metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - albertvillanova/legal_contracts
base_model: google/bert_uncased_L-4_H-512_A-8
model-index:
  - name: bert-small-finetuned-legal-contracts-larger4010
    results: []

bert-small-finetuned-legal-contracts-larger4010

This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the None dataset.

Model description

More information needed

Intended uses & limitations

The model was not trained on the whole dataset which is around 9.5 GB, but only The first 40% of train + the last 10% of train.

datasets_train = load_dataset('albertvillanova/legal_contracts' , split='train[:40%]')
datasets_validation = load_dataset('albertvillanova/legal_contracts' , split='train[-10%:]')

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 128
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1

Framework versions

Transformers 4.21.1
Pytorch 1.12.1+cu113
Datasets 2.4.0
Tokenizers 0.12.1