Day2

This model is a fine-tuned version of nlpaueb/legal-bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.4956
• eval_Accuracy: 0.9035
• eval_F1_macro: 0.5229
• eval_F1_class_0: 0.0
• eval_F1_class_1: 0.0
• eval_F1_class_2: 0.9289
• eval_F1_class_3: 0.6667
• eval_F1_class_4: 0.9910
• eval_F1_class_5: 0.6667
• eval_F1_class_6: 0.0
• eval_F1_class_7: 0.5
• eval_F1_class_8: 0.0
• eval_F1_class_9: 0.0
• eval_F1_class_10: 0.0
• eval_F1_class_11: 0.9
• eval_F1_class_12: 0.6
• eval_F1_class_13: 0.9645
• eval_F1_class_14: 0.7500
• eval_F1_class_15: 0.8667
• eval_F1_class_16: 0.9302
• eval_F1_class_17: 1.0
• eval_F1_class_18: 0.8605
• eval_F1_class_19: 0.6667
• eval_F1_class_20: 0.9508
• eval_F1_class_21: 0.7692
• eval_F1_class_22: 0.0
• eval_F1_class_23: 0.6667
• eval_F1_class_24: 0.8333
• eval_F1_class_25: 0.5714
• eval_F1_class_26: 0.9231
• eval_F1_class_27: 0.9828
• eval_F1_class_28: 0.0
• eval_F1_class_29: 0.9500
• eval_F1_class_30: 0.0
• eval_F1_class_31: 0.0
• eval_F1_class_32: 0.5
• eval_F1_class_33: 0.5
• eval_F1_class_34: 0.8571
• eval_F1_class_35: 0.8547
• eval_F1_class_36: 0.8727
• eval_F1_class_37: 0.5
• eval_F1_class_38: 0.8837
• eval_F1_class_39: 0.7619
• eval_F1_class_40: 0.0
• eval_F1_class_41: 0.4444
• eval_F1_class_42: 0.0
• eval_F1_class_43: 0.0
• eval_F1_class_44: 0.0
• eval_F1_class_45: 0.7500
• eval_F1_class_46: 0.3333
• eval_F1_class_47: 0.7556
• eval_F1_class_48: 0.8679
• eval_F1_class_49: 1.0
• eval_F1_class_50: 0.8571
• eval_F1_class_51: 0.0
• eval_F1_class_52: 0.0
• eval_F1_class_53: 0.8
• eval_F1_class_54: 0.4
• eval_F1_class_55: 0.0
• eval_F1_class_56: 0.9082
• eval_F1_class_57: 0.0
• eval_F1_class_58: 0.6667
• eval_F1_class_59: 1.0
• eval_F1_class_60: 0.0
• eval_F1_class_61: 0.0
• eval_F1_class_62: 0.9747
• eval_F1_class_63: 0.0
• eval_F1_class_64: 0.8966
• eval_F1_class_65: 1.0
• eval_F1_class_66: 0.8
• eval_F1_class_67: 0.0
• eval_F1_class_68: 0.0
• eval_F1_class_69: 0.0
• eval_F1_class_70: 1.0
• eval_runtime: 61.7038
• eval_samples_per_second: 32.235
• eval_steps_per_second: 2.026
• epoch: 6.64
• step: 3300

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3