Training in progress, epoch 5, checkpoint
Browse files
checkpoint-500/config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-large",
|
| 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": 1024,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 4096,
|
| 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": 16,
|
| 18 |
+
"num_hidden_layers": 24,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"pooler_dropout": 0,
|
| 21 |
+
"pooler_hidden_act": "gelu",
|
| 22 |
+
"pooler_hidden_size": 1024,
|
| 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:cbd0df4cea2fff7cb6d68864326e6e6e7dc18428fe6fa99d87e470ffa498336f
|
| 3 |
+
size 1624907048
|
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": 7.457424163818359,
|
| 14 |
+
"learning_rate": 9.166666666666666e-06,
|
| 15 |
+
"loss": 0.6324,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5,
|
| 20 |
+
"eval_sport_AUC": 0.8304120879120879,
|
| 21 |
+
"eval_sport_accuracy": 0.65,
|
| 22 |
+
"eval_sport_f1": 0.0,
|
| 23 |
+
"eval_sport_loss": 0.5250802040100098,
|
| 24 |
+
"eval_sport_macro_f1": 0.3939393939393939,
|
| 25 |
+
"eval_sport_micro_f1": 0.65,
|
| 26 |
+
"eval_sport_precision": 0.0,
|
| 27 |
+
"eval_sport_recall": 0.0,
|
| 28 |
+
"eval_sport_runtime": 1.1394,
|
| 29 |
+
"eval_sport_samples_per_second": 351.062,
|
| 30 |
+
"eval_sport_specificity": 1.0,
|
| 31 |
+
"eval_sport_steps_per_second": 0.878,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"grad_norm": 5.422190189361572,
|
| 37 |
+
"learning_rate": 8.333333333333334e-06,
|
| 38 |
+
"loss": 0.5076,
|
| 39 |
+
"step": 100
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0,
|
| 43 |
+
"eval_sport_AUC": 0.8500274725274726,
|
| 44 |
+
"eval_sport_accuracy": 0.7825,
|
| 45 |
+
"eval_sport_f1": 0.6448979591836734,
|
| 46 |
+
"eval_sport_loss": 0.5157707929611206,
|
| 47 |
+
"eval_sport_macro_f1": 0.7440706012134584,
|
| 48 |
+
"eval_sport_micro_f1": 0.7825,
|
| 49 |
+
"eval_sport_precision": 0.7523809523809524,
|
| 50 |
+
"eval_sport_recall": 0.5642857142857143,
|
| 51 |
+
"eval_sport_runtime": 1.1662,
|
| 52 |
+
"eval_sport_samples_per_second": 343.005,
|
| 53 |
+
"eval_sport_specificity": 0.9,
|
| 54 |
+
"eval_sport_steps_per_second": 0.858,
|
| 55 |
+
"step": 100
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 13.851410865783691,
|
| 60 |
+
"learning_rate": 7.500000000000001e-06,
|
| 61 |
+
"loss": 0.4243,
|
| 62 |
+
"step": 150
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.5,
|
| 66 |
+
"eval_sport_AUC": 0.8775,
|
| 67 |
+
"eval_sport_accuracy": 0.77,
|
| 68 |
+
"eval_sport_f1": 0.7341040462427746,
|
| 69 |
+
"eval_sport_loss": 0.435852974653244,
|
| 70 |
+
"eval_sport_macro_f1": 0.7657304372183036,
|
| 71 |
+
"eval_sport_micro_f1": 0.77,
|
| 72 |
+
"eval_sport_precision": 0.616504854368932,
|
| 73 |
+
"eval_sport_recall": 0.9071428571428571,
|
| 74 |
+
"eval_sport_runtime": 1.1657,
|
| 75 |
+
"eval_sport_samples_per_second": 343.153,
|
| 76 |
+
"eval_sport_specificity": 0.6961538461538461,
|
| 77 |
+
"eval_sport_steps_per_second": 0.858,
|
| 78 |
+
"step": 150
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.0,
|
| 82 |
+
"grad_norm": 13.909685134887695,
|
| 83 |
+
"learning_rate": 6.666666666666667e-06,
|
| 84 |
+
"loss": 0.4124,
|
| 85 |
+
"step": 200
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 2.0,
|
| 89 |
+
"eval_sport_AUC": 0.8836813186813187,
|
| 90 |
+
"eval_sport_accuracy": 0.8125,
|
| 91 |
+
"eval_sport_f1": 0.7232472324723247,
|
| 92 |
+
"eval_sport_loss": 0.43182089924812317,
|
| 93 |
+
"eval_sport_macro_f1": 0.7907351474270886,
|
| 94 |
+
"eval_sport_micro_f1": 0.8125,
|
| 95 |
+
"eval_sport_precision": 0.7480916030534351,
|
| 96 |
+
"eval_sport_recall": 0.7,
|
| 97 |
+
"eval_sport_runtime": 1.1592,
|
| 98 |
+
"eval_sport_samples_per_second": 345.066,
|
| 99 |
+
"eval_sport_specificity": 0.8730769230769231,
|
| 100 |
+
"eval_sport_steps_per_second": 0.863,
|
| 101 |
+
"step": 200
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.5,
|
| 105 |
+
"grad_norm": 6.286098003387451,
|
| 106 |
+
"learning_rate": 5.833333333333334e-06,
|
| 107 |
+
"loss": 0.3136,
|
| 108 |
+
"step": 250
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 2.5,
|
| 112 |
+
"eval_sport_AUC": 0.8988736263736266,
|
| 113 |
+
"eval_sport_accuracy": 0.8075,
|
| 114 |
+
"eval_sport_f1": 0.7353951890034365,
|
| 115 |
+
"eval_sport_loss": 0.410043865442276,
|
| 116 |
+
"eval_sport_macro_f1": 0.7920590876254903,
|
| 117 |
+
"eval_sport_micro_f1": 0.8075,
|
| 118 |
+
"eval_sport_precision": 0.7086092715231788,
|
| 119 |
+
"eval_sport_recall": 0.7642857142857142,
|
| 120 |
+
"eval_sport_runtime": 1.1531,
|
| 121 |
+
"eval_sport_samples_per_second": 346.882,
|
| 122 |
+
"eval_sport_specificity": 0.8307692307692308,
|
| 123 |
+
"eval_sport_steps_per_second": 0.867,
|
| 124 |
+
"step": 250
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0,
|
| 128 |
+
"grad_norm": 12.10146427154541,
|
| 129 |
+
"learning_rate": 5e-06,
|
| 130 |
+
"loss": 0.3356,
|
| 131 |
+
"step": 300
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.0,
|
| 135 |
+
"eval_sport_AUC": 0.8957967032967034,
|
| 136 |
+
"eval_sport_accuracy": 0.7925,
|
| 137 |
+
"eval_sport_f1": 0.7492447129909365,
|
| 138 |
+
"eval_sport_loss": 0.41360825300216675,
|
| 139 |
+
"eval_sport_macro_f1": 0.7861362157705216,
|
| 140 |
+
"eval_sport_micro_f1": 0.7925,
|
| 141 |
+
"eval_sport_precision": 0.6492146596858639,
|
| 142 |
+
"eval_sport_recall": 0.8857142857142857,
|
| 143 |
+
"eval_sport_runtime": 1.1571,
|
| 144 |
+
"eval_sport_samples_per_second": 345.688,
|
| 145 |
+
"eval_sport_specificity": 0.7423076923076923,
|
| 146 |
+
"eval_sport_steps_per_second": 0.864,
|
| 147 |
+
"step": 300
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 3.5,
|
| 151 |
+
"grad_norm": 20.41313362121582,
|
| 152 |
+
"learning_rate": 4.166666666666667e-06,
|
| 153 |
+
"loss": 0.1984,
|
| 154 |
+
"step": 350
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 3.5,
|
| 158 |
+
"eval_sport_AUC": 0.8906868131868132,
|
| 159 |
+
"eval_sport_accuracy": 0.8025,
|
| 160 |
+
"eval_sport_f1": 0.7375415282392026,
|
| 161 |
+
"eval_sport_loss": 0.572439968585968,
|
| 162 |
+
"eval_sport_macro_f1": 0.7896124474863347,
|
| 163 |
+
"eval_sport_micro_f1": 0.8025,
|
| 164 |
+
"eval_sport_precision": 0.6894409937888198,
|
| 165 |
+
"eval_sport_recall": 0.7928571428571428,
|
| 166 |
+
"eval_sport_runtime": 1.1428,
|
| 167 |
+
"eval_sport_samples_per_second": 350.014,
|
| 168 |
+
"eval_sport_specificity": 0.8076923076923077,
|
| 169 |
+
"eval_sport_steps_per_second": 0.875,
|
| 170 |
+
"step": 350
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.0,
|
| 174 |
+
"grad_norm": 6.701220989227295,
|
| 175 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 176 |
+
"loss": 0.2116,
|
| 177 |
+
"step": 400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 4.0,
|
| 181 |
+
"eval_sport_AUC": 0.8903571428571428,
|
| 182 |
+
"eval_sport_accuracy": 0.805,
|
| 183 |
+
"eval_sport_f1": 0.7310344827586207,
|
| 184 |
+
"eval_sport_loss": 0.5627374053001404,
|
| 185 |
+
"eval_sport_macro_f1": 0.7890466531440162,
|
| 186 |
+
"eval_sport_micro_f1": 0.805,
|
| 187 |
+
"eval_sport_precision": 0.7066666666666667,
|
| 188 |
+
"eval_sport_recall": 0.7571428571428571,
|
| 189 |
+
"eval_sport_runtime": 1.1389,
|
| 190 |
+
"eval_sport_samples_per_second": 351.223,
|
| 191 |
+
"eval_sport_specificity": 0.8307692307692308,
|
| 192 |
+
"eval_sport_steps_per_second": 0.878,
|
| 193 |
+
"step": 400
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 4.5,
|
| 197 |
+
"grad_norm": 7.881011962890625,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 0.1592,
|
| 200 |
+
"step": 450
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.5,
|
| 204 |
+
"eval_sport_AUC": 0.8919780219780219,
|
| 205 |
+
"eval_sport_accuracy": 0.815,
|
| 206 |
+
"eval_sport_f1": 0.7581699346405228,
|
| 207 |
+
"eval_sport_loss": 0.6977052092552185,
|
| 208 |
+
"eval_sport_macro_f1": 0.8041861818951602,
|
| 209 |
+
"eval_sport_micro_f1": 0.815,
|
| 210 |
+
"eval_sport_precision": 0.6987951807228916,
|
| 211 |
+
"eval_sport_recall": 0.8285714285714286,
|
| 212 |
+
"eval_sport_runtime": 1.1728,
|
| 213 |
+
"eval_sport_samples_per_second": 341.066,
|
| 214 |
+
"eval_sport_specificity": 0.8076923076923077,
|
| 215 |
+
"eval_sport_steps_per_second": 0.853,
|
| 216 |
+
"step": 450
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.0,
|
| 220 |
+
"grad_norm": 11.35710620880127,
|
| 221 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 222 |
+
"loss": 0.12,
|
| 223 |
+
"step": 500
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 5.0,
|
| 227 |
+
"eval_sport_AUC": 0.891565934065934,
|
| 228 |
+
"eval_sport_accuracy": 0.8225,
|
| 229 |
+
"eval_sport_f1": 0.7641196013289037,
|
| 230 |
+
"eval_sport_loss": 0.7499765157699585,
|
| 231 |
+
"eval_sport_macro_f1": 0.8109175160953135,
|
| 232 |
+
"eval_sport_micro_f1": 0.8225,
|
| 233 |
+
"eval_sport_precision": 0.7142857142857143,
|
| 234 |
+
"eval_sport_recall": 0.8214285714285714,
|
| 235 |
+
"eval_sport_runtime": 1.1465,
|
| 236 |
+
"eval_sport_samples_per_second": 348.894,
|
| 237 |
+
"eval_sport_specificity": 0.823076923076923,
|
| 238 |
+
"eval_sport_steps_per_second": 0.872,
|
| 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": 678669432573600.0,
|
| 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 sha256:89ea036e31dedf09ee41ba848e76f18b84cb49b9929a5f2845bdf3506e9220fa
|
| 3 |
+
size 5432
|