metadata
license: other
library_name: peft
tags:
- generated_from_trainer
base_model: Qwen/Qwen1.5-7B
metrics:
- accuracy
model-index:
- name: twitter_disaster
results: []
twitter_disaster
This model is a fine-tuned version of Qwen/Qwen1.5-7B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4902
- Accuracy: 0.7767
- F1 Macro: 0.7451
- F1 Micro: 0.7767
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
---|---|---|---|---|---|---|
0.8422 | 0.18 | 50 | 0.6453 | 0.7178 | 0.6372 | 0.7178 |
0.6082 | 0.37 | 100 | 0.5489 | 0.7472 | 0.7123 | 0.7472 |
0.4305 | 0.55 | 150 | 0.5572 | 0.7252 | 0.5777 | 0.7252 |
0.5021 | 0.74 | 200 | 0.5000 | 0.7721 | 0.7437 | 0.7721 |
0.4715 | 0.92 | 250 | 0.4902 | 0.7767 | 0.7451 | 0.7767 |
0.3937 | 1.1 | 300 | 0.5194 | 0.7601 | 0.7018 | 0.7601 |
0.4219 | 1.29 | 350 | 0.5228 | 0.7665 | 0.7228 | 0.7665 |
0.4315 | 1.47 | 400 | 0.5791 | 0.7555 | 0.6901 | 0.7555 |
0.4134 | 1.65 | 450 | 0.6182 | 0.7390 | 0.7196 | 0.7390 |
0.4173 | 1.84 | 500 | 0.5454 | 0.7638 | 0.7116 | 0.7638 |
0.3278 | 2.02 | 550 | 0.5477 | 0.7721 | 0.7219 | 0.7721 |
0.2641 | 2.21 | 600 | 0.6011 | 0.7528 | 0.7152 | 0.7528 |
0.2256 | 2.39 | 650 | 0.6485 | 0.7601 | 0.6962 | 0.7601 |
0.2544 | 2.57 | 700 | 0.6459 | 0.7629 | 0.7165 | 0.7629 |
0.2839 | 2.76 | 750 | 0.5922 | 0.7656 | 0.7253 | 0.7656 |
0.2634 | 2.94 | 800 | 0.6312 | 0.7638 | 0.7076 | 0.7638 |
Framework versions
- PEFT 0.9.0
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2