metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: TED_CLM_gpt2_tedlium6
results: []
TED_CLM_gpt2_tedlium6
This model is a fine-tuned version of Lakoc/TED_CLM_gpt2_tedlium4 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8904
- Accuracy: 0.5555
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: 0.005
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 512
- total_eval_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7738 | 1.0 | 180 | 1.8841 | 0.5533 |
| 1.7249 | 2.0 | 360 | 1.8786 | 0.5550 |
| 1.6667 | 3.0 | 540 | 1.8845 | 0.5548 |
| 1.6369 | 4.0 | 720 | 1.8904 | 0.5555 |
| 1.6309 | 5.0 | 900 | 1.8904 | 0.5555 |
| 1.6288 | 6.0 | 1080 | 1.8904 | 0.5555 |
| 1.6277 | 7.0 | 1260 | 1.8904 | 0.5555 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.13.1
- Tokenizers 0.13.3