metadata
license: apache-2.0
base_model: google-t5/t5-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: t5_base_twitter
results: []
t5_base_twitter
This model is a fine-tuned version of google-t5/t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4913
- Accuracy: 0.7656
- F1 Macro: 0.7266
- F1 Micro: 0.7656
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 |
---|---|---|---|---|---|---|
0.4808 | 0.18 | 50 | 0.5170 | 0.7445 | 0.6740 | 0.7445 |
0.5169 | 0.37 | 100 | 0.5100 | 0.7555 | 0.7269 | 0.7555 |
0.4548 | 0.55 | 150 | 0.4922 | 0.7647 | 0.7017 | 0.7647 |
0.498 | 0.74 | 200 | 0.5057 | 0.7518 | 0.6776 | 0.7518 |
0.4844 | 0.92 | 250 | 0.4913 | 0.7656 | 0.7266 | 0.7656 |
0.3949 | 1.1 | 300 | 0.5401 | 0.7482 | 0.6885 | 0.7482 |
0.4028 | 1.29 | 350 | 0.5463 | 0.7482 | 0.7209 | 0.7482 |
0.3778 | 1.47 | 400 | 0.5438 | 0.7555 | 0.7087 | 0.7555 |
0.4383 | 1.65 | 450 | 0.5412 | 0.7381 | 0.7095 | 0.7381 |
0.3984 | 1.84 | 500 | 0.5293 | 0.7555 | 0.7239 | 0.7555 |
0.3122 | 2.02 | 550 | 0.5272 | 0.7564 | 0.7212 | 0.7564 |
0.2764 | 2.21 | 600 | 0.5961 | 0.7463 | 0.7048 | 0.7463 |
0.236 | 2.39 | 650 | 0.6630 | 0.7454 | 0.6996 | 0.7454 |
0.1996 | 2.57 | 700 | 0.7070 | 0.7482 | 0.6967 | 0.7482 |
0.2245 | 2.76 | 750 | 0.6734 | 0.7454 | 0.7016 | 0.7454 |
0.2903 | 2.94 | 800 | 0.6760 | 0.7454 | 0.6954 | 0.7454 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2