Commit
•
11e3a02
1
Parent(s):
f3d83d4
Added supporting files
Browse files- config.json +178 -0
- generation_config.json +13 -0
- preprocessor_config.json +22 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +581 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/trocr-base-handwritten",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"VisionEncoderDecoderModel"
|
| 5 |
+
],
|
| 6 |
+
"decoder": {
|
| 7 |
+
"_name_or_path": "",
|
| 8 |
+
"activation_dropout": 0.0,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"add_cross_attention": true,
|
| 11 |
+
"architectures": null,
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bad_words_ids": null,
|
| 14 |
+
"begin_suppress_tokens": null,
|
| 15 |
+
"bos_token_id": 0,
|
| 16 |
+
"chunk_size_feed_forward": 0,
|
| 17 |
+
"classifier_dropout": 0.0,
|
| 18 |
+
"cross_attention_hidden_size": 768,
|
| 19 |
+
"d_model": 1024,
|
| 20 |
+
"decoder_attention_heads": 16,
|
| 21 |
+
"decoder_ffn_dim": 4096,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 12,
|
| 24 |
+
"decoder_start_token_id": 2,
|
| 25 |
+
"diversity_penalty": 0.0,
|
| 26 |
+
"do_sample": false,
|
| 27 |
+
"dropout": 0.1,
|
| 28 |
+
"early_stopping": false,
|
| 29 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 30 |
+
"eos_token_id": 2,
|
| 31 |
+
"exponential_decay_length_penalty": null,
|
| 32 |
+
"finetuning_task": null,
|
| 33 |
+
"forced_bos_token_id": null,
|
| 34 |
+
"forced_eos_token_id": null,
|
| 35 |
+
"id2label": {
|
| 36 |
+
"0": "LABEL_0",
|
| 37 |
+
"1": "LABEL_1"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_decoder": true,
|
| 41 |
+
"is_encoder_decoder": false,
|
| 42 |
+
"label2id": {
|
| 43 |
+
"LABEL_0": 0,
|
| 44 |
+
"LABEL_1": 1
|
| 45 |
+
},
|
| 46 |
+
"layernorm_embedding": true,
|
| 47 |
+
"length_penalty": 1.0,
|
| 48 |
+
"max_length": 20,
|
| 49 |
+
"max_position_embeddings": 512,
|
| 50 |
+
"min_length": 0,
|
| 51 |
+
"model_type": "trocr",
|
| 52 |
+
"no_repeat_ngram_size": 0,
|
| 53 |
+
"num_beam_groups": 1,
|
| 54 |
+
"num_beams": 1,
|
| 55 |
+
"num_return_sequences": 1,
|
| 56 |
+
"output_attentions": false,
|
| 57 |
+
"output_hidden_states": false,
|
| 58 |
+
"output_scores": false,
|
| 59 |
+
"pad_token_id": 1,
|
| 60 |
+
"prefix": null,
|
| 61 |
+
"problem_type": null,
|
| 62 |
+
"pruned_heads": {},
|
| 63 |
+
"remove_invalid_values": false,
|
| 64 |
+
"repetition_penalty": 1.0,
|
| 65 |
+
"return_dict": true,
|
| 66 |
+
"return_dict_in_generate": false,
|
| 67 |
+
"scale_embedding": false,
|
| 68 |
+
"sep_token_id": null,
|
| 69 |
+
"suppress_tokens": null,
|
| 70 |
+
"task_specific_params": null,
|
| 71 |
+
"temperature": 1.0,
|
| 72 |
+
"tf_legacy_loss": false,
|
| 73 |
+
"tie_encoder_decoder": false,
|
| 74 |
+
"tie_word_embeddings": true,
|
| 75 |
+
"tokenizer_class": null,
|
| 76 |
+
"top_k": 50,
|
| 77 |
+
"top_p": 1.0,
|
| 78 |
+
"torch_dtype": null,
|
| 79 |
+
"torchscript": false,
|
| 80 |
+
"typical_p": 1.0,
|
| 81 |
+
"use_bfloat16": false,
|
| 82 |
+
"use_cache": false,
|
| 83 |
+
"use_learned_position_embeddings": true,
|
| 84 |
+
"vocab_size": 50265
|
| 85 |
+
},
|
| 86 |
+
"decoder_start_token_id": 0,
|
| 87 |
+
"early_stopping": true,
|
| 88 |
+
"encoder": {
|
| 89 |
+
"_name_or_path": "",
|
| 90 |
+
"add_cross_attention": false,
|
| 91 |
+
"architectures": null,
|
| 92 |
+
"attention_probs_dropout_prob": 0.0,
|
| 93 |
+
"bad_words_ids": null,
|
| 94 |
+
"begin_suppress_tokens": null,
|
| 95 |
+
"bos_token_id": null,
|
| 96 |
+
"chunk_size_feed_forward": 0,
|
| 97 |
+
"cross_attention_hidden_size": null,
|
| 98 |
+
"decoder_start_token_id": null,
|
| 99 |
+
"diversity_penalty": 0.0,
|
| 100 |
+
"do_sample": false,
|
| 101 |
+
"early_stopping": false,
|
| 102 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 103 |
+
"encoder_stride": 16,
|
| 104 |
+
"eos_token_id": null,
|
| 105 |
+
"exponential_decay_length_penalty": null,
|
| 106 |
+
"finetuning_task": null,
|
| 107 |
+
"forced_bos_token_id": null,
|
| 108 |
+
"forced_eos_token_id": null,
|
| 109 |
+
"hidden_act": "gelu",
|
| 110 |
+
"hidden_dropout_prob": 0.0,
|
| 111 |
+
"hidden_size": 768,
|
| 112 |
+
"id2label": {
|
| 113 |
+
"0": "LABEL_0",
|
| 114 |
+
"1": "LABEL_1"
|
| 115 |
+
},
|
| 116 |
+
"image_size": 384,
|
| 117 |
+
"initializer_range": 0.02,
|
| 118 |
+
"intermediate_size": 3072,
|
| 119 |
+
"is_decoder": false,
|
| 120 |
+
"is_encoder_decoder": false,
|
| 121 |
+
"label2id": {
|
| 122 |
+
"LABEL_0": 0,
|
| 123 |
+
"LABEL_1": 1
|
| 124 |
+
},
|
| 125 |
+
"layer_norm_eps": 1e-12,
|
| 126 |
+
"length_penalty": 1.0,
|
| 127 |
+
"max_length": 20,
|
| 128 |
+
"min_length": 0,
|
| 129 |
+
"model_type": "vit",
|
| 130 |
+
"no_repeat_ngram_size": 0,
|
| 131 |
+
"num_attention_heads": 12,
|
| 132 |
+
"num_beam_groups": 1,
|
| 133 |
+
"num_beams": 1,
|
| 134 |
+
"num_channels": 3,
|
| 135 |
+
"num_hidden_layers": 12,
|
| 136 |
+
"num_return_sequences": 1,
|
| 137 |
+
"output_attentions": false,
|
| 138 |
+
"output_hidden_states": false,
|
| 139 |
+
"output_scores": false,
|
| 140 |
+
"pad_token_id": null,
|
| 141 |
+
"patch_size": 16,
|
| 142 |
+
"prefix": null,
|
| 143 |
+
"problem_type": null,
|
| 144 |
+
"pruned_heads": {},
|
| 145 |
+
"qkv_bias": false,
|
| 146 |
+
"remove_invalid_values": false,
|
| 147 |
+
"repetition_penalty": 1.0,
|
| 148 |
+
"return_dict": true,
|
| 149 |
+
"return_dict_in_generate": false,
|
| 150 |
+
"sep_token_id": null,
|
| 151 |
+
"suppress_tokens": null,
|
| 152 |
+
"task_specific_params": null,
|
| 153 |
+
"temperature": 1.0,
|
| 154 |
+
"tf_legacy_loss": false,
|
| 155 |
+
"tie_encoder_decoder": false,
|
| 156 |
+
"tie_word_embeddings": true,
|
| 157 |
+
"tokenizer_class": null,
|
| 158 |
+
"top_k": 50,
|
| 159 |
+
"top_p": 1.0,
|
| 160 |
+
"torch_dtype": null,
|
| 161 |
+
"torchscript": false,
|
| 162 |
+
"typical_p": 1.0,
|
| 163 |
+
"use_bfloat16": false
|
| 164 |
+
},
|
| 165 |
+
"eos_token_id": 2,
|
| 166 |
+
"is_encoder_decoder": true,
|
| 167 |
+
"length_penalty": 2.0,
|
| 168 |
+
"max_length": 64,
|
| 169 |
+
"model_type": "vision-encoder-decoder",
|
| 170 |
+
"no_repeat_ngram_size": 3,
|
| 171 |
+
"num_beams": 4,
|
| 172 |
+
"pad_token_id": 1,
|
| 173 |
+
"processor_class": "TrOCRProcessor",
|
| 174 |
+
"tie_word_embeddings": false,
|
| 175 |
+
"torch_dtype": "float32",
|
| 176 |
+
"transformers_version": "4.38.2",
|
| 177 |
+
"vocab_size": 50265
|
| 178 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"length_penalty": 2.0,
|
| 7 |
+
"max_length": 64,
|
| 8 |
+
"no_repeat_ngram_size": 3,
|
| 9 |
+
"num_beams": 4,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.38.2",
|
| 12 |
+
"use_cache": false
|
| 13 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 384,
|
| 20 |
+
"width": 384
|
| 21 |
+
}
|
| 22 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4e1790a384fa14b1e49faa88f9183cfcd3fa74835aff7b5414ba65f749ac175
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:481f461d53ee38dc27cc9006d53d051bc7d98728f6b39cfd8d7d040a4199fcae
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,581 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 35.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 875,
|
| 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": 6.834115982055664,
|
| 14 |
+
"learning_rate": 4.874285714285715e-05,
|
| 15 |
+
"loss": 3.1488,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_cer": 0.37662337662337664,
|
| 21 |
+
"eval_loss": 0.7865502834320068,
|
| 22 |
+
"eval_runtime": 8.8116,
|
| 23 |
+
"eval_samples_per_second": 2.27,
|
| 24 |
+
"eval_steps_per_second": 0.34,
|
| 25 |
+
"step": 25
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"grad_norm": 5.263820171356201,
|
| 30 |
+
"learning_rate": 4.7314285714285714e-05,
|
| 31 |
+
"loss": 0.8093,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"eval_cer": 0.4675324675324675,
|
| 37 |
+
"eval_loss": 0.8947464823722839,
|
| 38 |
+
"eval_runtime": 2.7975,
|
| 39 |
+
"eval_samples_per_second": 7.149,
|
| 40 |
+
"eval_steps_per_second": 1.072,
|
| 41 |
+
"step": 50
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"grad_norm": 5.471518039703369,
|
| 46 |
+
"learning_rate": 4.588571428571429e-05,
|
| 47 |
+
"loss": 0.724,
|
| 48 |
+
"step": 75
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"eval_cer": 0.4155844155844156,
|
| 53 |
+
"eval_loss": 0.8557536005973816,
|
| 54 |
+
"eval_runtime": 2.7561,
|
| 55 |
+
"eval_samples_per_second": 7.257,
|
| 56 |
+
"eval_steps_per_second": 1.088,
|
| 57 |
+
"step": 75
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 6.84343957901001,
|
| 62 |
+
"learning_rate": 4.445714285714286e-05,
|
| 63 |
+
"loss": 0.7405,
|
| 64 |
+
"step": 100
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_cer": 0.33766233766233766,
|
| 69 |
+
"eval_loss": 0.8277163505554199,
|
| 70 |
+
"eval_runtime": 2.8432,
|
| 71 |
+
"eval_samples_per_second": 7.034,
|
| 72 |
+
"eval_steps_per_second": 1.055,
|
| 73 |
+
"step": 100
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"grad_norm": 4.826574802398682,
|
| 78 |
+
"learning_rate": 4.302857142857143e-05,
|
| 79 |
+
"loss": 0.7361,
|
| 80 |
+
"step": 125
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.0,
|
| 84 |
+
"eval_cer": 0.36363636363636365,
|
| 85 |
+
"eval_loss": 0.8111216425895691,
|
| 86 |
+
"eval_runtime": 2.8495,
|
| 87 |
+
"eval_samples_per_second": 7.019,
|
| 88 |
+
"eval_steps_per_second": 1.053,
|
| 89 |
+
"step": 125
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"grad_norm": 4.757009983062744,
|
| 94 |
+
"learning_rate": 4.16e-05,
|
| 95 |
+
"loss": 0.7061,
|
| 96 |
+
"step": 150
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 6.0,
|
| 100 |
+
"eval_cer": 0.37662337662337664,
|
| 101 |
+
"eval_loss": 0.7869578003883362,
|
| 102 |
+
"eval_runtime": 2.8778,
|
| 103 |
+
"eval_samples_per_second": 6.95,
|
| 104 |
+
"eval_steps_per_second": 1.042,
|
| 105 |
+
"step": 150
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"grad_norm": 4.829946517944336,
|
| 110 |
+
"learning_rate": 4.017142857142857e-05,
|
| 111 |
+
"loss": 0.7159,
|
| 112 |
+
"step": 175
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 7.0,
|
| 116 |
+
"eval_cer": 0.45454545454545453,
|
| 117 |
+
"eval_loss": 0.8124594688415527,
|
| 118 |
+
"eval_runtime": 2.9455,
|
| 119 |
+
"eval_samples_per_second": 6.79,
|
| 120 |
+
"eval_steps_per_second": 1.019,
|
| 121 |
+
"step": 175
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 3.9984848499298096,
|
| 126 |
+
"learning_rate": 3.874285714285715e-05,
|
| 127 |
+
"loss": 0.707,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_cer": 0.37662337662337664,
|
| 133 |
+
"eval_loss": 0.8081725239753723,
|
| 134 |
+
"eval_runtime": 2.8597,
|
| 135 |
+
"eval_samples_per_second": 6.994,
|
| 136 |
+
"eval_steps_per_second": 1.049,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"grad_norm": 3.7987842559814453,
|
| 142 |
+
"learning_rate": 3.7314285714285715e-05,
|
| 143 |
+
"loss": 0.7073,
|
| 144 |
+
"step": 225
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.0,
|
| 148 |
+
"eval_cer": 0.45454545454545453,
|
| 149 |
+
"eval_loss": 0.8078359961509705,
|
| 150 |
+
"eval_runtime": 2.8752,
|
| 151 |
+
"eval_samples_per_second": 6.956,
|
| 152 |
+
"eval_steps_per_second": 1.043,
|
| 153 |
+
"step": 225
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 10.0,
|
| 157 |
+
"grad_norm": 4.148154258728027,
|
| 158 |
+
"learning_rate": 3.588571428571429e-05,
|
| 159 |
+
"loss": 0.6956,
|
| 160 |
+
"step": 250
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 10.0,
|
| 164 |
+
"eval_cer": 0.37662337662337664,
|
| 165 |
+
"eval_loss": 0.8172693848609924,
|
| 166 |
+
"eval_runtime": 2.8623,
|
| 167 |
+
"eval_samples_per_second": 6.987,
|
| 168 |
+
"eval_steps_per_second": 1.048,
|
| 169 |
+
"step": 250
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 11.0,
|
| 173 |
+
"grad_norm": 3.5226118564605713,
|
| 174 |
+
"learning_rate": 3.445714285714286e-05,
|
| 175 |
+
"loss": 0.7018,
|
| 176 |
+
"step": 275
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 11.0,
|
| 180 |
+
"eval_cer": 0.37662337662337664,
|
| 181 |
+
"eval_loss": 0.8221365809440613,
|
| 182 |
+
"eval_runtime": 2.8711,
|
| 183 |
+
"eval_samples_per_second": 6.966,
|
| 184 |
+
"eval_steps_per_second": 1.045,
|
| 185 |
+
"step": 275
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 3.736212730407715,
|
| 190 |
+
"learning_rate": 3.302857142857143e-05,
|
| 191 |
+
"loss": 0.6901,
|
| 192 |
+
"step": 300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_cer": 0.45454545454545453,
|
| 197 |
+
"eval_loss": 0.8348967432975769,
|
| 198 |
+
"eval_runtime": 2.9677,
|
| 199 |
+
"eval_samples_per_second": 6.739,
|
| 200 |
+
"eval_steps_per_second": 1.011,
|
| 201 |
+
"step": 300
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 13.0,
|
| 205 |
+
"grad_norm": 3.468735456466675,
|
| 206 |
+
"learning_rate": 3.16e-05,
|
| 207 |
+
"loss": 0.6862,
|
| 208 |
+
"step": 325
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 13.0,
|
| 212 |
+
"eval_cer": 0.36363636363636365,
|
| 213 |
+
"eval_loss": 0.7819182276725769,
|
| 214 |
+
"eval_runtime": 2.8624,
|
| 215 |
+
"eval_samples_per_second": 6.987,
|
| 216 |
+
"eval_steps_per_second": 1.048,
|
| 217 |
+
"step": 325
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 14.0,
|
| 221 |
+
"grad_norm": 4.044895648956299,
|
| 222 |
+
"learning_rate": 3.0171428571428572e-05,
|
| 223 |
+
"loss": 0.7024,
|
| 224 |
+
"step": 350
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.0,
|
| 228 |
+
"eval_cer": 0.36363636363636365,
|
| 229 |
+
"eval_loss": 0.7919606566429138,
|
| 230 |
+
"eval_runtime": 2.907,
|
| 231 |
+
"eval_samples_per_second": 6.88,
|
| 232 |
+
"eval_steps_per_second": 1.032,
|
| 233 |
+
"step": 350
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.0,
|
| 237 |
+
"grad_norm": 3.781301259994507,
|
| 238 |
+
"learning_rate": 2.8742857142857143e-05,
|
| 239 |
+
"loss": 0.6858,
|
| 240 |
+
"step": 375
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.0,
|
| 244 |
+
"eval_cer": 0.36363636363636365,
|
| 245 |
+
"eval_loss": 0.812893807888031,
|
| 246 |
+
"eval_runtime": 2.8775,
|
| 247 |
+
"eval_samples_per_second": 6.95,
|
| 248 |
+
"eval_steps_per_second": 1.043,
|
| 249 |
+
"step": 375
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 3.412642240524292,
|
| 254 |
+
"learning_rate": 2.7314285714285716e-05,
|
| 255 |
+
"loss": 0.6782,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_cer": 0.4025974025974026,
|
| 261 |
+
"eval_loss": 0.8198434710502625,
|
| 262 |
+
"eval_runtime": 2.89,
|
| 263 |
+
"eval_samples_per_second": 6.92,
|
| 264 |
+
"eval_steps_per_second": 1.038,
|
| 265 |
+
"step": 400
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 17.0,
|
| 269 |
+
"grad_norm": 2.96364426612854,
|
| 270 |
+
"learning_rate": 2.5885714285714286e-05,
|
| 271 |
+
"loss": 0.675,
|
| 272 |
+
"step": 425
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 17.0,
|
| 276 |
+
"eval_cer": 0.44155844155844154,
|
| 277 |
+
"eval_loss": 0.8002452850341797,
|
| 278 |
+
"eval_runtime": 2.9999,
|
| 279 |
+
"eval_samples_per_second": 6.667,
|
| 280 |
+
"eval_steps_per_second": 1.0,
|
| 281 |
+
"step": 425
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 18.0,
|
| 285 |
+
"grad_norm": 3.237696647644043,
|
| 286 |
+
"learning_rate": 2.445714285714286e-05,
|
| 287 |
+
"loss": 0.6749,
|
| 288 |
+
"step": 450
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 18.0,
|
| 292 |
+
"eval_cer": 0.36363636363636365,
|
| 293 |
+
"eval_loss": 0.7925994396209717,
|
| 294 |
+
"eval_runtime": 2.9067,
|
| 295 |
+
"eval_samples_per_second": 6.881,
|
| 296 |
+
"eval_steps_per_second": 1.032,
|
| 297 |
+
"step": 450
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 19.0,
|
| 301 |
+
"grad_norm": 3.5985169410705566,
|
| 302 |
+
"learning_rate": 2.302857142857143e-05,
|
| 303 |
+
"loss": 0.6723,
|
| 304 |
+
"step": 475
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 19.0,
|
| 308 |
+
"eval_cer": 0.4025974025974026,
|
| 309 |
+
"eval_loss": 0.8083222508430481,
|
| 310 |
+
"eval_runtime": 2.8862,
|
| 311 |
+
"eval_samples_per_second": 6.93,
|
| 312 |
+
"eval_steps_per_second": 1.039,
|
| 313 |
+
"step": 475
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 3.7446236610412598,
|
| 318 |
+
"learning_rate": 2.16e-05,
|
| 319 |
+
"loss": 0.6736,
|
| 320 |
+
"step": 500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_cer": 0.42857142857142855,
|
| 325 |
+
"eval_loss": 0.832375705242157,
|
| 326 |
+
"eval_runtime": 2.9714,
|
| 327 |
+
"eval_samples_per_second": 6.731,
|
| 328 |
+
"eval_steps_per_second": 1.01,
|
| 329 |
+
"step": 500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.0,
|
| 333 |
+
"grad_norm": 3.926819324493408,
|
| 334 |
+
"learning_rate": 2.0171428571428573e-05,
|
| 335 |
+
"loss": 0.6646,
|
| 336 |
+
"step": 525
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.0,
|
| 340 |
+
"eval_cer": 0.45454545454545453,
|
| 341 |
+
"eval_loss": 0.8589693903923035,
|
| 342 |
+
"eval_runtime": 2.9745,
|
| 343 |
+
"eval_samples_per_second": 6.724,
|
| 344 |
+
"eval_steps_per_second": 1.009,
|
| 345 |
+
"step": 525
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.0,
|
| 349 |
+
"grad_norm": 3.9272310733795166,
|
| 350 |
+
"learning_rate": 1.8742857142857143e-05,
|
| 351 |
+
"loss": 0.6586,
|
| 352 |
+
"step": 550
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 22.0,
|
| 356 |
+
"eval_cer": 0.36363636363636365,
|
| 357 |
+
"eval_loss": 0.803028404712677,
|
| 358 |
+
"eval_runtime": 2.9162,
|
| 359 |
+
"eval_samples_per_second": 6.858,
|
| 360 |
+
"eval_steps_per_second": 1.029,
|
| 361 |
+
"step": 550
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 23.0,
|
| 365 |
+
"grad_norm": 3.406860113143921,
|
| 366 |
+
"learning_rate": 1.7314285714285717e-05,
|
| 367 |
+
"loss": 0.6665,
|
| 368 |
+
"step": 575
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 23.0,
|
| 372 |
+
"eval_cer": 0.4675324675324675,
|
| 373 |
+
"eval_loss": 0.8490405082702637,
|
| 374 |
+
"eval_runtime": 2.8831,
|
| 375 |
+
"eval_samples_per_second": 6.937,
|
| 376 |
+
"eval_steps_per_second": 1.041,
|
| 377 |
+
"step": 575
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 3.1726770401000977,
|
| 382 |
+
"learning_rate": 1.5885714285714287e-05,
|
| 383 |
+
"loss": 0.655,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_cer": 0.4675324675324675,
|
| 389 |
+
"eval_loss": 0.8455414772033691,
|
| 390 |
+
"eval_runtime": 2.9414,
|
| 391 |
+
"eval_samples_per_second": 6.799,
|
| 392 |
+
"eval_steps_per_second": 1.02,
|
| 393 |
+
"step": 600
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 25.0,
|
| 397 |
+
"grad_norm": 3.506122589111328,
|
| 398 |
+
"learning_rate": 1.4457142857142857e-05,
|
| 399 |
+
"loss": 0.6641,
|
| 400 |
+
"step": 625
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 25.0,
|
| 404 |
+
"eval_cer": 0.42857142857142855,
|
| 405 |
+
"eval_loss": 0.8133634924888611,
|
| 406 |
+
"eval_runtime": 3.0374,
|
| 407 |
+
"eval_samples_per_second": 6.584,
|
| 408 |
+
"eval_steps_per_second": 0.988,
|
| 409 |
+
"step": 625
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 26.0,
|
| 413 |
+
"grad_norm": 4.370969772338867,
|
| 414 |
+
"learning_rate": 1.3028571428571429e-05,
|
| 415 |
+
"loss": 0.6597,
|
| 416 |
+
"step": 650
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 26.0,
|
| 420 |
+
"eval_cer": 0.4675324675324675,
|
| 421 |
+
"eval_loss": 0.8322270512580872,
|
| 422 |
+
"eval_runtime": 2.8213,
|
| 423 |
+
"eval_samples_per_second": 7.089,
|
| 424 |
+
"eval_steps_per_second": 1.063,
|
| 425 |
+
"step": 650
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 27.0,
|
| 429 |
+
"grad_norm": 3.851907253265381,
|
| 430 |
+
"learning_rate": 1.16e-05,
|
| 431 |
+
"loss": 0.6611,
|
| 432 |
+
"step": 675
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.0,
|
| 436 |
+
"eval_cer": 0.45454545454545453,
|
| 437 |
+
"eval_loss": 0.8302900195121765,
|
| 438 |
+
"eval_runtime": 2.8659,
|
| 439 |
+
"eval_samples_per_second": 6.979,
|
| 440 |
+
"eval_steps_per_second": 1.047,
|
| 441 |
+
"step": 675
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"grad_norm": 2.989511013031006,
|
| 446 |
+
"learning_rate": 1.0171428571428573e-05,
|
| 447 |
+
"loss": 0.6617,
|
| 448 |
+
"step": 700
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 28.0,
|
| 452 |
+
"eval_cer": 0.45454545454545453,
|
| 453 |
+
"eval_loss": 0.8202476501464844,
|
| 454 |
+
"eval_runtime": 2.9062,
|
| 455 |
+
"eval_samples_per_second": 6.882,
|
| 456 |
+
"eval_steps_per_second": 1.032,
|
| 457 |
+
"step": 700
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 29.0,
|
| 461 |
+
"grad_norm": 3.977627992630005,
|
| 462 |
+
"learning_rate": 8.742857142857143e-06,
|
| 463 |
+
"loss": 0.6531,
|
| 464 |
+
"step": 725
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 29.0,
|
| 468 |
+
"eval_cer": 0.45454545454545453,
|
| 469 |
+
"eval_loss": 0.8274255394935608,
|
| 470 |
+
"eval_runtime": 2.8627,
|
| 471 |
+
"eval_samples_per_second": 6.987,
|
| 472 |
+
"eval_steps_per_second": 1.048,
|
| 473 |
+
"step": 725
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 30.0,
|
| 477 |
+
"grad_norm": 3.391641855239868,
|
| 478 |
+
"learning_rate": 7.314285714285715e-06,
|
| 479 |
+
"loss": 0.6491,
|
| 480 |
+
"step": 750
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.0,
|
| 484 |
+
"eval_cer": 0.42857142857142855,
|
| 485 |
+
"eval_loss": 0.8236328959465027,
|
| 486 |
+
"eval_runtime": 2.8678,
|
| 487 |
+
"eval_samples_per_second": 6.974,
|
| 488 |
+
"eval_steps_per_second": 1.046,
|
| 489 |
+
"step": 750
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 31.0,
|
| 493 |
+
"grad_norm": 3.7944726943969727,
|
| 494 |
+
"learning_rate": 5.885714285714286e-06,
|
| 495 |
+
"loss": 0.6471,
|
| 496 |
+
"step": 775
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 31.0,
|
| 500 |
+
"eval_cer": 0.4155844155844156,
|
| 501 |
+
"eval_loss": 0.8234376907348633,
|
| 502 |
+
"eval_runtime": 2.9962,
|
| 503 |
+
"eval_samples_per_second": 6.675,
|
| 504 |
+
"eval_steps_per_second": 1.001,
|
| 505 |
+
"step": 775
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"grad_norm": 4.810728073120117,
|
| 510 |
+
"learning_rate": 4.457142857142858e-06,
|
| 511 |
+
"loss": 0.6446,
|
| 512 |
+
"step": 800
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 32.0,
|
| 516 |
+
"eval_cer": 0.42857142857142855,
|
| 517 |
+
"eval_loss": 0.83391934633255,
|
| 518 |
+
"eval_runtime": 2.8811,
|
| 519 |
+
"eval_samples_per_second": 6.942,
|
| 520 |
+
"eval_steps_per_second": 1.041,
|
| 521 |
+
"step": 800
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 33.0,
|
| 525 |
+
"grad_norm": 3.008390188217163,
|
| 526 |
+
"learning_rate": 3.028571428571429e-06,
|
| 527 |
+
"loss": 0.6439,
|
| 528 |
+
"step": 825
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 33.0,
|
| 532 |
+
"eval_cer": 0.4025974025974026,
|
| 533 |
+
"eval_loss": 0.8176433444023132,
|
| 534 |
+
"eval_runtime": 2.8474,
|
| 535 |
+
"eval_samples_per_second": 7.024,
|
| 536 |
+
"eval_steps_per_second": 1.054,
|
| 537 |
+
"step": 825
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 34.0,
|
| 541 |
+
"grad_norm": 4.271653652191162,
|
| 542 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 543 |
+
"loss": 0.6453,
|
| 544 |
+
"step": 850
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 34.0,
|
| 548 |
+
"eval_cer": 0.4025974025974026,
|
| 549 |
+
"eval_loss": 0.8156414031982422,
|
| 550 |
+
"eval_runtime": 2.8852,
|
| 551 |
+
"eval_samples_per_second": 6.932,
|
| 552 |
+
"eval_steps_per_second": 1.04,
|
| 553 |
+
"step": 850
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 35.0,
|
| 557 |
+
"grad_norm": 4.2927045822143555,
|
| 558 |
+
"learning_rate": 1.7142857142857143e-07,
|
| 559 |
+
"loss": 0.6404,
|
| 560 |
+
"step": 875
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 35.0,
|
| 564 |
+
"eval_cer": 0.35064935064935066,
|
| 565 |
+
"eval_loss": 0.8252768516540527,
|
| 566 |
+
"eval_runtime": 2.8782,
|
| 567 |
+
"eval_samples_per_second": 6.949,
|
| 568 |
+
"eval_steps_per_second": 1.042,
|
| 569 |
+
"step": 875
|
| 570 |
+
}
|
| 571 |
+
],
|
| 572 |
+
"logging_steps": 500,
|
| 573 |
+
"max_steps": 875,
|
| 574 |
+
"num_input_tokens_seen": 0,
|
| 575 |
+
"num_train_epochs": 35,
|
| 576 |
+
"save_steps": 500,
|
| 577 |
+
"total_flos": 5.237996336971776e+18,
|
| 578 |
+
"train_batch_size": 8,
|
| 579 |
+
"trial_name": null,
|
| 580 |
+
"trial_params": null
|
| 581 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67893d287f7b09d71e4feeb0b5000c89db68ddcc92bb963e8763ff2ef6fa84b5
|
| 3 |
+
size 5048
|