Spaces:
Runtime error
Runtime error
| tags: | |
| - generated_from_trainer | |
| metrics: | |
| - accuracy | |
| model-index: | |
| - name: essays | |
| results: [] | |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You | |
| should probably proofread and complete it, then remove this comment. --> | |
| # essays | |
| This model is a fine-tuned version of [sberbank-ai/rugpt3small_based_on_gpt2](https://huggingface.co/sberbank-ai/rugpt3small_based_on_gpt2) on an unknown dataset. | |
| It achieves the following results on the evaluation set: | |
| - Loss: 2.7676 | |
| - Accuracy: 0.4758 | |
| ## 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: 5e-05 | |
| - train_batch_size: 1 | |
| - eval_batch_size: 1 | |
| - seed: 42 | |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 | |
| - lr_scheduler_type: linear | |
| - num_epochs: 3.0 | |
| ### Training results | |
| ### Framework versions | |
| - Transformers 4.29.0.dev0 | |
| - Pytorch 1.9.1+cu111 | |
| - Datasets 2.11.0 | |
| - Tokenizers 0.13.3 | |