metadata
license: mit
base_model: gpt2-medium
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: gpt2-medium-supervised-summarize-checkpoint
results: []
gpt2-medium-supervised-summarize-checkpoint
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7422
- Rouge1: 0.6035
- Rouge2: 0.2047
- Rougel: 0.4141
- Rougelsum: 0.5319
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: 1e-05
- train_batch_size: 50
- eval_batch_size: 50
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 1.859 | 0.21 | 500 | 1.8105 | 0.5966 | 0.1961 | 0.4025 | 0.5237 |
| 1.852 | 0.43 | 1000 | 1.7900 | 0.5994 | 0.1981 | 0.4061 | 0.5271 |
| 1.8189 | 0.64 | 1500 | 1.7764 | 0.6000 | 0.2005 | 0.4082 | 0.5276 |
| 1.8191 | 0.86 | 2000 | 1.7695 | 0.6013 | 0.2009 | 0.4096 | 0.5290 |
| 1.7969 | 1.07 | 2500 | 1.7617 | 0.6038 | 0.2020 | 0.4108 | 0.5311 |
| 1.7967 | 1.28 | 3000 | 1.7578 | 0.6024 | 0.2024 | 0.4114 | 0.5304 |
| 1.7813 | 1.5 | 3500 | 1.7520 | 0.6038 | 0.2039 | 0.4128 | 0.5320 |
| 1.7704 | 1.71 | 4000 | 1.7480 | 0.6033 | 0.2045 | 0.4132 | 0.5310 |
| 1.7852 | 1.93 | 4500 | 1.7422 | 0.6035 | 0.2047 | 0.4141 | 0.5319 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0