metadata
license: apache-2.0
base_model: google/flan-t5-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: flan_t5_base_amazon
results: []
flan_t5_base_amazon
This model is a fine-tuned version of google/flan-t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5448
- Accuracy: 0.8412
- F1 Macro: 0.8142
- F1 Micro: 0.8412
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: 0.0005
- 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 |
---|---|---|---|---|---|---|
1.1669 | 0.13 | 50 | 0.9142 | 0.7404 | 0.6916 | 0.7404 |
0.8536 | 0.26 | 100 | 0.8417 | 0.7569 | 0.7197 | 0.7569 |
0.827 | 0.39 | 150 | 0.6893 | 0.7905 | 0.7471 | 0.7905 |
0.672 | 0.53 | 200 | 0.7235 | 0.7984 | 0.7730 | 0.7984 |
0.7424 | 0.66 | 250 | 0.6684 | 0.7945 | 0.7461 | 0.7945 |
0.6802 | 0.79 | 300 | 0.6008 | 0.8215 | 0.8014 | 0.8215 |
0.7847 | 0.92 | 350 | 0.6225 | 0.8123 | 0.7925 | 0.8123 |
0.5258 | 1.05 | 400 | 0.6656 | 0.8215 | 0.8000 | 0.8215 |
0.4945 | 1.18 | 450 | 0.6410 | 0.8235 | 0.7983 | 0.8235 |
0.4097 | 1.32 | 500 | 0.5937 | 0.8347 | 0.8110 | 0.8347 |
0.4116 | 1.45 | 550 | 0.5966 | 0.8314 | 0.8061 | 0.8314 |
0.4785 | 1.58 | 600 | 0.5696 | 0.8347 | 0.8107 | 0.8347 |
0.4821 | 1.71 | 650 | 0.5536 | 0.8366 | 0.8098 | 0.8366 |
0.4137 | 1.84 | 700 | 0.5612 | 0.8373 | 0.8116 | 0.8373 |
0.4623 | 1.97 | 750 | 0.5448 | 0.8412 | 0.8142 | 0.8412 |
0.1953 | 2.11 | 800 | 0.5984 | 0.8472 | 0.8201 | 0.8472 |
0.2114 | 2.24 | 850 | 0.6189 | 0.8432 | 0.8177 | 0.8432 |
0.2252 | 2.37 | 900 | 0.6411 | 0.8465 | 0.8199 | 0.8465 |
0.1937 | 2.5 | 950 | 0.6044 | 0.8524 | 0.8245 | 0.8524 |
0.2611 | 2.63 | 1000 | 0.6188 | 0.8472 | 0.8189 | 0.8472 |
0.3021 | 2.76 | 1050 | 0.6018 | 0.8472 | 0.8189 | 0.8472 |
0.2309 | 2.89 | 1100 | 0.5804 | 0.8478 | 0.8186 | 0.8478 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2