metadata
base_model: luqh/ClinicalT5-base
tags:
- generated_from_trainer
datasets:
- sem_eval_2024_task_2
metrics:
- rouge
model-index:
- name: run1
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: sem_eval_2024_task_2
type: sem_eval_2024_task_2
config: sem_eval_2024_task_2_source
split: validation
args: sem_eval_2024_task_2_source
metrics:
- name: Rouge1
type: rouge
value: 50
run1
This model is a fine-tuned version of luqh/ClinicalT5-base on the sem_eval_2024_task_2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2336
- Rouge1: 50.0
- Rouge2: 0.0
- Rougel: 50.0
- Rougelsum: 50.0
- Gen Len: 2.22
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: 4
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- 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 | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
No log | 1.0 | 212 | 0.2358 | 50.5 | 0.0 | 50.5 | 50.5 | 3.93 |
1.8021 | 2.0 | 425 | 0.2346 | 50.5 | 0.0 | 50.5 | 50.5 | 3.73 |
1.8021 | 3.0 | 637 | 0.2347 | 50.0 | 0.0 | 50.0 | 50.0 | 2.04 |
0.2555 | 4.0 | 850 | 0.2342 | 51.0 | 0.0 | 51.0 | 51.0 | 3.46 |
0.2555 | 5.0 | 1062 | 0.2333 | 50.5 | 0.0 | 50.5 | 50.5 | 2.33 |
0.2518 | 6.0 | 1275 | 0.2327 | 51.0 | 0.0 | 51.0 | 50.5 | 2.52 |
0.2518 | 7.0 | 1487 | 0.2351 | 50.0 | 0.0 | 50.0 | 50.0 | 2.0 |
0.2516 | 8.0 | 1700 | 0.2354 | 50.0 | 0.0 | 50.0 | 50.0 | 2.0 |
0.2516 | 9.0 | 1912 | 0.2329 | 52.0 | 0.0 | 52.0 | 51.5 | 2.42 |
0.2516 | 9.98 | 2120 | 0.2336 | 50.0 | 0.0 | 50.0 | 50.0 | 2.22 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0