|
--- |
|
language: |
|
- en |
|
license: apache-2.0 |
|
datasets: |
|
- glue |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: t5-base-finetuned-rte |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: GLUE RTE |
|
type: glue |
|
args: rte |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.5634 |
|
--- |
|
|
|
|
|
# T5-base-finetuned-rte |
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
This model is T5 fine-tuned on GLUE RTE dataset. It acheives the following results on the validation set |
|
- Accuracy: 0.7690 |
|
|
|
|
|
## Model Details |
|
T5 is an encoder-decoder model pre-trained on a multi-task mixture of unsupervised and supervised tasks and for which each task is converted into a text-to-text format. |
|
|
|
## Training procedure |
|
|
|
### Tokenization |
|
Since, T5 is a text-to-text model, the labels of the dataset are converted as follows: |
|
For each example, a sentence as been formed as **"rte sentence1: " + rte_sent1 + "sentence 2: " + rte_sent2** and fed to the tokenizer to get the **input_ids** and **attention_mask**. |
|
For each label, target is choosen as **"entailment"** if label is 0, else label is **"not_entailment"** and tokenized to get **input_ids** and **attention_mask** . |
|
During training, these inputs_ids having **pad** token are replaced with -100 so that loss is not calculated for them. Then these input ids are given as labels, and above attention_mask of labels |
|
is given as decoder attention mask. |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 3e-4 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: epsilon=1e-08 |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
|
|
|Epoch | Training Loss | Validation Accuracy | |
|
|:----:|:-------------:|:-------------------:| |
|
| 1 | 0.1099 | 0.7617 | |
|
| 2 | 0.0573 | 0.7617 | |
|
| 3 | 0.0276 | 0.7690 | |
|
|
