| --- |
| tags: |
| - generated_from_trainer |
| metrics: |
| - bleu |
| model-index: |
| - name: mBART_translator_json_3 |
| 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. --> |
|
|
| # mBART_translator_json_3 |
| |
| This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on the None dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.2480 |
| - Bleu: 72.3119 |
| - Gen Len: 38.8266 |
| |
| ## 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 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
| |:-------------:|:-----:|:----:|:---------------:|:-------:|:--------:| |
| | No log | 1.0 | 444 | 1.5654 | 37.2408 | 115.2556 | |
| | 4.2672 | 2.0 | 888 | 0.9088 | 58.5669 | 56.1363 | |
| | 1.754 | 3.0 | 1332 | 0.6627 | 56.8038 | 56.2753 | |
| | 1.2023 | 4.0 | 1776 | 0.5349 | 59.8569 | 35.3384 | |
| | 0.9387 | 5.0 | 2220 | 0.4390 | 66.4894 | 46.4797 | |
| | 0.7839 | 6.0 | 2664 | 0.3663 | 68.8133 | 46.3215 | |
| | 0.664 | 7.0 | 3108 | 0.3127 | 67.7323 | 37.8041 | |
| | 0.5833 | 8.0 | 3552 | 0.2790 | 69.3004 | 38.8193 | |
| | 0.5833 | 9.0 | 3996 | 0.2543 | 70.0163 | 38.4707 | |
| | 0.5206 | 10.0 | 4440 | 0.2480 | 72.3119 | 38.8266 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.23.1 |
| - Pytorch 1.12.1+cu113 |
| - Datasets 2.5.2 |
| - Tokenizers 0.13.1 |
| |
