Best F1=0.7795 (trial 10)

Hyper‑parameters: {'learning_rate': 4.437063943360846e-06, 'weight_decay': 0.0017391423354692903, 'batch_size': 32, 'warmup_ratio': 0.09464137213269672, 'hidden_dropout': 0.053138253462458006, 'attn_dropout': 0.2735895207101088, 'classifier_dropout': 0.39340169178729456, 'epoch': 10.447761194029852}

config.json

@@ -0,0 +1,35 @@
+{
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.2735895207101088,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.39340169178729456,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.053138253462458006,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "M",
+    "1": "NM"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "M": 0,
+    "NM": 1
+  },
+  "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": "4.51.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50262
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:238d8f17bfba9751b89dd4276821eff7d5046748d169b1e8df6722e19e88f51e
+size 498603608

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35a6e50574846c23126f8c5408f5a1047e59071de247c7c43c5f610c8750e64b
+size 997327098

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dcb5238779ec6957f466f5f3ad7dd55e13d492f8d41fbd8eb9e576c04c93a1e
+size 14244

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:042e25180ae7eb4b4b781d7cbe3a4bed50b71b320d1307baca737eadff985a68
+size 988

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53d20ac6487a3cfdeeb7da204127eda74e631c6c06e14703d5e999cedbeedc93
+size 1064

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,59 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "max_len": 512,
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,309 @@
+{
+  "best_global_step": 1400,
+  "best_metric": 0.7795431976166831,
+  "best_model_checkpoint": "./hpo_PlanTL-GOB-ES_roberta-base-bne/trial_10/checkpoint-1400",
+  "epoch": 10.447761194029852,
+  "eval_steps": 100,
+  "global_step": 1400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.746268656716418,
+      "grad_norm": 1.1734787225723267,
+      "learning_rate": 2.299839426139915e-06,
+      "loss": 0.6959,
+      "step": 100
+    },
+    {
+      "epoch": 0.746268656716418,
+      "eval_accuracy": 0.6081081081081081,
+      "eval_f1": 0.442484773258667,
+      "eval_loss": 0.6894811391830444,
+      "eval_precision": 0.4994814622763806,
+      "eval_recall": 0.4998257839721254,
+      "eval_runtime": 0.3253,
+      "eval_samples_per_second": 682.392,
+      "eval_steps_per_second": 21.517,
+      "step": 100
+    },
+    {
+      "epoch": 1.4925373134328357,
+      "grad_norm": 2.857347011566162,
+      "learning_rate": 4.41754964344502e-06,
+      "loss": 0.6866,
+      "step": 200
+    },
+    {
+      "epoch": 1.4925373134328357,
+      "eval_accuracy": 0.6936936936936937,
+      "eval_f1": 0.6659290077011596,
+      "eval_loss": 0.6791544556617737,
+      "eval_precision": 0.6693403028118241,
+      "eval_recall": 0.6636759581881533,
+      "eval_runtime": 0.3219,
+      "eval_samples_per_second": 689.684,
+      "eval_steps_per_second": 21.747,
+      "step": 200
+    },
+    {
+      "epoch": 2.2388059701492535,
+      "grad_norm": 5.069456100463867,
+      "learning_rate": 4.176060181986678e-06,
+      "loss": 0.6555,
+      "step": 300
+    },
+    {
+      "epoch": 2.2388059701492535,
+      "eval_accuracy": 0.6981981981981982,
+      "eval_f1": 0.5875093596605563,
+      "eval_loss": 0.5988879799842834,
+      "eval_precision": 0.7426395939086294,
+      "eval_recall": 0.6040940766550522,
+      "eval_runtime": 0.3255,
+      "eval_samples_per_second": 681.953,
+      "eval_steps_per_second": 21.503,
+      "step": 300
+    },
+    {
+      "epoch": 2.9850746268656714,
+      "grad_norm": 5.771562099456787,
+      "learning_rate": 3.9321314330388585e-06,
+      "loss": 0.5745,
+      "step": 400
+    },
+    {
+      "epoch": 2.9850746268656714,
+      "eval_accuracy": 0.7252252252252253,
+      "eval_f1": 0.637943480469481,
+      "eval_loss": 0.5637385249137878,
+      "eval_precision": 0.772757980070934,
+      "eval_recall": 0.6406794425087108,
+      "eval_runtime": 0.3245,
+      "eval_samples_per_second": 684.169,
+      "eval_steps_per_second": 21.573,
+      "step": 400
+    },
+    {
+      "epoch": 3.7313432835820897,
+      "grad_norm": 6.657003879547119,
+      "learning_rate": 3.6882026840910388e-06,
+      "loss": 0.5315,
+      "step": 500
+    },
+    {
+      "epoch": 3.7313432835820897,
+      "eval_accuracy": 0.7612612612612613,
+      "eval_f1": 0.724597991714065,
+      "eval_loss": 0.5057556629180908,
+      "eval_precision": 0.7563169387542893,
+      "eval_recall": 0.7147212543554007,
+      "eval_runtime": 0.3235,
+      "eval_samples_per_second": 686.192,
+      "eval_steps_per_second": 21.637,
+      "step": 500
+    },
+    {
+      "epoch": 4.477611940298507,
+      "grad_norm": 6.151370048522949,
+      "learning_rate": 3.4442739351432187e-06,
+      "loss": 0.4973,
+      "step": 600
+    },
+    {
+      "epoch": 4.477611940298507,
+      "eval_accuracy": 0.7747747747747747,
+      "eval_f1": 0.7448745058380069,
+      "eval_loss": 0.4851745069026947,
+      "eval_precision": 0.7674050632911392,
+      "eval_recall": 0.7355400696864112,
+      "eval_runtime": 0.3262,
+      "eval_samples_per_second": 680.614,
+      "eval_steps_per_second": 21.461,
+      "step": 600
+    },
+    {
+      "epoch": 5.223880597014926,
+      "grad_norm": 10.251242637634277,
+      "learning_rate": 3.2003451861953985e-06,
+      "loss": 0.4523,
+      "step": 700
+    },
+    {
+      "epoch": 5.223880597014926,
+      "eval_accuracy": 0.7567567567567568,
+      "eval_f1": 0.7204551814196436,
+      "eval_loss": 0.48035961389541626,
+      "eval_precision": 0.7493827160493827,
+      "eval_recall": 0.7111498257839721,
+      "eval_runtime": 0.3234,
+      "eval_samples_per_second": 686.473,
+      "eval_steps_per_second": 21.646,
+      "step": 700
+    },
+    {
+      "epoch": 5.970149253731344,
+      "grad_norm": 5.884487628936768,
+      "learning_rate": 2.9564164372475784e-06,
+      "loss": 0.4519,
+      "step": 800
+    },
+    {
+      "epoch": 5.970149253731344,
+      "eval_accuracy": 0.7612612612612613,
+      "eval_f1": 0.7444118605408928,
+      "eval_loss": 0.4888066351413727,
+      "eval_precision": 0.7438242177342463,
+      "eval_recall": 0.7450348432055749,
+      "eval_runtime": 0.3288,
+      "eval_samples_per_second": 675.222,
+      "eval_steps_per_second": 21.291,
+      "step": 800
+    },
+    {
+      "epoch": 6.7164179104477615,
+      "grad_norm": 5.6701226234436035,
+      "learning_rate": 2.7124876882997587e-06,
+      "loss": 0.4267,
+      "step": 900
+    },
+    {
+      "epoch": 6.7164179104477615,
+      "eval_accuracy": 0.7522522522522522,
+      "eval_f1": 0.7183715834390497,
+      "eval_loss": 0.470042884349823,
+      "eval_precision": 0.7407906558849955,
+      "eval_recall": 0.7101045296167248,
+      "eval_runtime": 0.3283,
+      "eval_samples_per_second": 676.253,
+      "eval_steps_per_second": 21.323,
+      "step": 900
+    },
+    {
+      "epoch": 7.462686567164179,
+      "grad_norm": 14.518937110900879,
+      "learning_rate": 2.4685589393519386e-06,
+      "loss": 0.4149,
+      "step": 1000
+    },
+    {
+      "epoch": 7.462686567164179,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_f1": 0.7628205128205128,
+      "eval_loss": 0.47342488169670105,
+      "eval_precision": 0.7702702702702703,
+      "eval_recall": 0.7578397212543554,
+      "eval_runtime": 0.3222,
+      "eval_samples_per_second": 688.983,
+      "eval_steps_per_second": 21.725,
+      "step": 1000
+    },
+    {
+      "epoch": 8.208955223880597,
+      "grad_norm": 6.513219356536865,
+      "learning_rate": 2.2246301904041184e-06,
+      "loss": 0.3897,
+      "step": 1100
+    },
+    {
+      "epoch": 8.208955223880597,
+      "eval_accuracy": 0.7702702702702703,
+      "eval_f1": 0.7501489572989076,
+      "eval_loss": 0.4738766849040985,
+      "eval_precision": 0.7540976265114196,
+      "eval_recall": 0.7471254355400697,
+      "eval_runtime": 0.3272,
+      "eval_samples_per_second": 678.418,
+      "eval_steps_per_second": 21.392,
+      "step": 1100
+    },
+    {
+      "epoch": 8.955223880597014,
+      "grad_norm": 19.53132438659668,
+      "learning_rate": 1.9807014414562983e-06,
+      "loss": 0.3806,
+      "step": 1200
+    },
+    {
+      "epoch": 8.955223880597014,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_f1": 0.759913482335977,
+      "eval_loss": 0.46905937790870667,
+      "eval_precision": 0.7727443609022557,
+      "eval_recall": 0.7527874564459931,
+      "eval_runtime": 0.3253,
+      "eval_samples_per_second": 682.365,
+      "eval_steps_per_second": 21.516,
+      "step": 1200
+    },
+    {
+      "epoch": 9.701492537313433,
+      "grad_norm": 12.14911937713623,
+      "learning_rate": 1.7367726925084784e-06,
+      "loss": 0.3609,
+      "step": 1300
+    },
+    {
+      "epoch": 9.701492537313433,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_f1": 0.7679442508710801,
+      "eval_loss": 0.48617300391197205,
+      "eval_precision": 0.7679442508710801,
+      "eval_recall": 0.7679442508710801,
+      "eval_runtime": 0.3248,
+      "eval_samples_per_second": 683.448,
+      "eval_steps_per_second": 21.55,
+      "step": 1300
+    },
+    {
+      "epoch": 10.447761194029852,
+      "grad_norm": 15.257988929748535,
+      "learning_rate": 1.4928439435606583e-06,
+      "loss": 0.3456,
+      "step": 1400
+    },
+    {
+      "epoch": 10.447761194029852,
+      "eval_accuracy": 0.7972972972972973,
+      "eval_f1": 0.7795431976166831,
+      "eval_loss": 0.4786612093448639,
+      "eval_precision": 0.783922973578146,
+      "eval_recall": 0.7761324041811846,
+      "eval_runtime": 0.3278,
+      "eval_samples_per_second": 677.161,
+      "eval_steps_per_second": 21.352,
+      "step": 1400
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 4,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.17373841796096e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99f3a3c337e0c63af9fc8d9493048fda0d36e478d1b317c852330539661cee96
+size 5304