Training in progress, epoch 5, checkpoint
Browse files
checkpoint-500/config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"layer_norm_eps": 1e-07,
|
| 13 |
+
"legacy": true,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"max_relative_positions": -1,
|
| 16 |
+
"model_type": "deberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"pooler_dropout": 0,
|
| 21 |
+
"pooler_hidden_act": "gelu",
|
| 22 |
+
"pooler_hidden_size": 768,
|
| 23 |
+
"pos_att_type": [
|
| 24 |
+
"c2p",
|
| 25 |
+
"p2c"
|
| 26 |
+
],
|
| 27 |
+
"position_biased_input": false,
|
| 28 |
+
"relative_attention": true,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.48.3",
|
| 31 |
+
"type_vocab_size": 0,
|
| 32 |
+
"vocab_size": 50265
|
| 33 |
+
}
|
checkpoint-500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f525bd75537a953f71238cf52c333865be51b56bb3b4830c43a5d030f74a1b4
|
| 3 |
+
size 556799560
|
checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,263 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5,
|
| 13 |
+
"grad_norm": 4.128329277038574,
|
| 14 |
+
"learning_rate": 9.166666666666666e-06,
|
| 15 |
+
"loss": 0.5694,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5,
|
| 20 |
+
"eval_pet_AUC": 0.877852385631619,
|
| 21 |
+
"eval_pet_accuracy": 0.7775,
|
| 22 |
+
"eval_pet_f1": 0.5898617511520737,
|
| 23 |
+
"eval_pet_loss": 0.4617109000682831,
|
| 24 |
+
"eval_pet_macro_f1": 0.7186015445297247,
|
| 25 |
+
"eval_pet_micro_f1": 0.7775,
|
| 26 |
+
"eval_pet_precision": 0.8533333333333334,
|
| 27 |
+
"eval_pet_recall": 0.4507042253521127,
|
| 28 |
+
"eval_pet_runtime": 0.434,
|
| 29 |
+
"eval_pet_samples_per_second": 921.733,
|
| 30 |
+
"eval_pet_specificity": 0.9573643410852714,
|
| 31 |
+
"eval_pet_steps_per_second": 2.304,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"grad_norm": 6.654369354248047,
|
| 37 |
+
"learning_rate": 8.333333333333334e-06,
|
| 38 |
+
"loss": 0.4313,
|
| 39 |
+
"step": 100
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0,
|
| 43 |
+
"eval_pet_AUC": 0.9084779997816355,
|
| 44 |
+
"eval_pet_accuracy": 0.815,
|
| 45 |
+
"eval_pet_f1": 0.7086614173228346,
|
| 46 |
+
"eval_pet_loss": 0.3827367424964905,
|
| 47 |
+
"eval_pet_macro_f1": 0.7865651408958496,
|
| 48 |
+
"eval_pet_micro_f1": 0.815,
|
| 49 |
+
"eval_pet_precision": 0.8035714285714286,
|
| 50 |
+
"eval_pet_recall": 0.6338028169014085,
|
| 51 |
+
"eval_pet_runtime": 0.4369,
|
| 52 |
+
"eval_pet_samples_per_second": 915.512,
|
| 53 |
+
"eval_pet_specificity": 0.9147286821705426,
|
| 54 |
+
"eval_pet_steps_per_second": 2.289,
|
| 55 |
+
"step": 100
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 3.3484041690826416,
|
| 60 |
+
"learning_rate": 7.500000000000001e-06,
|
| 61 |
+
"loss": 0.3259,
|
| 62 |
+
"step": 150
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.5,
|
| 66 |
+
"eval_pet_AUC": 0.9165028933289661,
|
| 67 |
+
"eval_pet_accuracy": 0.8325,
|
| 68 |
+
"eval_pet_f1": 0.7681660899653979,
|
| 69 |
+
"eval_pet_loss": 0.36348748207092285,
|
| 70 |
+
"eval_pet_macro_f1": 0.8185253150414074,
|
| 71 |
+
"eval_pet_micro_f1": 0.8325,
|
| 72 |
+
"eval_pet_precision": 0.7551020408163265,
|
| 73 |
+
"eval_pet_recall": 0.7816901408450704,
|
| 74 |
+
"eval_pet_runtime": 0.4775,
|
| 75 |
+
"eval_pet_samples_per_second": 837.78,
|
| 76 |
+
"eval_pet_specificity": 0.8604651162790697,
|
| 77 |
+
"eval_pet_steps_per_second": 2.094,
|
| 78 |
+
"step": 150
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.0,
|
| 82 |
+
"grad_norm": 5.001344203948975,
|
| 83 |
+
"learning_rate": 6.666666666666667e-06,
|
| 84 |
+
"loss": 0.2924,
|
| 85 |
+
"step": 200
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 2.0,
|
| 89 |
+
"eval_pet_AUC": 0.9257287913527678,
|
| 90 |
+
"eval_pet_accuracy": 0.85,
|
| 91 |
+
"eval_pet_f1": 0.8,
|
| 92 |
+
"eval_pet_loss": 0.346834272146225,
|
| 93 |
+
"eval_pet_macro_f1": 0.8400000000000001,
|
| 94 |
+
"eval_pet_micro_f1": 0.85,
|
| 95 |
+
"eval_pet_precision": 0.759493670886076,
|
| 96 |
+
"eval_pet_recall": 0.8450704225352113,
|
| 97 |
+
"eval_pet_runtime": 0.4621,
|
| 98 |
+
"eval_pet_samples_per_second": 865.539,
|
| 99 |
+
"eval_pet_specificity": 0.8527131782945736,
|
| 100 |
+
"eval_pet_steps_per_second": 2.164,
|
| 101 |
+
"step": 200
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.5,
|
| 105 |
+
"grad_norm": 9.390085220336914,
|
| 106 |
+
"learning_rate": 5.833333333333334e-06,
|
| 107 |
+
"loss": 0.2524,
|
| 108 |
+
"step": 250
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.5,
|
| 112 |
+
"eval_pet_AUC": 0.9255377224587837,
|
| 113 |
+
"eval_pet_accuracy": 0.8275,
|
| 114 |
+
"eval_pet_f1": 0.7206477732793523,
|
| 115 |
+
"eval_pet_loss": 0.40142711997032166,
|
| 116 |
+
"eval_pet_macro_f1": 0.797936906531177,
|
| 117 |
+
"eval_pet_micro_f1": 0.8275,
|
| 118 |
+
"eval_pet_precision": 0.8476190476190476,
|
| 119 |
+
"eval_pet_recall": 0.6267605633802817,
|
| 120 |
+
"eval_pet_runtime": 0.4565,
|
| 121 |
+
"eval_pet_samples_per_second": 876.237,
|
| 122 |
+
"eval_pet_specificity": 0.937984496124031,
|
| 123 |
+
"eval_pet_steps_per_second": 2.191,
|
| 124 |
+
"step": 250
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0,
|
| 128 |
+
"grad_norm": 5.205839157104492,
|
| 129 |
+
"learning_rate": 5e-06,
|
| 130 |
+
"loss": 0.2088,
|
| 131 |
+
"step": 300
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.0,
|
| 135 |
+
"eval_pet_AUC": 0.9287313025439459,
|
| 136 |
+
"eval_pet_accuracy": 0.85,
|
| 137 |
+
"eval_pet_f1": 0.7931034482758621,
|
| 138 |
+
"eval_pet_loss": 0.3718249797821045,
|
| 139 |
+
"eval_pet_macro_f1": 0.8377281947261663,
|
| 140 |
+
"eval_pet_micro_f1": 0.85,
|
| 141 |
+
"eval_pet_precision": 0.777027027027027,
|
| 142 |
+
"eval_pet_recall": 0.8098591549295775,
|
| 143 |
+
"eval_pet_runtime": 0.4379,
|
| 144 |
+
"eval_pet_samples_per_second": 913.543,
|
| 145 |
+
"eval_pet_specificity": 0.872093023255814,
|
| 146 |
+
"eval_pet_steps_per_second": 2.284,
|
| 147 |
+
"step": 300
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 3.5,
|
| 151 |
+
"grad_norm": 21.944002151489258,
|
| 152 |
+
"learning_rate": 4.166666666666667e-06,
|
| 153 |
+
"loss": 0.2056,
|
| 154 |
+
"step": 350
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 3.5,
|
| 158 |
+
"eval_pet_AUC": 0.9244459002074462,
|
| 159 |
+
"eval_pet_accuracy": 0.845,
|
| 160 |
+
"eval_pet_f1": 0.7615384615384615,
|
| 161 |
+
"eval_pet_loss": 0.45458969473838806,
|
| 162 |
+
"eval_pet_macro_f1": 0.8233618233618234,
|
| 163 |
+
"eval_pet_micro_f1": 0.845,
|
| 164 |
+
"eval_pet_precision": 0.8389830508474576,
|
| 165 |
+
"eval_pet_recall": 0.6971830985915493,
|
| 166 |
+
"eval_pet_runtime": 1.5537,
|
| 167 |
+
"eval_pet_samples_per_second": 257.455,
|
| 168 |
+
"eval_pet_specificity": 0.9263565891472868,
|
| 169 |
+
"eval_pet_steps_per_second": 0.644,
|
| 170 |
+
"step": 350
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.0,
|
| 174 |
+
"grad_norm": 8.805855751037598,
|
| 175 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 176 |
+
"loss": 0.1453,
|
| 177 |
+
"step": 400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 4.0,
|
| 181 |
+
"eval_pet_AUC": 0.9274484113986243,
|
| 182 |
+
"eval_pet_accuracy": 0.84,
|
| 183 |
+
"eval_pet_f1": 0.7557251908396947,
|
| 184 |
+
"eval_pet_loss": 0.4785776436328888,
|
| 185 |
+
"eval_pet_macro_f1": 0.8183830415165017,
|
| 186 |
+
"eval_pet_micro_f1": 0.84,
|
| 187 |
+
"eval_pet_precision": 0.825,
|
| 188 |
+
"eval_pet_recall": 0.6971830985915493,
|
| 189 |
+
"eval_pet_runtime": 0.4252,
|
| 190 |
+
"eval_pet_samples_per_second": 940.724,
|
| 191 |
+
"eval_pet_specificity": 0.9186046511627907,
|
| 192 |
+
"eval_pet_steps_per_second": 2.352,
|
| 193 |
+
"step": 400
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 4.5,
|
| 197 |
+
"grad_norm": 6.169196605682373,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 0.1278,
|
| 200 |
+
"step": 450
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.5,
|
| 204 |
+
"eval_pet_AUC": 0.9321705426356589,
|
| 205 |
+
"eval_pet_accuracy": 0.86,
|
| 206 |
+
"eval_pet_f1": 0.8193548387096774,
|
| 207 |
+
"eval_pet_loss": 0.472931444644928,
|
| 208 |
+
"eval_pet_macro_f1": 0.8525345622119815,
|
| 209 |
+
"eval_pet_micro_f1": 0.86,
|
| 210 |
+
"eval_pet_precision": 0.7559523809523809,
|
| 211 |
+
"eval_pet_recall": 0.8943661971830986,
|
| 212 |
+
"eval_pet_runtime": 0.4403,
|
| 213 |
+
"eval_pet_samples_per_second": 908.565,
|
| 214 |
+
"eval_pet_specificity": 0.8410852713178295,
|
| 215 |
+
"eval_pet_steps_per_second": 2.271,
|
| 216 |
+
"step": 450
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.0,
|
| 220 |
+
"grad_norm": 3.6303980350494385,
|
| 221 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 222 |
+
"loss": 0.1313,
|
| 223 |
+
"step": 500
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 5.0,
|
| 227 |
+
"eval_pet_AUC": 0.9292772136696146,
|
| 228 |
+
"eval_pet_accuracy": 0.8475,
|
| 229 |
+
"eval_pet_f1": 0.7715355805243446,
|
| 230 |
+
"eval_pet_loss": 0.5229131579399109,
|
| 231 |
+
"eval_pet_macro_f1": 0.8285445257218346,
|
| 232 |
+
"eval_pet_micro_f1": 0.8475,
|
| 233 |
+
"eval_pet_precision": 0.824,
|
| 234 |
+
"eval_pet_recall": 0.7253521126760564,
|
| 235 |
+
"eval_pet_runtime": 0.4211,
|
| 236 |
+
"eval_pet_samples_per_second": 949.835,
|
| 237 |
+
"eval_pet_specificity": 0.9147286821705426,
|
| 238 |
+
"eval_pet_steps_per_second": 2.375,
|
| 239 |
+
"step": 500
|
| 240 |
+
}
|
| 241 |
+
],
|
| 242 |
+
"logging_steps": 50,
|
| 243 |
+
"max_steps": 600,
|
| 244 |
+
"num_input_tokens_seen": 0,
|
| 245 |
+
"num_train_epochs": 6,
|
| 246 |
+
"save_steps": 500,
|
| 247 |
+
"stateful_callbacks": {
|
| 248 |
+
"TrainerControl": {
|
| 249 |
+
"args": {
|
| 250 |
+
"should_epoch_stop": false,
|
| 251 |
+
"should_evaluate": false,
|
| 252 |
+
"should_log": false,
|
| 253 |
+
"should_save": true,
|
| 254 |
+
"should_training_stop": false
|
| 255 |
+
},
|
| 256 |
+
"attributes": {}
|
| 257 |
+
}
|
| 258 |
+
},
|
| 259 |
+
"total_flos": 191623499520000.0,
|
| 260 |
+
"train_batch_size": 16,
|
| 261 |
+
"trial_name": null,
|
| 262 |
+
"trial_params": null
|
| 263 |
+
}
|
checkpoint-500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e32dc27c557e0f441bcd20b19146da9bd1fd47ccb03d8413c22d107b236c2a36
|
| 3 |
+
size 5432
|