|
--- |
|
license: apache-2.0 |
|
base_model: t5-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- glue |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: t5-base_qnli_dense_epochs-8 |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: glue |
|
type: glue |
|
config: qnli |
|
split: validation |
|
args: qnli |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.9269632070291048 |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# t5-base_qnli_dense_epochs-8 |
|
|
|
This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the glue dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1982 |
|
- Accuracy: 0.9270 |
|
|
|
## 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: 64 |
|
- seed: 0 |
|
- distributed_type: multi-GPU |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 20 |
|
- num_epochs: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 0.6742 | 0.01 | 50 | 0.6559 | 0.5380 | |
|
| 0.4309 | 0.02 | 100 | 0.4215 | 0.8433 | |
|
| 0.4535 | 0.02 | 150 | 0.3441 | 0.8644 | |
|
| 0.2523 | 0.03 | 200 | 0.2892 | 0.8957 | |
|
| 0.2616 | 0.04 | 250 | 0.2927 | 0.8986 | |
|
| 0.2088 | 0.05 | 300 | 0.3608 | 0.8796 | |
|
| 0.2454 | 0.05 | 350 | 0.2730 | 0.9087 | |
|
| 0.2824 | 0.06 | 400 | 0.2819 | 0.8900 | |
|
| 0.308 | 0.07 | 450 | 0.2904 | 0.8966 | |
|
| 0.2035 | 0.08 | 500 | 0.3073 | 0.8951 | |
|
| 0.2096 | 0.08 | 550 | 0.2743 | 0.9061 | |
|
| 0.338 | 0.09 | 600 | 0.2520 | 0.9072 | |
|
| 0.2484 | 0.1 | 650 | 0.2702 | 0.9030 | |
|
| 0.2042 | 0.11 | 700 | 0.2476 | 0.9138 | |
|
| 0.2908 | 0.11 | 750 | 0.2194 | 0.9180 | |
|
| 0.1985 | 0.12 | 800 | 0.2432 | 0.9169 | |
|
| 0.19 | 0.13 | 850 | 0.2615 | 0.9112 | |
|
| 0.2186 | 0.14 | 900 | 0.2289 | 0.9215 | |
|
| 0.2208 | 0.15 | 950 | 0.2272 | 0.9204 | |
|
| 0.3049 | 0.15 | 1000 | 0.3508 | 0.8880 | |
|
| 0.3373 | 0.16 | 1050 | 0.2363 | 0.9105 | |
|
| 0.2493 | 0.17 | 1100 | 0.2196 | 0.9206 | |
|
| 0.2359 | 0.18 | 1150 | 0.2160 | 0.9237 | |
|
| 0.2207 | 0.18 | 1200 | 0.2211 | 0.9217 | |
|
| 0.2824 | 0.19 | 1250 | 0.2386 | 0.9182 | |
|
| 0.3605 | 0.2 | 1300 | 0.2548 | 0.9112 | |
|
| 0.2763 | 0.21 | 1350 | 0.2579 | 0.9149 | |
|
| 0.2299 | 0.21 | 1400 | 0.2104 | 0.9226 | |
|
| 0.1787 | 0.22 | 1450 | 0.2280 | 0.9224 | |
|
| 0.1961 | 0.23 | 1500 | 0.2244 | 0.9233 | |
|
| 0.1923 | 0.24 | 1550 | 0.2245 | 0.9231 | |
|
| 0.1844 | 0.24 | 1600 | 0.2735 | 0.9123 | |
|
| 0.1714 | 0.25 | 1650 | 0.3108 | 0.9121 | |
|
| 0.2606 | 0.26 | 1700 | 0.2238 | 0.9189 | |
|
| 0.3326 | 0.27 | 1750 | 0.2363 | 0.9132 | |
|
| 0.1379 | 0.27 | 1800 | 0.2429 | 0.9094 | |
|
| 0.2266 | 0.28 | 1850 | 0.2416 | 0.9224 | |
|
| 0.2654 | 0.29 | 1900 | 0.2277 | 0.9242 | |
|
| 0.6668 | 0.3 | 1950 | 0.2808 | 0.9092 | |
|
| 0.1875 | 0.31 | 2000 | 0.1982 | 0.9270 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.14.1 |
|
|