first commit

all_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 3.0,
+    "eval_f1": 0.6645335958651243,
+    "eval_loss": 0.6792270531125815,
+    "eval_runtime": 1.8612,
+    "eval_samples": 72,
+    "eval_samples_per_second": 38.686,
+    "eval_steps_per_second": 38.686,
+    "test_f1": 0.8574610244988863,
+    "test_loss": 0.41005876731686347,
+    "test_runtime": 1.8863,
+    "test_samples_per_second": 38.171,
+    "test_steps_per_second": 38.171,
+    "train_loss": 0.34470706577448884,
+    "train_runtime": 143.1766,
+    "train_samples": 516,
+    "train_samples_per_second": 10.812,
+    "train_steps_per_second": 10.812
+}

config.json

@@ -0,0 +1,41 @@
+{
+  "architectures": [
+    "SentenceClassificationModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classification_head": {
+    "num_labels": 1,
+    "pooling_method": "average"
+  },
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "frozen_layers": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8
+  ],
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.12.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

post-training eval_results.json

@@ -0,0 +1,8 @@
+{
+    "eval_samples": 72,
+    "test_f1": 0.8574610244988863,
+    "test_loss": 0.41005876731686347,
+    "test_runtime": 1.8863,
+    "test_samples_per_second": 38.171,
+    "test_steps_per_second": 38.171
+}

pre-training eval_results.json

@@ -0,0 +1,7 @@
+{
+    "eval_f1": 0.6645335958651243,
+    "eval_loss": 0.6792270531125815,
+    "eval_runtime": 1.8612,
+    "eval_samples_per_second": 38.686,
+    "eval_steps_per_second": 38.686
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:333c922af70e47797b9d7dd218342a55d79ae925c8f7327c293388aa505fff2f
+size 498671085

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.34470706577448884,
+    "train_runtime": 143.1766,
+    "train_samples": 516,
+    "train_samples_per_second": 10.812,
+    "train_steps_per_second": 10.812
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 1548,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.19,
+      "eval_f1": 0.849254555494202,
+      "eval_loss": 0.43653005974375836,
+      "eval_runtime": 1.8732,
+      "eval_samples_per_second": 38.438,
+      "eval_steps_per_second": 38.438,
+      "step": 100
+    },
+    {
+      "epoch": 0.39,
+      "eval_f1": 0.8503301540719003,
+      "eval_loss": 0.41970557127046154,
+      "eval_runtime": 2.0707,
+      "eval_samples_per_second": 34.771,
+      "eval_steps_per_second": 34.771,
+      "step": 200
+    },
+    {
+      "epoch": 0.58,
+      "eval_f1": 0.8297065282491516,
+      "eval_loss": 0.40088379935422785,
+      "eval_runtime": 1.8754,
+      "eval_samples_per_second": 38.393,
+      "eval_steps_per_second": 38.393,
+      "step": 300
+    },
+    {
+      "epoch": 0.78,
+      "eval_f1": 0.8422688422688422,
+      "eval_loss": 0.3884492161937947,
+      "eval_runtime": 1.9854,
+      "eval_samples_per_second": 36.264,
+      "eval_steps_per_second": 36.264,
+      "step": 400
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.3850129198966415e-05,
+      "loss": 0.4311,
+      "step": 500
+    },
+    {
+      "epoch": 0.97,
+      "eval_f1": 0.8528028933092224,
+      "eval_loss": 0.38350051814712116,
+      "eval_runtime": 1.8888,
+      "eval_samples_per_second": 38.119,
+      "eval_steps_per_second": 38.119,
+      "step": 500
+    },
+    {
+      "epoch": 1.16,
+      "eval_f1": 0.850596842510589,
+      "eval_loss": 0.37396426411921446,
+      "eval_runtime": 1.8758,
+      "eval_samples_per_second": 38.383,
+      "eval_steps_per_second": 38.383,
+      "step": 600
+    },
+    {
+      "epoch": 1.36,
+      "eval_f1": 0.8452932098765432,
+      "eval_loss": 0.4050087825284831,
+      "eval_runtime": 1.8815,
+      "eval_samples_per_second": 38.268,
+      "eval_steps_per_second": 38.268,
+      "step": 700
+    },
+    {
+      "epoch": 1.55,
+      "eval_f1": 0.8532502919423901,
+      "eval_loss": 0.3930495809773195,
+      "eval_runtime": 1.8501,
+      "eval_samples_per_second": 38.918,
+      "eval_steps_per_second": 38.918,
+      "step": 800
+    },
+    {
+      "epoch": 1.74,
+      "eval_f1": 0.8646658074019518,
+      "eval_loss": 0.37751149353174285,
+      "eval_runtime": 1.87,
+      "eval_samples_per_second": 38.504,
+      "eval_steps_per_second": 38.504,
+      "step": 900
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.7700258397932818e-05,
+      "loss": 0.3357,
+      "step": 1000
+    },
+    {
+      "epoch": 1.94,
+      "eval_f1": 0.8435221536735839,
+      "eval_loss": 0.40316357022531535,
+      "eval_runtime": 1.9305,
+      "eval_samples_per_second": 37.296,
+      "eval_steps_per_second": 37.296,
+      "step": 1000
+    },
+    {
+      "epoch": 2.13,
+      "eval_f1": 0.8601869158878505,
+      "eval_loss": 0.37796433318775113,
+      "eval_runtime": 1.8724,
+      "eval_samples_per_second": 38.453,
+      "eval_steps_per_second": 38.453,
+      "step": 1100
+    },
+    {
+      "epoch": 2.33,
+      "eval_f1": 0.8579914609244479,
+      "eval_loss": 0.4167007547913436,
+      "eval_runtime": 1.8767,
+      "eval_samples_per_second": 38.365,
+      "eval_steps_per_second": 38.365,
+      "step": 1200
+    },
+    {
+      "epoch": 2.52,
+      "eval_f1": 0.8557028187418332,
+      "eval_loss": 0.4009078407659654,
+      "eval_runtime": 1.8798,
+      "eval_samples_per_second": 38.302,
+      "eval_steps_per_second": 38.302,
+      "step": 1300
+    },
+    {
+      "epoch": 2.71,
+      "eval_f1": 0.853885616568497,
+      "eval_loss": 0.4169967998459562,
+      "eval_runtime": 1.8663,
+      "eval_samples_per_second": 38.579,
+      "eval_steps_per_second": 38.579,
+      "step": 1400
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 1.550387596899225e-06,
+      "loss": 0.2747,
+      "step": 1500
+    },
+    {
+      "epoch": 2.91,
+      "eval_f1": 0.8545181852268466,
+      "eval_loss": 0.41253904368235106,
+      "eval_runtime": 1.8671,
+      "eval_samples_per_second": 38.563,
+      "eval_steps_per_second": 38.563,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "step": 1548,
+      "total_flos": 1928644274657952.0,
+      "train_loss": 0.34470706577448884,
+      "train_runtime": 143.1766,
+      "train_samples_per_second": 10.812,
+      "train_steps_per_second": 10.812
+    }
+  ],
+  "max_steps": 1548,
+  "num_train_epochs": 3,
+  "total_flos": 1928644274657952.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:034e6e5d4ed1c3949a2059e1509ff20f730a1624b75fb393f31abdb7248cb1b3
+size 2863