GPT-2-agriparts / README.md
Gan1108's picture
Update README.md
3c15286 verified
metadata
license: mit
base_model: malteos/gpt2-uk
tags:
  - generated_from_trainer
model-index:
  - name: GPT-2 agriparts
    results: []
language:
  - uk
datasets:
  - Gan1108/agriparts
widget:
  - text: '0007347852 - 734785.0 - Підшипник кульковий закритий спеціальний '

agriparts-4

This model is a fine-tuned version of malteos/gpt2-uk on Gan1108/agriparts dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8610

Model description

The model is trained to generate a text description in Ukrainian of spare parts for agricultural machinery based on their name.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
1.1912 1.0 1340 0.9919
0.9461 2.0 2680 0.8901
0.8646 3.0 4020 0.8610

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1