Training in progress, epoch 3, checkpoint

checkpoint-300/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "microsoft/deberta-large",
+  "architectures": [
+    "DebertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 1024,
+  "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-300/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfa029c90cf38a77e4956cc89dd35cb97414c52226d09d9e310694add87a0515
+size 1624907048

checkpoint-300/trainer_state.json

@@ -0,0 +1,171 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 50,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "grad_norm": 11.574865341186523,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.6129,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_sport_AUC": 0.7282902151975048,
+      "eval_sport_accuracy": 0.7025,
+      "eval_sport_f1": 0.5734767025089605,
+      "eval_sport_loss": 0.6572355628013611,
+      "eval_sport_macro_f1": 0.6725348963600465,
+      "eval_sport_micro_f1": 0.7025,
+      "eval_sport_precision": 0.6557377049180327,
+      "eval_sport_recall": 0.5095541401273885,
+      "eval_sport_runtime": 1.1513,
+      "eval_sport_samples_per_second": 347.442,
+      "eval_sport_specificity": 0.8271604938271605,
+      "eval_sport_steps_per_second": 0.869,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.967200756072998,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.588,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_sport_AUC": 0.7831511624859111,
+      "eval_sport_accuracy": 0.745,
+      "eval_sport_f1": 0.5819672131147541,
+      "eval_sport_loss": 0.5513046979904175,
+      "eval_sport_macro_f1": 0.6992569878523411,
+      "eval_sport_micro_f1": 0.745,
+      "eval_sport_precision": 0.8160919540229885,
+      "eval_sport_recall": 0.45222929936305734,
+      "eval_sport_runtime": 1.1692,
+      "eval_sport_samples_per_second": 342.11,
+      "eval_sport_specificity": 0.934156378600823,
+      "eval_sport_steps_per_second": 0.855,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 4.148317813873291,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.4635,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_sport_AUC": 0.8006867447773324,
+      "eval_sport_accuracy": 0.78,
+      "eval_sport_f1": 0.6879432624113475,
+      "eval_sport_loss": 0.523375391960144,
+      "eval_sport_macro_f1": 0.7590295462635888,
+      "eval_sport_micro_f1": 0.78,
+      "eval_sport_precision": 0.776,
+      "eval_sport_recall": 0.6178343949044586,
+      "eval_sport_runtime": 1.1639,
+      "eval_sport_samples_per_second": 343.677,
+      "eval_sport_specificity": 0.8847736625514403,
+      "eval_sport_steps_per_second": 0.859,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 6.097774982452393,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.4886,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_sport_AUC": 0.8396896542685643,
+      "eval_sport_accuracy": 0.7675,
+      "eval_sport_f1": 0.6910299003322259,
+      "eval_sport_loss": 0.5737913250923157,
+      "eval_sport_macro_f1": 0.752328577420622,
+      "eval_sport_micro_f1": 0.7675,
+      "eval_sport_precision": 0.7222222222222222,
+      "eval_sport_recall": 0.6624203821656051,
+      "eval_sport_runtime": 1.1526,
+      "eval_sport_samples_per_second": 347.033,
+      "eval_sport_specificity": 0.8353909465020576,
+      "eval_sport_steps_per_second": 0.868,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 2.1180505752563477,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.4257,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_sport_AUC": 0.825836806374669,
+      "eval_sport_accuracy": 0.7725,
+      "eval_sport_f1": 0.6915254237288135,
+      "eval_sport_loss": 0.6375756859779358,
+      "eval_sport_macro_f1": 0.7556637019634167,
+      "eval_sport_micro_f1": 0.7725,
+      "eval_sport_precision": 0.7391304347826086,
+      "eval_sport_recall": 0.6496815286624203,
+      "eval_sport_runtime": 1.1419,
+      "eval_sport_samples_per_second": 350.281,
+      "eval_sport_specificity": 0.8518518518518519,
+      "eval_sport_steps_per_second": 0.876,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 8.99002742767334,
+      "learning_rate": 1e-05,
+      "loss": 0.4228,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_sport_AUC": 0.837107808445388,
+      "eval_sport_accuracy": 0.755,
+      "eval_sport_f1": 0.711764705882353,
+      "eval_sport_loss": 0.7621915936470032,
+      "eval_sport_macro_f1": 0.7493606138107417,
+      "eval_sport_micro_f1": 0.755,
+      "eval_sport_precision": 0.6612021857923497,
+      "eval_sport_recall": 0.7707006369426752,
+      "eval_sport_runtime": 1.1477,
+      "eval_sport_samples_per_second": 348.521,
+      "eval_sport_specificity": 0.7448559670781894,
+      "eval_sport_steps_per_second": 0.871,
+      "step": 300
+    }
+  ],
+  "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": 407201659544160.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65d0e85882425e1f397469e25cd0fb2f8d00c20c484516af4471b0ee816d77f8
+size 5432