metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: flant5
results: []
flant5
This model is a fine-tuned version of google/flan-t5-base on an unknown dataset.
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: 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: 2
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2
Model Recycling
Evaluation on 36 datasets using talhaa/flant5 as a base model yields average score of 77.86 in comparison to 68.82 by google/t5-v1_1-base.
The model is ranked 1st among all tested models for the google/t5-v1_1-base architecture as of 10/01/2023 Results:
| 20_newsgroup | ag_news | amazon_reviews_multi | anli | boolq | cb | cola | copa | dbpedia | esnli | financial_phrasebank | imdb | isear | mnli | mrpc | multirc | poem_sentiment | qnli | qqp | rotten_tomatoes | rte | sst2 | sst_5bins | stsb | trec_coarse | trec_fine | tweet_ev_emoji | tweet_ev_emotion | tweet_ev_hate | tweet_ev_irony | tweet_ev_offensive | tweet_ev_sentiment | wic | wnli | wsc | yahoo_answers |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 87.0685 | 89.5333 | 67.14 | 52.1875 | 82.844 | 78.5714 | 80.1534 | 70 | 77.2667 | 90.6963 | 84.9 | 93.512 | 72.4902 | 87.4797 | 86.2745 | 61.8399 | 87.5 | 93.1173 | 90.7173 | 89.6811 | 85.9206 | 93.8073 | 56.5611 | 89.4438 | 97.4 | 91.6 | 47.054 | 80.5067 | 52.5926 | 74.8724 | 84.7674 | 71.76 | 68.8088 | 56.338 | 55.7692 | 72.6333 |
For more information, see: Model Recycling