
+
"learning_rate": 7.343873517786562e-05,
|
| 474 |
+
"loss": 0.0263,
|
| 475 |
+
"step": 7224
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 28.0,
|
| 479 |
+
"eval_cer": 0.08710172683961442,
|
| 480 |
+
"eval_loss": 0.5883095860481262,
|
| 481 |
+
"eval_runtime": 29.7516,
|
| 482 |
+
"eval_samples_per_second": 27.763,
|
| 483 |
+
"eval_steps_per_second": 6.958,
|
| 484 |
+
"eval_wer": 0.4093951709299546,
|
| 485 |
+
"step": 7224
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 29.0,
|
| 489 |
+
"grad_norm": 0.0025164771359413862,
|
| 490 |
+
"learning_rate": 7.242292490118577e-05,
|
| 491 |
+
"loss": 0.0268,
|
| 492 |
+
"step": 7482
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 29.0,
|
| 496 |
+
"eval_cer": 0.0871492109720002,
|
| 497 |
+
"eval_loss": 0.6013171076774597,
|
| 498 |
+
"eval_runtime": 29.0617,
|
| 499 |
+
"eval_samples_per_second": 28.422,
|
| 500 |
+
"eval_steps_per_second": 7.123,
|
| 501 |
+
"eval_wer": 0.41477408558450873,
|
| 502 |
+
"step": 7482
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 30.0,
|
| 506 |
+
"grad_norm": 0.012442449107766151,
|
| 507 |
+
"learning_rate": 7.140316205533597e-05,
|
| 508 |
+
"loss": 0.0269,
|
| 509 |
+
"step": 7740
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 30.0,
|
| 513 |
+
"eval_cer": 0.09121701831304706,
|
| 514 |
+
"eval_loss": 0.6137306690216064,
|
| 515 |
+
"eval_runtime": 29.4462,
|
| 516 |
+
"eval_samples_per_second": 28.051,
|
| 517 |
+
"eval_steps_per_second": 7.03,
|
| 518 |
+
"eval_wer": 0.42182644035381306,
|
| 519 |
+
"step": 7740
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 31.0,
|
| 523 |
+
"grad_norm": 2.4109809398651123,
|
| 524 |
+
"learning_rate": 7.038339920948617e-05,
|
| 525 |
+
"loss": 0.0234,
|
| 526 |
+
"step": 7998
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 31.0,
|
| 530 |
+
"eval_cer": 0.08730749141328606,
|
| 531 |
+
"eval_loss": 0.5840467214584351,
|
| 532 |
+
"eval_runtime": 29.7315,
|
| 533 |
+
"eval_samples_per_second": 27.782,
|
| 534 |
+
"eval_steps_per_second": 6.962,
|
| 535 |
+
"eval_wer": 0.40987329667702604,
|
| 536 |
+
"step": 7998
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 32.0,
|
| 540 |
+
"grad_norm": 1.8709559440612793,
|
| 541 |
+
"learning_rate": 6.936363636363637e-05,
|
| 542 |
+
"loss": 0.0229,
|
| 543 |
+
"step": 8256
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 32.0,
|
| 547 |
+
"eval_cer": 0.08613621614777062,
|
| 548 |
+
"eval_loss": 0.628635823726654,
|
| 549 |
+
"eval_runtime": 29.901,
|
| 550 |
+
"eval_samples_per_second": 27.624,
|
| 551 |
+
"eval_steps_per_second": 6.923,
|
| 552 |
+
"eval_wer": 0.4041357877121683,
|
| 553 |
+
"step": 8256
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 33.0,
|
| 557 |
+
"grad_norm": 21.66620445251465,
|
| 558 |
+
"learning_rate": 6.834387351778657e-05,
|
| 559 |
+
"loss": 0.0205,
|
| 560 |
+
"step": 8514
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 33.0,
|
| 564 |
+
"eval_cer": 0.0859146235299704,
|
| 565 |
+
"eval_loss": 0.5922638773918152,
|
| 566 |
+
"eval_runtime": 29.394,
|
| 567 |
+
"eval_samples_per_second": 28.101,
|
| 568 |
+
"eval_steps_per_second": 7.042,
|
| 569 |
+
"eval_wer": 0.3968443700693282,
|
| 570 |
+
"step": 8514
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 34.0,
|
| 574 |
+
"grad_norm": 0.023612063378095627,
|
| 575 |
+
"learning_rate": 6.732411067193675e-05,
|
| 576 |
+
"loss": 0.0196,
|
| 577 |
+
"step": 8772
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 34.0,
|
| 581 |
+
"eval_cer": 0.08950758954715966,
|
| 582 |
+
"eval_loss": 0.6187863945960999,
|
| 583 |
+
"eval_runtime": 29.6115,
|
| 584 |
+
"eval_samples_per_second": 27.895,
|
| 585 |
+
"eval_steps_per_second": 6.991,
|
| 586 |
+
"eval_wer": 0.4049725077695434,
|
| 587 |
+
"step": 8772
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 35.0,
|
| 591 |
+
"grad_norm": 0.011560407467186451,
|
| 592 |
+
"learning_rate": 6.630434782608695e-05,
|
| 593 |
+
"loss": 0.0191,
|
| 594 |
+
"step": 9030
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 35.0,
|
| 598 |
+
"eval_cer": 0.08724417923677172,
|
| 599 |
+
"eval_loss": 0.6254743933677673,
|
| 600 |
+
"eval_runtime": 30.6227,
|
| 601 |
+
"eval_samples_per_second": 26.973,
|
| 602 |
+
"eval_steps_per_second": 6.76,
|
| 603 |
+
"eval_wer": 0.4148936170212766,
|
| 604 |
+
"step": 9030
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 36.0,
|
| 608 |
+
"grad_norm": 0.30145859718322754,
|
| 609 |
+
"learning_rate": 6.528458498023716e-05,
|
| 610 |
+
"loss": 0.0185,
|
| 611 |
+
"step": 9288
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 36.0,
|
| 615 |
+
"eval_cer": 0.08858956298770161,
|
| 616 |
+
"eval_loss": 0.5937672853469849,
|
| 617 |
+
"eval_runtime": 29.5915,
|
| 618 |
+
"eval_samples_per_second": 27.913,
|
| 619 |
+
"eval_steps_per_second": 6.995,
|
| 620 |
+
"eval_wer": 0.41035142242409756,
|
| 621 |
+
"step": 9288
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 37.0,
|
| 625 |
+
"grad_norm": 0.013502486981451511,
|
| 626 |
+
"learning_rate": 6.426482213438736e-05,
|
| 627 |
+
"loss": 0.0163,
|
| 628 |
+
"step": 9546
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 37.0,
|
| 632 |
+
"eval_cer": 0.08642112094208519,
|
| 633 |
+
"eval_loss": 0.6004359126091003,
|
| 634 |
+
"eval_runtime": 29.3597,
|
| 635 |
+
"eval_samples_per_second": 28.134,
|
| 636 |
+
"eval_steps_per_second": 7.05,
|
| 637 |
+
"eval_wer": 0.4076021993784365,
|
| 638 |
+
"step": 9546
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 38.0,
|
| 642 |
+
"grad_norm": 4.638584136962891,
|
| 643 |
+
"learning_rate": 6.324505928853756e-05,
|
| 644 |
+
"loss": 0.0171,
|
| 645 |
+
"step": 9804
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 38.0,
|
| 649 |
+
"eval_cer": 0.08646860507447095,
|
| 650 |
+
"eval_loss": 0.6485220193862915,
|
| 651 |
+
"eval_runtime": 29.8389,
|
| 652 |
+
"eval_samples_per_second": 27.682,
|
| 653 |
+
"eval_steps_per_second": 6.937,
|
| 654 |
+
"eval_wer": 0.4010279703562037,
|
| 655 |
+
"step": 9804
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 39.0,
|
| 659 |
+
"grad_norm": 0.03485848009586334,
|
| 660 |
+
"learning_rate": 6.222529644268775e-05,
|
| 661 |
+
"loss": 0.0163,
|
| 662 |
+
"step": 10062
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 39.0,
|
| 666 |
+
"eval_cer": 0.08507573719115528,
|
| 667 |
+
"eval_loss": 0.6360179781913757,
|
| 668 |
+
"eval_runtime": 29.5268,
|
| 669 |
+
"eval_samples_per_second": 27.975,
|
| 670 |
+
"eval_steps_per_second": 7.011,
|
| 671 |
+
"eval_wer": 0.40353813052832893,
|
| 672 |
+
"step": 10062
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 40.0,
|
| 676 |
+
"grad_norm": 0.004591576289385557,
|
| 677 |
+
"learning_rate": 6.120553359683795e-05,
|
| 678 |
+
"loss": 0.0144,
|
| 679 |
+
"step": 10320
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 40.0,
|
| 683 |
+
"eval_cer": 0.0878773010019152,
|
| 684 |
+
"eval_loss": 0.6229643821716309,
|
| 685 |
+
"eval_runtime": 29.6603,
|
| 686 |
+
"eval_samples_per_second": 27.849,
|
| 687 |
+
"eval_steps_per_second": 6.979,
|
| 688 |
+
"eval_wer": 0.41071001673440116,
|
| 689 |
+
"step": 10320
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 41.0,
|
| 693 |
+
"grad_norm": 0.006813856773078442,
|
| 694 |
+
"learning_rate": 6.0185770750988146e-05,
|
| 695 |
+
"loss": 0.0133,
|
| 696 |
+
"step": 10578
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 41.0,
|
| 700 |
+
"eval_cer": 0.08781398882540084,
|
| 701 |
+
"eval_loss": 0.6123417615890503,
|
| 702 |
+
"eval_runtime": 29.6582,
|
| 703 |
+
"eval_samples_per_second": 27.851,
|
| 704 |
+
"eval_steps_per_second": 6.98,
|
| 705 |
+
"eval_wer": 0.4066459478842936,
|
| 706 |
+
"step": 10578
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 42.0,
|
| 710 |
+
"grad_norm": 0.0054159704595804214,
|
| 711 |
+
"learning_rate": 5.9166007905138345e-05,
|
| 712 |
+
"loss": 0.0131,
|
| 713 |
+
"step": 10836
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 42.0,
|
| 717 |
+
"eval_cer": 0.08721252314851453,
|
| 718 |
+
"eval_loss": 0.6532118916511536,
|
| 719 |
+
"eval_runtime": 29.9909,
|
| 720 |
+
"eval_samples_per_second": 27.542,
|
| 721 |
+
"eval_steps_per_second": 6.902,
|
| 722 |
+
"eval_wer": 0.40365766196509684,
|
| 723 |
+
"step": 10836
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 43.0,
|
| 727 |
+
"grad_norm": 0.07446209341287613,
|
| 728 |
+
"learning_rate": 5.814624505928854e-05,
|
| 729 |
+
"loss": 0.0122,
|
| 730 |
+
"step": 11094
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 43.0,
|
| 734 |
+
"eval_cer": 0.0865160892068567,
|
| 735 |
+
"eval_loss": 0.6612946391105652,
|
| 736 |
+
"eval_runtime": 29.2687,
|
| 737 |
+
"eval_samples_per_second": 28.221,
|
| 738 |
+
"eval_steps_per_second": 7.072,
|
| 739 |
+
"eval_wer": 0.40640688501075783,
|
| 740 |
+
"step": 11094
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 44.0,
|
| 744 |
+
"grad_norm": 5.483983993530273,
|
| 745 |
+
"learning_rate": 5.712648221343874e-05,
|
| 746 |
+
"loss": 0.0127,
|
| 747 |
+
"step": 11352
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 44.0,
|
| 751 |
+
"eval_cer": 0.08488580066161225,
|
| 752 |
+
"eval_loss": 0.627931535243988,
|
| 753 |
+
"eval_runtime": 30.3672,
|
| 754 |
+
"eval_samples_per_second": 27.2,
|
| 755 |
+
"eval_steps_per_second": 6.817,
|
| 756 |
+
"eval_wer": 0.40234281616065026,
|
| 757 |
+
"step": 11352
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 45.0,
|
| 761 |
+
"grad_norm": 0.0012587299570441246,
|
| 762 |
+
"learning_rate": 5.6106719367588936e-05,
|
| 763 |
+
"loss": 0.0115,
|
| 764 |
+
"step": 11610
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 45.0,
|
| 768 |
+
"eval_cer": 0.08453758369078333,
|
| 769 |
+
"eval_loss": 0.6949998140335083,
|
| 770 |
+
"eval_runtime": 29.5908,
|
| 771 |
+
"eval_samples_per_second": 27.914,
|
| 772 |
+
"eval_steps_per_second": 6.995,
|
| 773 |
+
"eval_wer": 0.3993545302414535,
|
| 774 |
+
"step": 11610
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 46.0,
|
| 778 |
+
"grad_norm": 0.0007135935593396425,
|
| 779 |
+
"learning_rate": 5.508695652173913e-05,
|
| 780 |
+
"loss": 0.01,
|
| 781 |
+
"step": 11868
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 46.0,
|
| 785 |
+
"eval_cer": 0.08490162870574083,
|
| 786 |
+
"eval_loss": 0.7085043787956238,
|
| 787 |
+
"eval_runtime": 29.5898,
|
| 788 |
+
"eval_samples_per_second": 27.915,
|
| 789 |
+
"eval_steps_per_second": 6.996,
|
| 790 |
+
"eval_wer": 0.39242170690891703,
|
| 791 |
+
"step": 11868
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 47.0,
|
| 795 |
+
"grad_norm": 0.0029387420509010553,
|
| 796 |
+
"learning_rate": 5.406719367588933e-05,
|
| 797 |
+
"loss": 0.0093,
|
| 798 |
+
"step": 12126
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 47.0,
|
| 802 |
+
"eval_cer": 0.08711755488374301,
|
| 803 |
+
"eval_loss": 0.672882080078125,
|
| 804 |
+
"eval_runtime": 29.3653,
|
| 805 |
+
"eval_samples_per_second": 28.128,
|
| 806 |
+
"eval_steps_per_second": 7.049,
|
| 807 |
+
"eval_wer": 0.41310064546975855,
|
| 808 |
+
"step": 12126
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 48.0,
|
| 812 |
+
"grad_norm": 0.0017440381925553083,
|
| 813 |
+
"learning_rate": 5.304743083003953e-05,
|
| 814 |
+
"loss": 0.0106,
|
| 815 |
+
"step": 12384
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 48.0,
|
| 819 |
+
"eval_cer": 0.08915937257633075,
|
| 820 |
+
"eval_loss": 0.6983070373535156,
|
| 821 |
+
"eval_runtime": 29.8557,
|
| 822 |
+
"eval_samples_per_second": 27.666,
|
| 823 |
+
"eval_steps_per_second": 6.933,
|
| 824 |
+
"eval_wer": 0.40963423380349034,
|
| 825 |
+
"step": 12384
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 49.0,
|
| 829 |
+
"grad_norm": 0.0008035471546463668,
|
| 830 |
+
"learning_rate": 5.202766798418972e-05,
|
| 831 |
+
"loss": 0.0087,
|
| 832 |
+
"step": 12642
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 49.0,
|
| 836 |
+
"eval_cer": 0.08846293863467292,
|
| 837 |
+
"eval_loss": 0.678435742855072,
|
| 838 |
+
"eval_runtime": 30.1041,
|
| 839 |
+
"eval_samples_per_second": 27.438,
|
| 840 |
+
"eval_steps_per_second": 6.876,
|
| 841 |
+
"eval_wer": 0.4139373655271336,
|
| 842 |
+
"step": 12642
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 50.0,
|
| 846 |
+
"grad_norm": 0.002892086748033762,
|
| 847 |
+
"learning_rate": 5.100790513833993e-05,
|
| 848 |
+
"loss": 0.0093,
|
| 849 |
+
"step": 12900
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 50.0,
|
| 853 |
+
"eval_cer": 0.08513904936766964,
|
| 854 |
+
"eval_loss": 0.6546027064323425,
|
| 855 |
+
"eval_runtime": 29.6129,
|
| 856 |
+
"eval_samples_per_second": 27.893,
|
| 857 |
+
"eval_steps_per_second": 6.99,
|
| 858 |
+
"eval_wer": 0.4024623475974181,
|
| 859 |
+
"step": 12900
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 51.0,
|
| 863 |
+
"grad_norm": 0.03883075714111328,
|
| 864 |
+
"learning_rate": 4.998814229249012e-05,
|
| 865 |
+
"loss": 0.0088,
|
| 866 |
+
"step": 13158
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 51.0,
|
| 870 |
+
"eval_cer": 0.08455341173491192,
|
| 871 |
+
"eval_loss": 0.6771977543830872,
|
| 872 |
+
"eval_runtime": 29.4829,
|
| 873 |
+
"eval_samples_per_second": 28.016,
|
| 874 |
+
"eval_steps_per_second": 7.021,
|
| 875 |
+
"eval_wer": 0.39983265598852497,
|
| 876 |
+
"step": 13158
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 52.0,
|
| 880 |
+
"grad_norm": 0.0004216069064568728,
|
| 881 |
+
"learning_rate": 4.8972332015810276e-05,
|
| 882 |
+
"loss": 0.0086,
|
| 883 |
+
"step": 13416
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 52.0,
|
| 887 |
+
"eval_cer": 0.0862786685449279,
|
| 888 |
+
"eval_loss": 0.6763046979904175,
|
| 889 |
+
"eval_runtime": 29.8208,
|
| 890 |
+
"eval_samples_per_second": 27.699,
|
| 891 |
+
"eval_steps_per_second": 6.941,
|
| 892 |
+
"eval_wer": 0.39911546736791775,
|
| 893 |
+
"step": 13416
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 53.0,
|
| 897 |
+
"grad_norm": 0.006624546833336353,
|
| 898 |
+
"learning_rate": 4.7952569169960475e-05,
|
| 899 |
+
"loss": 0.0075,
|
| 900 |
+
"step": 13674
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 53.0,
|
| 904 |
+
"eval_cer": 0.08407857041105431,
|
| 905 |
+
"eval_loss": 0.6990380883216858,
|
| 906 |
+
"eval_runtime": 29.3013,
|
| 907 |
+
"eval_samples_per_second": 28.19,
|
| 908 |
+
"eval_steps_per_second": 7.065,
|
| 909 |
+
"eval_wer": 0.39517092995457803,
|
| 910 |
+
"step": 13674
|
| 911 |
+
}
|
| 912 |
+
],
|
| 913 |
+
"logging_steps": 500,
|
| 914 |
+
"max_steps": 25800,
|
| 915 |
+
"num_input_tokens_seen": 0,
|
| 916 |
+
"num_train_epochs": 100,
|
| 917 |
+
"save_steps": 500,
|
| 918 |
+
"stateful_callbacks": {
|
| 919 |
+
"EarlyStoppingCallback": {
|
| 920 |
+
"args": {
|
| 921 |
+
"early_stopping_patience": 20,
|
| 922 |
+
"early_stopping_threshold": 0.001
|
| 923 |
+
},
|
| 924 |
+
"attributes": {
|
| 925 |
+
"early_stopping_patience_counter": 7
|
| 926 |
+
}
|
| 927 |
+
},
|
| 928 |
+
"TrainerControl": {
|
| 929 |
+
"args": {
|
| 930 |
+
"should_epoch_stop": false,
|
| 931 |
+
"should_evaluate": false,
|
| 932 |
+
"should_log": false,
|
| 933 |
+
"should_save": true,
|
| 934 |
+
"should_training_stop": false
|
| 935 |
+
},
|
| 936 |
+
"attributes": {}
|
| 937 |
+
}
|
| 938 |
+
},
|
| 939 |
+
"total_flos": 2.895592213882363e+19,
|
| 940 |
+
"train_batch_size": 8,
|
| 941 |
+
"trial_name": null,
|
| 942 |
+
"trial_params": null
|
| 943 |
+
}
|
checkpoint-13674/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:badc6886eb2995443a897620ab2c36decc3b66898e976799a420ad7fb69a195c
|
| 3 |
+
size 5304
|