Upload 12 files
Browse files- added_tokens.json +4 -0
- config.json +116 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +10 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +48 -0
- trainer_state.json +491 -0
- training_args.bin +3 -0
- vocab.json +40 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 39,
|
| 3 |
+
"<s>": 38
|
| 4 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-greek",
|
| 3 |
+
"activation_dropout": 0.05,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": true,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_size": 1024,
|
| 59 |
+
"initializer_range": 0.02,
|
| 60 |
+
"intermediate_size": 4096,
|
| 61 |
+
"layer_norm_eps": 1e-05,
|
| 62 |
+
"layerdrop": 0.1,
|
| 63 |
+
"mask_channel_length": 10,
|
| 64 |
+
"mask_channel_min_space": 1,
|
| 65 |
+
"mask_channel_other": 0.0,
|
| 66 |
+
"mask_channel_prob": 0.0,
|
| 67 |
+
"mask_channel_selection": "static",
|
| 68 |
+
"mask_feature_length": 10,
|
| 69 |
+
"mask_feature_min_masks": 0,
|
| 70 |
+
"mask_feature_prob": 0.0,
|
| 71 |
+
"mask_time_length": 10,
|
| 72 |
+
"mask_time_min_masks": 2,
|
| 73 |
+
"mask_time_min_space": 1,
|
| 74 |
+
"mask_time_other": 0.0,
|
| 75 |
+
"mask_time_prob": 0.05,
|
| 76 |
+
"mask_time_selection": "static",
|
| 77 |
+
"model_type": "wav2vec2",
|
| 78 |
+
"num_adapter_layers": 3,
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_codevector_groups": 2,
|
| 81 |
+
"num_codevectors_per_group": 320,
|
| 82 |
+
"num_conv_pos_embedding_groups": 16,
|
| 83 |
+
"num_conv_pos_embeddings": 128,
|
| 84 |
+
"num_feat_extract_layers": 7,
|
| 85 |
+
"num_hidden_layers": 24,
|
| 86 |
+
"num_negatives": 100,
|
| 87 |
+
"output_hidden_size": 1024,
|
| 88 |
+
"pad_token_id": 37,
|
| 89 |
+
"proj_codevector_dim": 768,
|
| 90 |
+
"tdnn_dilation": [
|
| 91 |
+
1,
|
| 92 |
+
2,
|
| 93 |
+
3,
|
| 94 |
+
1,
|
| 95 |
+
1
|
| 96 |
+
],
|
| 97 |
+
"tdnn_dim": [
|
| 98 |
+
512,
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
1500
|
| 103 |
+
],
|
| 104 |
+
"tdnn_kernel": [
|
| 105 |
+
5,
|
| 106 |
+
3,
|
| 107 |
+
3,
|
| 108 |
+
1,
|
| 109 |
+
1
|
| 110 |
+
],
|
| 111 |
+
"torch_dtype": "float32",
|
| 112 |
+
"transformers_version": "4.37.2",
|
| 113 |
+
"use_weighted_layer_sum": false,
|
| 114 |
+
"vocab_size": 41,
|
| 115 |
+
"xvector_output_dim": 512
|
| 116 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:539cf50570693a7394729a7b81ab30c69d2728b0f397a56891b465393ffc7c21
|
| 3 |
+
size 1261975580
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:050695608042d9ad1055e3ca97c8fe74b161450f691b270f01ab4a7cf0d24c09
|
| 3 |
+
size 2490495926
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"processor_class": "Wav2Vec2Processor",
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8a8feff87f4c7dd11bc89821bc8a947b0cb9d877ca95ee20a1ff571b636f72a
|
| 3 |
+
size 14308
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:403b8a9aae6815ef1a4c55e7c12a79b2814300fc437235dbe1a90cbe1d4c3c72
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"36": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": true,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": true,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": false
|
| 10 |
+
},
|
| 11 |
+
"37": {
|
| 12 |
+
"content": "[PAD]",
|
| 13 |
+
"lstrip": true,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": true,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": false
|
| 18 |
+
},
|
| 19 |
+
"38": {
|
| 20 |
+
"content": "<s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"39": {
|
| 28 |
+
"content": "</s>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"bos_token": "<s>",
|
| 37 |
+
"clean_up_tokenization_spaces": true,
|
| 38 |
+
"do_lower_case": false,
|
| 39 |
+
"eos_token": "</s>",
|
| 40 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 41 |
+
"pad_token": "[PAD]",
|
| 42 |
+
"processor_class": "Wav2Vec2Processor",
|
| 43 |
+
"replace_word_delimiter_char": " ",
|
| 44 |
+
"target_lang": null,
|
| 45 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 46 |
+
"unk_token": "[UNK]",
|
| 47 |
+
"word_delimiter_token": "|"
|
| 48 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,491 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.31333859510655093,
|
| 3 |
+
"best_model_checkpoint": "xls-r-greek-cretan/checkpoint-3894",
|
| 4 |
+
"epoch": 35.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4130,
|
| 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 |
+
"eval_cer": 0.9992756772417789,
|
| 14 |
+
"eval_loss": 3.404292583465576,
|
| 15 |
+
"eval_runtime": 6.277,
|
| 16 |
+
"eval_samples_per_second": 37.597,
|
| 17 |
+
"eval_steps_per_second": 4.779,
|
| 18 |
+
"eval_wer": 1.0,
|
| 19 |
+
"step": 118
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 1.69,
|
| 23 |
+
"learning_rate": 0.00011999999999999999,
|
| 24 |
+
"loss": 5.627,
|
| 25 |
+
"step": 200
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_cer": 0.8620889468347096,
|
| 30 |
+
"eval_loss": 2.5985116958618164,
|
| 31 |
+
"eval_runtime": 6.1017,
|
| 32 |
+
"eval_samples_per_second": 38.678,
|
| 33 |
+
"eval_steps_per_second": 4.917,
|
| 34 |
+
"eval_wer": 0.9952644041041832,
|
| 35 |
+
"step": 236
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 3.0,
|
| 39 |
+
"eval_cer": 0.17398232652469942,
|
| 40 |
+
"eval_loss": 0.6798810958862305,
|
| 41 |
+
"eval_runtime": 5.9946,
|
| 42 |
+
"eval_samples_per_second": 39.369,
|
| 43 |
+
"eval_steps_per_second": 5.004,
|
| 44 |
+
"eval_wer": 0.5603788476716653,
|
| 45 |
+
"step": 354
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.39,
|
| 49 |
+
"learning_rate": 0.00023999999999999998,
|
| 50 |
+
"loss": 1.9128,
|
| 51 |
+
"step": 400
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 4.0,
|
| 55 |
+
"eval_cer": 0.15094886281326958,
|
| 56 |
+
"eval_loss": 0.5401029586791992,
|
| 57 |
+
"eval_runtime": 5.974,
|
| 58 |
+
"eval_samples_per_second": 39.504,
|
| 59 |
+
"eval_steps_per_second": 5.022,
|
| 60 |
+
"eval_wer": 0.5027624309392266,
|
| 61 |
+
"step": 472
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.0,
|
| 65 |
+
"eval_cer": 0.13906996957844414,
|
| 66 |
+
"eval_loss": 0.49666497111320496,
|
| 67 |
+
"eval_runtime": 5.4532,
|
| 68 |
+
"eval_samples_per_second": 43.277,
|
| 69 |
+
"eval_steps_per_second": 5.501,
|
| 70 |
+
"eval_wer": 0.48066298342541436,
|
| 71 |
+
"step": 590
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 5.08,
|
| 75 |
+
"learning_rate": 0.0002917355371900826,
|
| 76 |
+
"loss": 1.029,
|
| 77 |
+
"step": 600
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.0,
|
| 81 |
+
"eval_cer": 0.1282051282051282,
|
| 82 |
+
"eval_loss": 0.5171706080436707,
|
| 83 |
+
"eval_runtime": 5.3095,
|
| 84 |
+
"eval_samples_per_second": 44.449,
|
| 85 |
+
"eval_steps_per_second": 5.65,
|
| 86 |
+
"eval_wer": 0.4451460142067877,
|
| 87 |
+
"step": 708
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 6.78,
|
| 91 |
+
"learning_rate": 0.0002752066115702479,
|
| 92 |
+
"loss": 0.8154,
|
| 93 |
+
"step": 800
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 7.0,
|
| 97 |
+
"eval_cer": 0.13066782558307982,
|
| 98 |
+
"eval_loss": 0.5484737753868103,
|
| 99 |
+
"eval_runtime": 6.0637,
|
| 100 |
+
"eval_samples_per_second": 38.92,
|
| 101 |
+
"eval_steps_per_second": 4.947,
|
| 102 |
+
"eval_wer": 0.43804262036306235,
|
| 103 |
+
"step": 826
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 8.0,
|
| 107 |
+
"eval_cer": 0.11299435028248588,
|
| 108 |
+
"eval_loss": 0.455568790435791,
|
| 109 |
+
"eval_runtime": 5.7805,
|
| 110 |
+
"eval_samples_per_second": 40.827,
|
| 111 |
+
"eval_steps_per_second": 5.19,
|
| 112 |
+
"eval_wer": 0.40331491712707185,
|
| 113 |
+
"step": 944
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 8.47,
|
| 117 |
+
"learning_rate": 0.00025867768595041324,
|
| 118 |
+
"loss": 0.68,
|
| 119 |
+
"step": 1000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 9.0,
|
| 123 |
+
"eval_cer": 0.11806460959003331,
|
| 124 |
+
"eval_loss": 0.47916799783706665,
|
| 125 |
+
"eval_runtime": 5.4838,
|
| 126 |
+
"eval_samples_per_second": 43.036,
|
| 127 |
+
"eval_steps_per_second": 5.471,
|
| 128 |
+
"eval_wer": 0.4112075769534333,
|
| 129 |
+
"step": 1062
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 10.0,
|
| 133 |
+
"eval_cer": 0.10922787193973635,
|
| 134 |
+
"eval_loss": 0.43157267570495605,
|
| 135 |
+
"eval_runtime": 5.6241,
|
| 136 |
+
"eval_samples_per_second": 41.962,
|
| 137 |
+
"eval_steps_per_second": 5.334,
|
| 138 |
+
"eval_wer": 0.39068666140489344,
|
| 139 |
+
"step": 1180
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 10.17,
|
| 143 |
+
"learning_rate": 0.0002421487603305785,
|
| 144 |
+
"loss": 0.5853,
|
| 145 |
+
"step": 1200
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 11.0,
|
| 149 |
+
"eval_cer": 0.10705490366507316,
|
| 150 |
+
"eval_loss": 0.46444636583328247,
|
| 151 |
+
"eval_runtime": 5.4324,
|
| 152 |
+
"eval_samples_per_second": 43.443,
|
| 153 |
+
"eval_steps_per_second": 5.522,
|
| 154 |
+
"eval_wer": 0.36937647987371747,
|
| 155 |
+
"step": 1298
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 11.86,
|
| 159 |
+
"learning_rate": 0.00022561983471074378,
|
| 160 |
+
"loss": 0.534,
|
| 161 |
+
"step": 1400
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 12.0,
|
| 165 |
+
"eval_cer": 0.11313921483413009,
|
| 166 |
+
"eval_loss": 0.5147783756256104,
|
| 167 |
+
"eval_runtime": 5.7969,
|
| 168 |
+
"eval_samples_per_second": 40.711,
|
| 169 |
+
"eval_steps_per_second": 5.175,
|
| 170 |
+
"eval_wer": 0.3764798737174428,
|
| 171 |
+
"step": 1416
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 13.0,
|
| 175 |
+
"eval_cer": 0.10893814283644793,
|
| 176 |
+
"eval_loss": 0.49593451619148254,
|
| 177 |
+
"eval_runtime": 5.8062,
|
| 178 |
+
"eval_samples_per_second": 40.647,
|
| 179 |
+
"eval_steps_per_second": 5.167,
|
| 180 |
+
"eval_wer": 0.3772691397000789,
|
| 181 |
+
"step": 1534
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 13.56,
|
| 185 |
+
"learning_rate": 0.0002090909090909091,
|
| 186 |
+
"loss": 0.4826,
|
| 187 |
+
"step": 1600
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 14.0,
|
| 191 |
+
"eval_cer": 0.10850354918151528,
|
| 192 |
+
"eval_loss": 0.48721909523010254,
|
| 193 |
+
"eval_runtime": 5.799,
|
| 194 |
+
"eval_samples_per_second": 40.697,
|
| 195 |
+
"eval_steps_per_second": 5.173,
|
| 196 |
+
"eval_wer": 0.3788476716653512,
|
| 197 |
+
"step": 1652
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 15.0,
|
| 201 |
+
"eval_cer": 0.10256410256410256,
|
| 202 |
+
"eval_loss": 0.45876312255859375,
|
| 203 |
+
"eval_runtime": 5.8742,
|
| 204 |
+
"eval_samples_per_second": 40.176,
|
| 205 |
+
"eval_steps_per_second": 5.107,
|
| 206 |
+
"eval_wer": 0.36306235201262826,
|
| 207 |
+
"step": 1770
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.25,
|
| 211 |
+
"learning_rate": 0.00019256198347107438,
|
| 212 |
+
"loss": 0.4211,
|
| 213 |
+
"step": 1800
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 16.0,
|
| 217 |
+
"eval_cer": 0.10155005070259307,
|
| 218 |
+
"eval_loss": 0.4505126476287842,
|
| 219 |
+
"eval_runtime": 5.2544,
|
| 220 |
+
"eval_samples_per_second": 44.915,
|
| 221 |
+
"eval_steps_per_second": 5.71,
|
| 222 |
+
"eval_wer": 0.3606945540647198,
|
| 223 |
+
"step": 1888
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.95,
|
| 227 |
+
"learning_rate": 0.00017603305785123967,
|
| 228 |
+
"loss": 0.396,
|
| 229 |
+
"step": 2000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 17.0,
|
| 233 |
+
"eval_cer": 0.09923221787628567,
|
| 234 |
+
"eval_loss": 0.46898314356803894,
|
| 235 |
+
"eval_runtime": 5.9134,
|
| 236 |
+
"eval_samples_per_second": 39.91,
|
| 237 |
+
"eval_steps_per_second": 5.073,
|
| 238 |
+
"eval_wer": 0.3606945540647198,
|
| 239 |
+
"step": 2006
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 18.0,
|
| 243 |
+
"eval_cer": 0.09937708242792989,
|
| 244 |
+
"eval_loss": 0.45210039615631104,
|
| 245 |
+
"eval_runtime": 6.1856,
|
| 246 |
+
"eval_samples_per_second": 38.153,
|
| 247 |
+
"eval_steps_per_second": 4.85,
|
| 248 |
+
"eval_wer": 0.3638516179952644,
|
| 249 |
+
"step": 2124
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 18.64,
|
| 253 |
+
"learning_rate": 0.00015950413223140495,
|
| 254 |
+
"loss": 0.3619,
|
| 255 |
+
"step": 2200
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 19.0,
|
| 259 |
+
"eval_cer": 0.09937708242792989,
|
| 260 |
+
"eval_loss": 0.4832761585712433,
|
| 261 |
+
"eval_runtime": 6.2572,
|
| 262 |
+
"eval_samples_per_second": 37.716,
|
| 263 |
+
"eval_steps_per_second": 4.794,
|
| 264 |
+
"eval_wer": 0.35438042620363064,
|
| 265 |
+
"step": 2242
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 20.0,
|
| 269 |
+
"eval_cer": 0.09097493843256556,
|
| 270 |
+
"eval_loss": 0.44832131266593933,
|
| 271 |
+
"eval_runtime": 5.7413,
|
| 272 |
+
"eval_samples_per_second": 41.106,
|
| 273 |
+
"eval_steps_per_second": 5.225,
|
| 274 |
+
"eval_wer": 0.3346487766377269,
|
| 275 |
+
"step": 2360
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 20.34,
|
| 279 |
+
"learning_rate": 0.00014297520661157024,
|
| 280 |
+
"loss": 0.33,
|
| 281 |
+
"step": 2400
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 21.0,
|
| 285 |
+
"eval_cer": 0.09865275966970882,
|
| 286 |
+
"eval_loss": 0.4619905650615692,
|
| 287 |
+
"eval_runtime": 5.4696,
|
| 288 |
+
"eval_samples_per_second": 43.148,
|
| 289 |
+
"eval_steps_per_second": 5.485,
|
| 290 |
+
"eval_wer": 0.3425414364640884,
|
| 291 |
+
"step": 2478
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 22.0,
|
| 295 |
+
"eval_cer": 0.09532087498189193,
|
| 296 |
+
"eval_loss": 0.4783581793308258,
|
| 297 |
+
"eval_runtime": 5.7494,
|
| 298 |
+
"eval_samples_per_second": 41.048,
|
| 299 |
+
"eval_steps_per_second": 5.218,
|
| 300 |
+
"eval_wer": 0.33859510655090763,
|
| 301 |
+
"step": 2596
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 22.03,
|
| 305 |
+
"learning_rate": 0.00012644628099173552,
|
| 306 |
+
"loss": 0.3199,
|
| 307 |
+
"step": 2600
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 23.0,
|
| 311 |
+
"eval_cer": 0.09633492684340142,
|
| 312 |
+
"eval_loss": 0.5089753866195679,
|
| 313 |
+
"eval_runtime": 6.2898,
|
| 314 |
+
"eval_samples_per_second": 37.521,
|
| 315 |
+
"eval_steps_per_second": 4.77,
|
| 316 |
+
"eval_wer": 0.34333070244672453,
|
| 317 |
+
"step": 2714
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 23.73,
|
| 321 |
+
"learning_rate": 0.00010991735537190081,
|
| 322 |
+
"loss": 0.2793,
|
| 323 |
+
"step": 2800
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 24.0,
|
| 327 |
+
"eval_cer": 0.0948862813269593,
|
| 328 |
+
"eval_loss": 0.5048560500144958,
|
| 329 |
+
"eval_runtime": 6.0494,
|
| 330 |
+
"eval_samples_per_second": 39.012,
|
| 331 |
+
"eval_steps_per_second": 4.959,
|
| 332 |
+
"eval_wer": 0.33859510655090763,
|
| 333 |
+
"step": 2832
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 25.0,
|
| 337 |
+
"eval_cer": 0.09198899029407503,
|
| 338 |
+
"eval_loss": 0.5005324482917786,
|
| 339 |
+
"eval_runtime": 6.5065,
|
| 340 |
+
"eval_samples_per_second": 36.271,
|
| 341 |
+
"eval_steps_per_second": 4.611,
|
| 342 |
+
"eval_wer": 0.324388318863457,
|
| 343 |
+
"step": 2950
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 25.42,
|
| 347 |
+
"learning_rate": 9.338842975206611e-05,
|
| 348 |
+
"loss": 0.2769,
|
| 349 |
+
"step": 3000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 26.0,
|
| 353 |
+
"eval_cer": 0.08851224105461393,
|
| 354 |
+
"eval_loss": 0.503860354423523,
|
| 355 |
+
"eval_runtime": 5.4619,
|
| 356 |
+
"eval_samples_per_second": 43.208,
|
| 357 |
+
"eval_steps_per_second": 5.493,
|
| 358 |
+
"eval_wer": 0.3267561168113654,
|
| 359 |
+
"step": 3068
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 27.0,
|
| 363 |
+
"eval_cer": 0.0938722294654498,
|
| 364 |
+
"eval_loss": 0.5316025614738464,
|
| 365 |
+
"eval_runtime": 5.994,
|
| 366 |
+
"eval_samples_per_second": 39.373,
|
| 367 |
+
"eval_steps_per_second": 5.005,
|
| 368 |
+
"eval_wer": 0.32991318074191,
|
| 369 |
+
"step": 3186
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 27.12,
|
| 373 |
+
"learning_rate": 7.68595041322314e-05,
|
| 374 |
+
"loss": 0.2329,
|
| 375 |
+
"step": 3200
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 28.0,
|
| 379 |
+
"eval_cer": 0.09256844850065189,
|
| 380 |
+
"eval_loss": 0.5238826274871826,
|
| 381 |
+
"eval_runtime": 5.6891,
|
| 382 |
+
"eval_samples_per_second": 41.483,
|
| 383 |
+
"eval_steps_per_second": 5.273,
|
| 384 |
+
"eval_wer": 0.3314917127071823,
|
| 385 |
+
"step": 3304
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 28.81,
|
| 389 |
+
"learning_rate": 6.033057851239669e-05,
|
| 390 |
+
"loss": 0.2327,
|
| 391 |
+
"step": 3400
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 29.0,
|
| 395 |
+
"eval_cer": 0.09184412574243082,
|
| 396 |
+
"eval_loss": 0.4961460530757904,
|
| 397 |
+
"eval_runtime": 5.7688,
|
| 398 |
+
"eval_samples_per_second": 40.91,
|
| 399 |
+
"eval_steps_per_second": 5.2,
|
| 400 |
+
"eval_wer": 0.3228097868981847,
|
| 401 |
+
"step": 3422
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 30.0,
|
| 405 |
+
"eval_cer": 0.09068520932927712,
|
| 406 |
+
"eval_loss": 0.4970700740814209,
|
| 407 |
+
"eval_runtime": 6.2622,
|
| 408 |
+
"eval_samples_per_second": 37.686,
|
| 409 |
+
"eval_steps_per_second": 4.791,
|
| 410 |
+
"eval_wer": 0.3267561168113654,
|
| 411 |
+
"step": 3540
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 30.51,
|
| 415 |
+
"learning_rate": 4.380165289256198e-05,
|
| 416 |
+
"loss": 0.2087,
|
| 417 |
+
"step": 3600
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 31.0,
|
| 421 |
+
"eval_cer": 0.09083007388092133,
|
| 422 |
+
"eval_loss": 0.5133862495422363,
|
| 423 |
+
"eval_runtime": 5.6526,
|
| 424 |
+
"eval_samples_per_second": 41.751,
|
| 425 |
+
"eval_steps_per_second": 5.307,
|
| 426 |
+
"eval_wer": 0.3259668508287293,
|
| 427 |
+
"step": 3658
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 32.0,
|
| 431 |
+
"eval_cer": 0.08851224105461393,
|
| 432 |
+
"eval_loss": 0.522043764591217,
|
| 433 |
+
"eval_runtime": 5.6904,
|
| 434 |
+
"eval_samples_per_second": 41.474,
|
| 435 |
+
"eval_steps_per_second": 5.272,
|
| 436 |
+
"eval_wer": 0.3212312549329124,
|
| 437 |
+
"step": 3776
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 32.2,
|
| 441 |
+
"learning_rate": 2.727272727272727e-05,
|
| 442 |
+
"loss": 0.1856,
|
| 443 |
+
"step": 3800
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 33.0,
|
| 447 |
+
"eval_cer": 0.08836737650296972,
|
| 448 |
+
"eval_loss": 0.5135778188705444,
|
| 449 |
+
"eval_runtime": 6.0356,
|
| 450 |
+
"eval_samples_per_second": 39.101,
|
| 451 |
+
"eval_steps_per_second": 4.971,
|
| 452 |
+
"eval_wer": 0.31333859510655093,
|
| 453 |
+
"step": 3894
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.9,
|
| 457 |
+
"learning_rate": 1.0743801652892562e-05,
|
| 458 |
+
"loss": 0.1803,
|
| 459 |
+
"step": 4000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 34.0,
|
| 463 |
+
"eval_cer": 0.08749818919310445,
|
| 464 |
+
"eval_loss": 0.5257639288902283,
|
| 465 |
+
"eval_runtime": 5.8083,
|
| 466 |
+
"eval_samples_per_second": 40.631,
|
| 467 |
+
"eval_steps_per_second": 5.165,
|
| 468 |
+
"eval_wer": 0.31333859510655093,
|
| 469 |
+
"step": 4012
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 35.0,
|
| 473 |
+
"eval_cer": 0.08764305374474866,
|
| 474 |
+
"eval_loss": 0.5268439650535583,
|
| 475 |
+
"eval_runtime": 6.0191,
|
| 476 |
+
"eval_samples_per_second": 39.208,
|
| 477 |
+
"eval_steps_per_second": 4.984,
|
| 478 |
+
"eval_wer": 0.3164956590370955,
|
| 479 |
+
"step": 4130
|
| 480 |
+
}
|
| 481 |
+
],
|
| 482 |
+
"logging_steps": 200,
|
| 483 |
+
"max_steps": 4130,
|
| 484 |
+
"num_input_tokens_seen": 0,
|
| 485 |
+
"num_train_epochs": 35,
|
| 486 |
+
"save_steps": 500,
|
| 487 |
+
"total_flos": 4.2609001975048284e+18,
|
| 488 |
+
"train_batch_size": 8,
|
| 489 |
+
"trial_name": null,
|
| 490 |
+
"trial_params": null
|
| 491 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:150052b7abe6a66b23fea1be7ce76627fa639b95da53afcc922b25f481abf433
|
| 3 |
+
size 4728
|
vocab.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 37,
|
| 3 |
+
"[UNK]": 36,
|
| 4 |
+
"o": 1,
|
| 5 |
+
"|": 0,
|
| 6 |
+
"ΐ": 2,
|
| 7 |
+
"ά": 3,
|
| 8 |
+
"έ": 4,
|
| 9 |
+
"ή": 5,
|
| 10 |
+
"ί": 6,
|
| 11 |
+
"α": 7,
|
| 12 |
+
"β": 8,
|
| 13 |
+
"γ": 9,
|
| 14 |
+
"δ": 10,
|
| 15 |
+
"ε": 11,
|
| 16 |
+
"ζ": 12,
|
| 17 |
+
"η": 13,
|
| 18 |
+
"θ": 14,
|
| 19 |
+
"ι": 15,
|
| 20 |
+
"κ": 16,
|
| 21 |
+
"λ": 17,
|
| 22 |
+
"μ": 18,
|
| 23 |
+
"ν": 19,
|
| 24 |
+
"ξ": 20,
|
| 25 |
+
"ο": 21,
|
| 26 |
+
"π": 22,
|
| 27 |
+
"ρ": 23,
|
| 28 |
+
"ς": 24,
|
| 29 |
+
"σ": 25,
|
| 30 |
+
"τ": 26,
|
| 31 |
+
"υ": 27,
|
| 32 |
+
"φ": 28,
|
| 33 |
+
"χ": 29,
|
| 34 |
+
"ψ": 30,
|
| 35 |
+
"ω": 31,
|
| 36 |
+
"ϊ": 32,
|
| 37 |
+
"ό": 33,
|
| 38 |
+
"ύ": 34,
|
| 39 |
+
"ώ": 35
|
| 40 |
+
}
|