Add model and its dependencies

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.8897058823529411,
+    "eval_combined_score": 0.9062594591867621,
+    "eval_f1": 0.9228130360205832,
+    "eval_loss": 0.802143931388855,
+    "eval_runtime": 1.9078,
+    "eval_samples": 408,
+    "eval_samples_per_second": 213.859,
+    "eval_steps_per_second": 26.732,
+    "train_loss": 0.09379986195784548,
+    "train_runtime": 1024.4983,
+    "train_samples": 3668,
+    "train_samples_per_second": 71.606,
+    "train_steps_per_second": 2.245
+}

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "google/electra-base-discriminator",
+  "architectures": [
+    "ElectraForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "embedding_size": 768,
+  "finetuning_task": "mrpc",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.9.2",
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.8897058823529411,
+    "eval_combined_score": 0.9062594591867621,
+    "eval_f1": 0.9228130360205832,
+    "eval_loss": 0.802143931388855,
+    "eval_runtime": 1.9078,
+    "eval_samples": 408,
+    "eval_samples_per_second": 213.859,
+    "eval_steps_per_second": 26.732
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36f088bc9f36722f69f23d860e503f48aabc9361379812aff9a1635d97a2cce2
+size 438020461

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "google/electra-base-discriminator", "tokenizer_class": "ElectraTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 0.09379986195784548,
+    "train_runtime": 1024.4983,
+    "train_samples": 3668,
+    "train_samples_per_second": 71.606,
+    "train_steps_per_second": 2.245
+}

trainer_state.json

@@ -0,0 +1,365 @@
+{
+  "best_metric": 0.8897058823529411,
+  "best_model_checkpoint": "./bert-base-uncased/fine_tuned_models/checkpoint-1725",
+  "epoch": 20.0,
+  "global_step": 2300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 1e-05,
+      "loss": 0.6358,
+      "step": 115
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7599857450398557,
+      "eval_f1": 0.8189910979228486,
+      "eval_loss": 0.5365363359451294,
+      "eval_runtime": 1.8712,
+      "eval_samples_per_second": 218.045,
+      "eval_steps_per_second": 27.256,
+      "step": 115
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2e-05,
+      "loss": 0.4545,
+      "step": 230
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8161764705882353,
+      "eval_combined_score": 0.84750665209541,
+      "eval_f1": 0.8788368336025848,
+      "eval_loss": 0.41578346490859985,
+      "eval_runtime": 1.8846,
+      "eval_samples_per_second": 216.486,
+      "eval_steps_per_second": 27.061,
+      "step": 230
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.888888888888889e-05,
+      "loss": 0.2835,
+      "step": 345
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8823529411764706,
+      "eval_combined_score": 0.8995098039215687,
+      "eval_f1": 0.9166666666666667,
+      "eval_loss": 0.3312576413154602,
+      "eval_runtime": 1.8685,
+      "eval_samples_per_second": 218.351,
+      "eval_steps_per_second": 27.294,
+      "step": 345
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 1.7777777777777777e-05,
+      "loss": 0.1495,
+      "step": 460
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8725490196078431,
+      "eval_combined_score": 0.891446923597025,
+      "eval_f1": 0.9103448275862069,
+      "eval_loss": 0.4420200288295746,
+      "eval_runtime": 1.8756,
+      "eval_samples_per_second": 217.526,
+      "eval_steps_per_second": 27.191,
+      "step": 460
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.103,
+      "step": 575
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8676470588235294,
+      "eval_combined_score": 0.8840080312567832,
+      "eval_f1": 0.9003690036900369,
+      "eval_loss": 0.5437020063400269,
+      "eval_runtime": 1.8749,
+      "eval_samples_per_second": 217.61,
+      "eval_steps_per_second": 27.201,
+      "step": 575
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 1.555555555555556e-05,
+      "loss": 0.0683,
+      "step": 690
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8627450980392157,
+      "eval_combined_score": 0.8835909790537375,
+      "eval_f1": 0.9044368600682594,
+      "eval_loss": 0.6472938656806946,
+      "eval_runtime": 1.8885,
+      "eval_samples_per_second": 216.048,
+      "eval_steps_per_second": 27.006,
+      "step": 690
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 1.4444444444444446e-05,
+      "loss": 0.053,
+      "step": 805
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8676470588235294,
+      "eval_combined_score": 0.8877484440875327,
+      "eval_f1": 0.9078498293515359,
+      "eval_loss": 0.5814068913459778,
+      "eval_runtime": 1.8738,
+      "eval_samples_per_second": 217.736,
+      "eval_steps_per_second": 27.217,
+      "step": 805
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.0292,
+      "step": 920
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8578431372549019,
+      "eval_combined_score": 0.880426585349859,
+      "eval_f1": 0.903010033444816,
+      "eval_loss": 0.7452187538146973,
+      "eval_runtime": 1.8877,
+      "eval_samples_per_second": 216.139,
+      "eval_steps_per_second": 27.017,
+      "step": 920
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.2222222222222224e-05,
+      "loss": 0.0212,
+      "step": 1035
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.8774509803921569,
+      "eval_combined_score": 0.8951714832274373,
+      "eval_f1": 0.9128919860627178,
+      "eval_loss": 0.683774471282959,
+      "eval_runtime": 1.8947,
+      "eval_samples_per_second": 215.341,
+      "eval_steps_per_second": 26.918,
+      "step": 1035
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.1111111111111113e-05,
+      "loss": 0.0103,
+      "step": 1150
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.8602941176470589,
+      "eval_combined_score": 0.8814291101055808,
+      "eval_f1": 0.9025641025641027,
+      "eval_loss": 0.789932906627655,
+      "eval_runtime": 1.8751,
+      "eval_samples_per_second": 217.593,
+      "eval_steps_per_second": 27.199,
+      "step": 1150
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 1e-05,
+      "loss": 0.0153,
+      "step": 1265
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.8676470588235294,
+      "eval_combined_score": 0.8859511889862328,
+      "eval_f1": 0.9042553191489361,
+      "eval_loss": 0.7904257774353027,
+      "eval_runtime": 1.8798,
+      "eval_samples_per_second": 217.048,
+      "eval_steps_per_second": 27.131,
+      "step": 1265
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 8.888888888888888e-06,
+      "loss": 0.0136,
+      "step": 1380
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.8799019607843137,
+      "eval_combined_score": 0.8980706385118149,
+      "eval_f1": 0.9162393162393162,
+      "eval_loss": 0.6983678340911865,
+      "eval_runtime": 1.8896,
+      "eval_samples_per_second": 215.921,
+      "eval_steps_per_second": 26.99,
+      "step": 1380
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 7.77777777777778e-06,
+      "loss": 0.0083,
+      "step": 1495
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.8823529411764706,
+      "eval_combined_score": 0.9002208391889179,
+      "eval_f1": 0.9180887372013652,
+      "eval_loss": 0.7833622097969055,
+      "eval_runtime": 1.8915,
+      "eval_samples_per_second": 215.705,
+      "eval_steps_per_second": 26.963,
+      "step": 1495
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.0078,
+      "step": 1610
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.875,
+      "eval_combined_score": 0.8939102564102563,
+      "eval_f1": 0.9128205128205128,
+      "eval_loss": 0.8284361362457275,
+      "eval_runtime": 1.8794,
+      "eval_samples_per_second": 217.087,
+      "eval_steps_per_second": 27.136,
+      "step": 1610
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 5.555555555555557e-06,
+      "loss": 0.0065,
+      "step": 1725
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.8897058823529411,
+      "eval_combined_score": 0.9062594591867621,
+      "eval_f1": 0.9228130360205832,
+      "eval_loss": 0.802143931388855,
+      "eval_runtime": 1.8792,
+      "eval_samples_per_second": 217.111,
+      "eval_steps_per_second": 27.139,
+      "step": 1725
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 4.444444444444444e-06,
+      "loss": 0.0046,
+      "step": 1840
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.8725490196078431,
+      "eval_combined_score": 0.8922067131937521,
+      "eval_f1": 0.9118644067796611,
+      "eval_loss": 0.890489399433136,
+      "eval_runtime": 1.8786,
+      "eval_samples_per_second": 217.185,
+      "eval_steps_per_second": 27.148,
+      "step": 1840
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.005,
+      "step": 1955
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.875,
+      "eval_combined_score": 0.8931521739130435,
+      "eval_f1": 0.9113043478260869,
+      "eval_loss": 0.8655802607536316,
+      "eval_runtime": 1.8821,
+      "eval_samples_per_second": 216.775,
+      "eval_steps_per_second": 27.097,
+      "step": 1955
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 2.222222222222222e-06,
+      "loss": 0.004,
+      "step": 2070
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.8700980392156863,
+      "eval_combined_score": 0.889280453113888,
+      "eval_f1": 0.9084628670120898,
+      "eval_loss": 0.9055202007293701,
+      "eval_runtime": 1.8752,
+      "eval_samples_per_second": 217.58,
+      "eval_steps_per_second": 27.198,
+      "step": 2070
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 1.111111111111111e-06,
+      "loss": 0.0019,
+      "step": 2185
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.8725490196078431,
+      "eval_combined_score": 0.8911356209150327,
+      "eval_f1": 0.9097222222222222,
+      "eval_loss": 0.9129809141159058,
+      "eval_runtime": 1.8883,
+      "eval_samples_per_second": 216.063,
+      "eval_steps_per_second": 27.008,
+      "step": 2185
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0,
+      "loss": 0.0008,
+      "step": 2300
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.8725490196078431,
+      "eval_combined_score": 0.8916009702850212,
+      "eval_f1": 0.9106529209621993,
+      "eval_loss": 0.9021580815315247,
+      "eval_runtime": 1.8731,
+      "eval_samples_per_second": 217.821,
+      "eval_steps_per_second": 27.228,
+      "step": 2300
+    },
+    {
+      "epoch": 20.0,
+      "step": 2300,
+      "total_flos": 4825456755302400.0,
+      "train_loss": 0.09379986195784548,
+      "train_runtime": 1024.4983,
+      "train_samples_per_second": 71.606,
+      "train_steps_per_second": 2.245
+    }
+  ],
+  "max_steps": 2300,
+  "num_train_epochs": 20,
+  "total_flos": 4825456755302400.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:804be5ec229e6a67e2a64e66260a5d1fa7842a43f9b93b7ffc78c6b5af0f13f5
+size 2671