model
Browse files- config.json +37 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- trainer_state.json +274 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"id2label": {
|
| 11 |
+
"0": "cardiovascular",
|
| 12 |
+
"1": "hepatorenal",
|
| 13 |
+
"2": "neurological",
|
| 14 |
+
"3": "oncological"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"cardiovascular": 0,
|
| 20 |
+
"hepatorenal": 1,
|
| 21 |
+
"neurological": 2,
|
| 22 |
+
"oncological": 3
|
| 23 |
+
},
|
| 24 |
+
"layer_norm_eps": 1e-12,
|
| 25 |
+
"max_position_embeddings": 512,
|
| 26 |
+
"model_type": "bert",
|
| 27 |
+
"num_attention_heads": 12,
|
| 28 |
+
"num_hidden_layers": 12,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"position_embedding_type": "absolute",
|
| 31 |
+
"problem_type": "multi_label_classification",
|
| 32 |
+
"torch_dtype": "float32",
|
| 33 |
+
"transformers_version": "4.52.4",
|
| 34 |
+
"type_vocab_size": 2,
|
| 35 |
+
"use_cache": true,
|
| 36 |
+
"vocab_size": 31090
|
| 37 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e26268954bd9906cd275473e97d8c93e2c09d0825bb294150faea71fd9c26ac
|
| 3 |
+
size 439709696
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"101": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"102": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"103": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"104": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,274 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1092,
|
| 3 |
+
"best_metric": 0.9566188454020108,
|
| 4 |
+
"best_model_checkpoint": "/kaggle/working/hpo_scibert_uncased/trial_6/checkpoint-1092",
|
| 5 |
+
"epoch": 7.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1092,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.32051282051282054,
|
| 14 |
+
"grad_norm": 121545.390625,
|
| 15 |
+
"learning_rate": 4.628257110973899e-05,
|
| 16 |
+
"loss": 0.6602,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.6410256410256411,
|
| 21 |
+
"grad_norm": 165566.34375,
|
| 22 |
+
"learning_rate": 4.406384766248401e-05,
|
| 23 |
+
"loss": 0.2825,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.9615384615384616,
|
| 28 |
+
"grad_norm": 183398.71875,
|
| 29 |
+
"learning_rate": 4.184512421522902e-05,
|
| 30 |
+
"loss": 0.2105,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0,
|
| 35 |
+
"eval_f1_macro": 0.9405513571748276,
|
| 36 |
+
"eval_f1_micro": 0.9374337221633087,
|
| 37 |
+
"eval_f1_weighted": 0.9373994117527249,
|
| 38 |
+
"eval_loss": 0.19546222686767578,
|
| 39 |
+
"eval_roc_auc_macro": 0.9773648844952434,
|
| 40 |
+
"eval_runtime": 6.6101,
|
| 41 |
+
"eval_samples_per_second": 54.008,
|
| 42 |
+
"eval_steps_per_second": 0.908,
|
| 43 |
+
"step": 156
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 1.282051282051282,
|
| 47 |
+
"grad_norm": 119565.84375,
|
| 48 |
+
"learning_rate": 3.962640076797404e-05,
|
| 49 |
+
"loss": 0.1367,
|
| 50 |
+
"step": 200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.6025641025641026,
|
| 54 |
+
"grad_norm": 162723.625,
|
| 55 |
+
"learning_rate": 3.740767732071905e-05,
|
| 56 |
+
"loss": 0.1247,
|
| 57 |
+
"step": 250
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.9230769230769231,
|
| 61 |
+
"grad_norm": 39087.80859375,
|
| 62 |
+
"learning_rate": 3.5188953873464064e-05,
|
| 63 |
+
"loss": 0.1182,
|
| 64 |
+
"step": 300
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.0,
|
| 68 |
+
"eval_f1_macro": 0.9436427931387608,
|
| 69 |
+
"eval_f1_micro": 0.94375,
|
| 70 |
+
"eval_f1_weighted": 0.9438667437399432,
|
| 71 |
+
"eval_loss": 0.16677653789520264,
|
| 72 |
+
"eval_roc_auc_macro": 0.9820117282613596,
|
| 73 |
+
"eval_runtime": 6.6121,
|
| 74 |
+
"eval_samples_per_second": 53.992,
|
| 75 |
+
"eval_steps_per_second": 0.907,
|
| 76 |
+
"step": 312
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.2435897435897436,
|
| 80 |
+
"grad_norm": 125978.8359375,
|
| 81 |
+
"learning_rate": 3.297023042620908e-05,
|
| 82 |
+
"loss": 0.0894,
|
| 83 |
+
"step": 350
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.564102564102564,
|
| 87 |
+
"grad_norm": 264165.40625,
|
| 88 |
+
"learning_rate": 3.0751506978954095e-05,
|
| 89 |
+
"loss": 0.0687,
|
| 90 |
+
"step": 400
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.8846153846153846,
|
| 94 |
+
"grad_norm": 68308.296875,
|
| 95 |
+
"learning_rate": 2.853278353169911e-05,
|
| 96 |
+
"loss": 0.0651,
|
| 97 |
+
"step": 450
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 3.0,
|
| 101 |
+
"eval_f1_macro": 0.9445835024754138,
|
| 102 |
+
"eval_f1_micro": 0.948024948024948,
|
| 103 |
+
"eval_f1_weighted": 0.9482797012589066,
|
| 104 |
+
"eval_loss": 0.15806053578853607,
|
| 105 |
+
"eval_roc_auc_macro": 0.9879374351632613,
|
| 106 |
+
"eval_runtime": 6.6786,
|
| 107 |
+
"eval_samples_per_second": 53.454,
|
| 108 |
+
"eval_steps_per_second": 0.898,
|
| 109 |
+
"step": 468
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 3.2051282051282053,
|
| 113 |
+
"grad_norm": 15062.0087890625,
|
| 114 |
+
"learning_rate": 2.6314060084444123e-05,
|
| 115 |
+
"loss": 0.0452,
|
| 116 |
+
"step": 500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.5256410256410255,
|
| 120 |
+
"grad_norm": 10435.2138671875,
|
| 121 |
+
"learning_rate": 2.409533663718914e-05,
|
| 122 |
+
"loss": 0.0429,
|
| 123 |
+
"step": 550
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 3.8461538461538463,
|
| 127 |
+
"grad_norm": 240173.984375,
|
| 128 |
+
"learning_rate": 2.1876613189934154e-05,
|
| 129 |
+
"loss": 0.0495,
|
| 130 |
+
"step": 600
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 4.0,
|
| 134 |
+
"eval_f1_macro": 0.9468900491419501,
|
| 135 |
+
"eval_f1_micro": 0.9495798319327731,
|
| 136 |
+
"eval_f1_weighted": 0.9496287427368676,
|
| 137 |
+
"eval_loss": 0.19388458132743835,
|
| 138 |
+
"eval_roc_auc_macro": 0.9843316823250352,
|
| 139 |
+
"eval_runtime": 6.5837,
|
| 140 |
+
"eval_samples_per_second": 54.225,
|
| 141 |
+
"eval_steps_per_second": 0.911,
|
| 142 |
+
"step": 624
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 4.166666666666667,
|
| 146 |
+
"grad_norm": 3220.84521484375,
|
| 147 |
+
"learning_rate": 1.965788974267917e-05,
|
| 148 |
+
"loss": 0.0278,
|
| 149 |
+
"step": 650
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 4.487179487179487,
|
| 153 |
+
"grad_norm": 77989.6796875,
|
| 154 |
+
"learning_rate": 1.7439166295424185e-05,
|
| 155 |
+
"loss": 0.0328,
|
| 156 |
+
"step": 700
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 4.8076923076923075,
|
| 160 |
+
"grad_norm": 17635.943359375,
|
| 161 |
+
"learning_rate": 1.5220442848169199e-05,
|
| 162 |
+
"loss": 0.0194,
|
| 163 |
+
"step": 750
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 5.0,
|
| 167 |
+
"eval_f1_macro": 0.9542678396871944,
|
| 168 |
+
"eval_f1_micro": 0.9544513457556936,
|
| 169 |
+
"eval_f1_weighted": 0.954747368183765,
|
| 170 |
+
"eval_loss": 0.17806702852249146,
|
| 171 |
+
"eval_roc_auc_macro": 0.9850715189437066,
|
| 172 |
+
"eval_runtime": 6.6153,
|
| 173 |
+
"eval_samples_per_second": 53.966,
|
| 174 |
+
"eval_steps_per_second": 0.907,
|
| 175 |
+
"step": 780
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 5.128205128205128,
|
| 179 |
+
"grad_norm": 7326.7744140625,
|
| 180 |
+
"learning_rate": 1.3001719400914214e-05,
|
| 181 |
+
"loss": 0.0209,
|
| 182 |
+
"step": 800
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 5.448717948717949,
|
| 186 |
+
"grad_norm": 42033.6640625,
|
| 187 |
+
"learning_rate": 1.0782995953659228e-05,
|
| 188 |
+
"loss": 0.0112,
|
| 189 |
+
"step": 850
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 5.769230769230769,
|
| 193 |
+
"grad_norm": 20092.216796875,
|
| 194 |
+
"learning_rate": 8.564272506404243e-06,
|
| 195 |
+
"loss": 0.0179,
|
| 196 |
+
"step": 900
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 6.0,
|
| 200 |
+
"eval_f1_macro": 0.9521823752294121,
|
| 201 |
+
"eval_f1_micro": 0.955067920585162,
|
| 202 |
+
"eval_f1_weighted": 0.9553084318062802,
|
| 203 |
+
"eval_loss": 0.22233474254608154,
|
| 204 |
+
"eval_roc_auc_macro": 0.9832222195348375,
|
| 205 |
+
"eval_runtime": 6.606,
|
| 206 |
+
"eval_samples_per_second": 54.042,
|
| 207 |
+
"eval_steps_per_second": 0.908,
|
| 208 |
+
"step": 936
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 6.089743589743589,
|
| 212 |
+
"grad_norm": 630.0228881835938,
|
| 213 |
+
"learning_rate": 6.345549059149257e-06,
|
| 214 |
+
"loss": 0.0115,
|
| 215 |
+
"step": 950
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 6.410256410256411,
|
| 219 |
+
"grad_norm": 469.14447021484375,
|
| 220 |
+
"learning_rate": 4.126825611894273e-06,
|
| 221 |
+
"loss": 0.005,
|
| 222 |
+
"step": 1000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 6.730769230769231,
|
| 226 |
+
"grad_norm": 564.0679931640625,
|
| 227 |
+
"learning_rate": 1.9081021646392873e-06,
|
| 228 |
+
"loss": 0.0125,
|
| 229 |
+
"step": 1050
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 7.0,
|
| 233 |
+
"eval_f1_macro": 0.9548856387728133,
|
| 234 |
+
"eval_f1_micro": 0.9563409563409563,
|
| 235 |
+
"eval_f1_weighted": 0.9566188454020108,
|
| 236 |
+
"eval_loss": 0.2219453752040863,
|
| 237 |
+
"eval_roc_auc_macro": 0.9854251710108592,
|
| 238 |
+
"eval_runtime": 6.6485,
|
| 239 |
+
"eval_samples_per_second": 53.697,
|
| 240 |
+
"eval_steps_per_second": 0.902,
|
| 241 |
+
"step": 1092
|
| 242 |
+
}
|
| 243 |
+
],
|
| 244 |
+
"logging_steps": 50,
|
| 245 |
+
"max_steps": 1092,
|
| 246 |
+
"num_input_tokens_seen": 0,
|
| 247 |
+
"num_train_epochs": 7,
|
| 248 |
+
"save_steps": 500,
|
| 249 |
+
"stateful_callbacks": {
|
| 250 |
+
"EarlyStoppingCallback": {
|
| 251 |
+
"args": {
|
| 252 |
+
"early_stopping_patience": 1,
|
| 253 |
+
"early_stopping_threshold": 0.0
|
| 254 |
+
},
|
| 255 |
+
"attributes": {
|
| 256 |
+
"early_stopping_patience_counter": 0
|
| 257 |
+
}
|
| 258 |
+
},
|
| 259 |
+
"TrainerControl": {
|
| 260 |
+
"args": {
|
| 261 |
+
"should_epoch_stop": false,
|
| 262 |
+
"should_evaluate": false,
|
| 263 |
+
"should_log": false,
|
| 264 |
+
"should_save": true,
|
| 265 |
+
"should_training_stop": true
|
| 266 |
+
},
|
| 267 |
+
"attributes": {}
|
| 268 |
+
}
|
| 269 |
+
},
|
| 270 |
+
"total_flos": 9155639800000512.0,
|
| 271 |
+
"train_batch_size": 32,
|
| 272 |
+
"trial_name": null,
|
| 273 |
+
"trial_params": null
|
| 274 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|