metadata
library_name: transformers
license: mit
base_model: facebook/mbart-large-50
tags:
- generated_from_trainer
model-index:
- name: image_captioning_model
results: []
image_captioning_model
This model is a fine-tuned version of facebook/mbart-large-50 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.9561
- Bleu4: 4.6602
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 350
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu4 |
|---|---|---|---|---|
| 12.4864 | 1.0 | 113 | 12.2093 | 0.9329 |
| 5.5024 | 2.0 | 226 | 5.1222 | 1.4283 |
| 4.8103 | 3.0 | 339 | 4.5379 | 2.3729 |
| 4.4941 | 4.0 | 452 | 4.2987 | 3.4861 |
| 4.2601 | 5.0 | 565 | 4.1685 | 4.2715 |
| 4.0996 | 6.0 | 678 | 4.1062 | 4.5653 |
| 4.0649 | 7.0 | 791 | 4.0509 | 4.4991 |
| 3.9869 | 8.0 | 904 | 4.0288 | 4.2192 |
| 3.8730 | 9.0 | 1017 | 3.9991 | 4.2306 |
| 3.7958 | 10.0 | 1130 | 3.9973 | 4.1558 |
| 3.7842 | 11.0 | 1243 | 3.9754 | 5.1463 |
| 3.7156 | 12.0 | 1356 | 3.9678 | 4.0594 |
| 3.6771 | 13.0 | 1469 | 3.9686 | 4.5274 |
| 3.6874 | 14.0 | 1582 | 3.9671 | 4.4852 |
| 3.6374 | 15.0 | 1695 | 3.9548 | 4.309 |
| 3.5811 | 16.0 | 1808 | 3.9561 | 4.6602 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2