Upload folder using huggingface_hub

training_checkpoints/checkpoint-5144/config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "nlpaueb/legal-bert-base-uncased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_ids": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Issue",
+    "1": "Court Discourse",
+    "2": "Conclusion",
+    "3": "Precedent Analysis",
+    "4": "Section Analysis",
+    "5": "Argument by Petitioner",
+    "6": "Fact",
+    "7": "Argument by Respondent",
+    "8": "Ratio",
+    "9": "Appellant",
+    "10": "Respondent",
+    "11": "Argument by Appellant",
+    "12": "Petitioner",
+    "13": "Judge",
+    "14": "Argument by Defendant"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Appellant": 9,
+    "Argument by Appellant": 11,
+    "Argument by Defendant": 14,
+    "Argument by Petitioner": 5,
+    "Argument by Respondent": 7,
+    "Conclusion": 2,
+    "Court Discourse": 1,
+    "Fact": 6,
+    "Issue": 0,
+    "Judge": 13,
+    "Petitioner": 12,
+    "Precedent Analysis": 3,
+    "Ratio": 8,
+    "Respondent": 10,
+    "Section Analysis": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

training_checkpoints/checkpoint-5144/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc809df5989df6850641fc9ec9044908a2f8ec6cf0c83c25b3c4e05b63803fa1
+size 437998636

training_checkpoints/checkpoint-5144/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca563db4fdd59e78c2dfe9f637f195786445b90d4ce2ded402862869ce911d99
+size 876118266

training_checkpoints/checkpoint-5144/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30f025f06094c8276db0b4600a0cb1cf1bc7f8a1a91d94cd8bb4a99db5ca2c39
+size 14244

training_checkpoints/checkpoint-5144/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8aa68408c126f1c30bfd7fb24462954f9213b8c6066139d17139e84482d913a1
+size 1064

training_checkpoints/checkpoint-5144/trainer_state.json

@@ -0,0 +1,267 @@
+{
+  "best_metric": 0.7839844036299938,
+  "best_model_checkpoint": "legal-bert-base-uncased/checkpoint-3858",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 5144,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.78,
+      "grad_norm": 6.735908031463623,
+      "learning_rate": 4.742353551062727e-05,
+      "loss": 1.1096,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7978311386522076,
+      "eval_f1_macro": 0.613418966848975,
+      "eval_f1_micro": 0.7978311386522076,
+      "eval_f1_weighted": 0.7839880886224406,
+      "eval_loss": 0.6747979521751404,
+      "eval_macro_fpr": 0.018830502815301575,
+      "eval_macro_sensitivity": 0.6339578261148517,
+      "eval_macro_specificity": 0.9845345883924012,
+      "eval_precision": 0.7854860044786733,
+      "eval_precision_macro": 0.6239440349375213,
+      "eval_recall": 0.7978311386522076,
+      "eval_recall_macro": 0.6339578261148517,
+      "eval_runtime": 30.9921,
+      "eval_samples_per_second": 41.656,
+      "eval_steps_per_second": 5.227,
+      "eval_weighted_fpr": 0.017778080512226686,
+      "eval_weighted_sensitivity": 0.7978311386522076,
+      "eval_weighted_specificity": 0.9701876872338092,
+      "step": 643
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 6.824150562286377,
+      "learning_rate": 4.48315189217211e-05,
+      "loss": 0.6187,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8109992254066615,
+      "eval_f1_macro": 0.7268228216676939,
+      "eval_f1_micro": 0.8109992254066615,
+      "eval_f1_weighted": 0.8089875323563979,
+      "eval_loss": 0.6449354887008667,
+      "eval_macro_fpr": 0.016938802987367406,
+      "eval_macro_sensitivity": 0.7327307314578203,
+      "eval_macro_specificity": 0.9857686909763926,
+      "eval_precision": 0.8196084470842065,
+      "eval_precision_macro": 0.7805936449528967,
+      "eval_recall": 0.8109992254066615,
+      "eval_recall_macro": 0.7327307314578203,
+      "eval_runtime": 58.3071,
+      "eval_samples_per_second": 22.141,
+      "eval_steps_per_second": 2.778,
+      "eval_weighted_fpr": 0.016373641121997046,
+      "eval_weighted_sensitivity": 0.8109992254066615,
+      "eval_weighted_specificity": 0.9755311392392266,
+      "step": 1286
+    },
+    {
+      "epoch": 2.33,
+      "grad_norm": 20.75279998779297,
+      "learning_rate": 4.223950233281493e-05,
+      "loss": 0.4747,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.814872192099148,
+      "eval_f1_macro": 0.7369957890904685,
+      "eval_f1_micro": 0.814872192099148,
+      "eval_f1_weighted": 0.8124596447054482,
+      "eval_loss": 0.8151384592056274,
+      "eval_macro_fpr": 0.016571707879599946,
+      "eval_macro_sensitivity": 0.739039840019417,
+      "eval_macro_specificity": 0.9860629698102752,
+      "eval_precision": 0.8192276160546627,
+      "eval_precision_macro": 0.7659183928188651,
+      "eval_recall": 0.814872192099148,
+      "eval_recall_macro": 0.739039840019417,
+      "eval_runtime": 55.9094,
+      "eval_samples_per_second": 23.091,
+      "eval_steps_per_second": 2.898,
+      "eval_weighted_fpr": 0.015968463954032203,
+      "eval_weighted_sensitivity": 0.814872192099148,
+      "eval_weighted_specificity": 0.9760723550549795,
+      "step": 1929
+    },
+    {
+      "epoch": 3.11,
+      "grad_norm": 10.737870216369629,
+      "learning_rate": 3.9652669777086575e-05,
+      "loss": 0.3866,
+      "step": 2000
+    },
+    {
+      "epoch": 3.89,
+      "grad_norm": 13.75886344909668,
+      "learning_rate": 3.7060653188180404e-05,
+      "loss": 0.2645,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8218435321456236,
+      "eval_f1_macro": 0.738478964900184,
+      "eval_f1_micro": 0.8218435321456236,
+      "eval_f1_weighted": 0.8189030195253059,
+      "eval_loss": 0.9344585537910461,
+      "eval_macro_fpr": 0.015763618193954002,
+      "eval_macro_sensitivity": 0.7412892486668318,
+      "eval_macro_specificity": 0.986618111597957,
+      "eval_precision": 0.8198244063808544,
+      "eval_precision_macro": 0.7446229748697999,
+      "eval_recall": 0.8218435321456236,
+      "eval_recall_macro": 0.7412892486668318,
+      "eval_runtime": 75.578,
+      "eval_samples_per_second": 17.082,
+      "eval_steps_per_second": 2.143,
+      "eval_weighted_fpr": 0.015247944842216919,
+      "eval_weighted_sensitivity": 0.8218435321456236,
+      "eval_weighted_specificity": 0.9774281418237326,
+      "step": 2572
+    },
+    {
+      "epoch": 4.67,
+      "grad_norm": 8.32991886138916,
+      "learning_rate": 3.4468636599274234e-05,
+      "loss": 0.1901,
+      "step": 3000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8195197521301317,
+      "eval_f1_macro": 0.7594855831156184,
+      "eval_f1_micro": 0.8195197521301317,
+      "eval_f1_weighted": 0.81658937168797,
+      "eval_loss": 1.0929232835769653,
+      "eval_macro_fpr": 0.01614879224826684,
+      "eval_macro_sensitivity": 0.743232314276478,
+      "eval_macro_specificity": 0.9863035723232563,
+      "eval_precision": 0.8241924882578779,
+      "eval_precision_macro": 0.8264084869739744,
+      "eval_recall": 0.8195197521301317,
+      "eval_recall_macro": 0.743232314276478,
+      "eval_runtime": 65.7326,
+      "eval_samples_per_second": 19.64,
+      "eval_steps_per_second": 2.465,
+      "eval_weighted_fpr": 0.01548687271518777,
+      "eval_weighted_sensitivity": 0.8195197521301317,
+      "eval_weighted_specificity": 0.9750338327187125,
+      "step": 3215
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 50.74209976196289,
+      "learning_rate": 3.187662001036807e-05,
+      "loss": 0.1131,
+      "step": 3500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.820294345468629,
+      "eval_f1_macro": 0.7839844036299938,
+      "eval_f1_micro": 0.8202943454686291,
+      "eval_f1_weighted": 0.8197212343639901,
+      "eval_loss": 1.1536158323287964,
+      "eval_macro_fpr": 0.01593810671400651,
+      "eval_macro_sensitivity": 0.7786487516932519,
+      "eval_macro_specificity": 0.9864617261648754,
+      "eval_precision": 0.821246410485655,
+      "eval_precision_macro": 0.7968460348821496,
+      "eval_recall": 0.820294345468629,
+      "eval_recall_macro": 0.7786487516932519,
+      "eval_runtime": 54.9389,
+      "eval_samples_per_second": 23.499,
+      "eval_steps_per_second": 2.949,
+      "eval_weighted_fpr": 0.015407092575375215,
+      "eval_weighted_sensitivity": 0.820294345468629,
+      "eval_weighted_specificity": 0.9766315470045015,
+      "step": 3858
+    },
+    {
+      "epoch": 6.22,
+      "grad_norm": 0.02055634744465351,
+      "learning_rate": 2.928978745463971e-05,
+      "loss": 0.0975,
+      "step": 4000
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 32.151573181152344,
+      "learning_rate": 2.669777086573354e-05,
+      "loss": 0.063,
+      "step": 4500
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8117738187451587,
+      "eval_f1_macro": 0.7495479890901372,
+      "eval_f1_micro": 0.8117738187451587,
+      "eval_f1_weighted": 0.8135965211162124,
+      "eval_loss": 1.3217942714691162,
+      "eval_macro_fpr": 0.01664162427116977,
+      "eval_macro_sensitivity": 0.7526435450791772,
+      "eval_macro_specificity": 0.9859379584321952,
+      "eval_precision": 0.818420676521013,
+      "eval_precision_macro": 0.7517839435012339,
+      "eval_recall": 0.8117738187451587,
+      "eval_recall_macro": 0.7526435450791772,
+      "eval_runtime": 59.9636,
+      "eval_samples_per_second": 21.53,
+      "eval_steps_per_second": 2.702,
+      "eval_weighted_fpr": 0.016292323164599398,
+      "eval_weighted_sensitivity": 0.8117738187451587,
+      "eval_weighted_specificity": 0.9772955577377688,
+      "step": 4501
+    },
+    {
+      "epoch": 7.78,
+      "grad_norm": 0.0045046876184642315,
+      "learning_rate": 2.4105754276827372e-05,
+      "loss": 0.0264,
+      "step": 5000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.82571649883811,
+      "eval_f1_macro": 0.7730278469427766,
+      "eval_f1_micro": 0.82571649883811,
+      "eval_f1_weighted": 0.8247069887539016,
+      "eval_loss": 1.3862823247909546,
+      "eval_macro_fpr": 0.015458317393184977,
+      "eval_macro_sensitivity": 0.7767515241074096,
+      "eval_macro_specificity": 0.9868314420976501,
+      "eval_precision": 0.8261549954416628,
+      "eval_precision_macro": 0.7783699125562222,
+      "eval_recall": 0.82571649883811,
+      "eval_recall_macro": 0.7767515241074096,
+      "eval_runtime": 58.9779,
+      "eval_samples_per_second": 21.89,
+      "eval_steps_per_second": 2.747,
+      "eval_weighted_fpr": 0.01485246550927454,
+      "eval_weighted_sensitivity": 0.82571649883811,
+      "eval_weighted_specificity": 0.97675513262664,
+      "step": 5144
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9645,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 1.0816179155582976e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_checkpoints/checkpoint-5144/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21938ff48cdf92422f693d69544854f8291bee6a4c7091c49d48bd0e59616530
+size 4920