patentClassificationLongFormer3

This model is a fine-tuned version of eclec/patentClassificationLongFormer2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4056
• Accuracy: 0.8115
• F1: 0.5806

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: 1.3494108617032343e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 11
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.17770742346220472
• lr_scheduler_warmup_steps: 59
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.4232	1.0	1107	0.4202	0.8041	0.5957
0.3753	2.0	2215	0.4056	0.8115	0.5806
0.3157	3.0	3321	0.4318	0.8073	0.5949

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.0
• Datasets 2.14.4
• Tokenizers 0.13.3