Training in progress, epoch 4, checkpoint

checkpoint-400/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "microsoft/deberta-base",
+  "architectures": [
+    "DebertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "c2p",
+    "p2c"
+  ],
+  "position_biased_input": false,
+  "relative_attention": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.3",
+  "type_vocab_size": 0,
+  "vocab_size": 50265
+}

checkpoint-400/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fc603ea5ed4207eefa60db0422a6c58cc58cbbad28c5153553f0520bf197473
+size 556799560

checkpoint-400/trainer_state.json

@@ -0,0 +1,217 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 50,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "grad_norm": 6.658211708068848,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.5628,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_sport_AUC": 0.8575549450549451,
+      "eval_sport_accuracy": 0.795,
+      "eval_sport_f1": 0.6917293233082706,
+      "eval_sport_loss": 0.45002254843711853,
+      "eval_sport_macro_f1": 0.7690856354369069,
+      "eval_sport_micro_f1": 0.795,
+      "eval_sport_precision": 0.7301587301587301,
+      "eval_sport_recall": 0.6571428571428571,
+      "eval_sport_runtime": 0.6957,
+      "eval_sport_samples_per_second": 574.989,
+      "eval_sport_specificity": 0.8692307692307693,
+      "eval_sport_steps_per_second": 1.437,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 7.818618297576904,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.4453,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_sport_AUC": 0.8724725274725275,
+      "eval_sport_accuracy": 0.79,
+      "eval_sport_f1": 0.7162162162162162,
+      "eval_sport_loss": 0.42653143405914307,
+      "eval_sport_macro_f1": 0.7747747747747749,
+      "eval_sport_micro_f1": 0.79,
+      "eval_sport_precision": 0.6794871794871795,
+      "eval_sport_recall": 0.7571428571428571,
+      "eval_sport_runtime": 0.6966,
+      "eval_sport_samples_per_second": 574.228,
+      "eval_sport_specificity": 0.8076923076923077,
+      "eval_sport_steps_per_second": 1.436,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 11.515430450439453,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.3635,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_sport_AUC": 0.867912087912088,
+      "eval_sport_accuracy": 0.8,
+      "eval_sport_f1": 0.7202797202797203,
+      "eval_sport_loss": 0.4973774254322052,
+      "eval_sport_macro_f1": 0.7823188484667084,
+      "eval_sport_micro_f1": 0.8,
+      "eval_sport_precision": 0.7054794520547946,
+      "eval_sport_recall": 0.7357142857142858,
+      "eval_sport_runtime": 0.7061,
+      "eval_sport_samples_per_second": 566.501,
+      "eval_sport_specificity": 0.8346153846153846,
+      "eval_sport_steps_per_second": 1.416,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 14.943413734436035,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.362,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_sport_AUC": 0.8794505494505495,
+      "eval_sport_accuracy": 0.78,
+      "eval_sport_f1": 0.7349397590361446,
+      "eval_sport_loss": 0.5080252885818481,
+      "eval_sport_macro_f1": 0.7734527855009783,
+      "eval_sport_micro_f1": 0.78,
+      "eval_sport_precision": 0.6354166666666666,
+      "eval_sport_recall": 0.8714285714285714,
+      "eval_sport_runtime": 0.7311,
+      "eval_sport_samples_per_second": 547.097,
+      "eval_sport_specificity": 0.7307692307692307,
+      "eval_sport_steps_per_second": 1.368,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 1.4722031354904175,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.2393,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_sport_AUC": 0.879532967032967,
+      "eval_sport_accuracy": 0.7925,
+      "eval_sport_f1": 0.7331189710610932,
+      "eval_sport_loss": 0.519324779510498,
+      "eval_sport_macro_f1": 0.781692409865925,
+      "eval_sport_micro_f1": 0.7925,
+      "eval_sport_precision": 0.6666666666666666,
+      "eval_sport_recall": 0.8142857142857143,
+      "eval_sport_runtime": 0.6948,
+      "eval_sport_samples_per_second": 575.731,
+      "eval_sport_specificity": 0.7807692307692308,
+      "eval_sport_steps_per_second": 1.439,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 7.780460357666016,
+      "learning_rate": 1e-05,
+      "loss": 0.2739,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_sport_AUC": 0.8780769230769231,
+      "eval_sport_accuracy": 0.7925,
+      "eval_sport_f1": 0.7446153846153846,
+      "eval_sport_loss": 0.5590013861656189,
+      "eval_sport_macro_f1": 0.7849392712550607,
+      "eval_sport_micro_f1": 0.7925,
+      "eval_sport_precision": 0.654054054054054,
+      "eval_sport_recall": 0.8642857142857143,
+      "eval_sport_runtime": 0.7031,
+      "eval_sport_samples_per_second": 568.879,
+      "eval_sport_specificity": 0.7538461538461538,
+      "eval_sport_steps_per_second": 1.422,
+      "step": 300
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 10.377619743347168,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.165,
+      "step": 350
+    },
+    {
+      "epoch": 3.5,
+      "eval_sport_AUC": 0.8746153846153847,
+      "eval_sport_accuracy": 0.7975,
+      "eval_sport_f1": 0.7412140575079872,
+      "eval_sport_loss": 0.5987029075622559,
+      "eval_sport_macro_f1": 0.787444811094856,
+      "eval_sport_micro_f1": 0.7975,
+      "eval_sport_precision": 0.6705202312138728,
+      "eval_sport_recall": 0.8285714285714286,
+      "eval_sport_runtime": 0.6942,
+      "eval_sport_samples_per_second": 576.207,
+      "eval_sport_specificity": 0.7807692307692308,
+      "eval_sport_steps_per_second": 1.441,
+      "step": 350
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.792548179626465,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.1496,
+      "step": 400
+    },
+    {
+      "epoch": 4.0,
+      "eval_sport_AUC": 0.8720329670329671,
+      "eval_sport_accuracy": 0.7975,
+      "eval_sport_f1": 0.7428571428571429,
+      "eval_sport_loss": 0.7307766675949097,
+      "eval_sport_macro_f1": 0.7879234167893963,
+      "eval_sport_micro_f1": 0.7975,
+      "eval_sport_precision": 0.6685714285714286,
+      "eval_sport_recall": 0.8357142857142857,
+      "eval_sport_runtime": 0.6967,
+      "eval_sport_samples_per_second": 574.17,
+      "eval_sport_specificity": 0.7769230769230769,
+      "eval_sport_steps_per_second": 1.435,
+      "step": 400
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 600,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 153202987866240.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1bc94fc718c18b1d22783932c9849ceddacd004e03b9e5fce890a3ddf29409c
+size 5432