metadata
license: mit
tags:
- generated_from_trainer
datasets:
- it5/datasets
metrics:
- rouge
model-index:
- name: it5-efficient-small-el32-qa-0.0003
results:
- task:
name: Summarization
type: summarization
dataset:
name: it5/datasets qa
type: it5/datasets
args: qa
metrics:
- name: Rouge1
type: rouge
value: 74.2234
it5-efficient-small-el32-qa-0.0003
This model is a fine-tuned version of stefan-it/it5-efficient-small-el32 on the it5/datasets qa dataset. It achieves the following results on the evaluation set:
- Loss: 0.8225
- Rouge1: 74.2234
- Rouge2: 40.5909
- Rougel: 74.1327
- Rougelsum: 74.2081
- Gen Len: 4.7055
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 1.1164 | 0.8 | 5000 | 0.8244 | 66.4678 | 35.3554 | 66.4543 | 66.4522 | 4.541 |
| 0.9097 | 1.59 | 10000 | 0.7299 | 70.0574 | 37.5535 | 69.9512 | 70.0084 | 4.5548 |
| 0.6637 | 2.39 | 15000 | 0.7314 | 72.0767 | 39.2263 | 72.0257 | 72.0473 | 4.703 |
| 0.5015 | 3.19 | 20000 | 0.7147 | 73.0185 | 39.9998 | 72.9347 | 72.9576 | 4.75 |
| 0.5101 | 3.99 | 25000 | 0.7055 | 73.7898 | 40.5481 | 73.7235 | 73.7901 | 4.8728 |
| 0.3903 | 4.78 | 30000 | 0.7442 | 74.0845 | 39.9841 | 74.0172 | 74.0635 | 4.5938 |
| 0.2993 | 5.58 | 35000 | 0.8184 | 73.8405 | 40.2569 | 73.7756 | 73.7972 | 4.7412 |
| 0.2227 | 6.38 | 40000 | 0.8278 | 74.0159 | 40.6403 | 73.9412 | 73.9722 | 4.742 |
Framework versions
- Transformers 4.15.0
- Pytorch 1.10.0+cu102
- Datasets 1.17.0
- Tokenizers 0.10.3