---
library_name: transformers
license: apache-2.0
base_model: google/flan-t5-base
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: flan-t5-rouge-durga-3
  results: []
---

<!-- 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. -->

# flan-t5-rouge-durga-3

This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0024
- Rouge1: 0.5907
- Rouge2: 0.5623
- Rougel: 0.5912
- Rougelsum: 0.5922

## 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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
| 2.4258        | 1.0   | 50   | 2.0073          | 0.2733 | 0.1044 | 0.2514 | 0.2517    |
| 2.2156        | 2.0   | 100  | 1.5972          | 0.2600 | 0.0904 | 0.2466 | 0.2473    |
| 1.9538        | 3.0   | 150  | 1.2838          | 0.2999 | 0.1401 | 0.2867 | 0.2877    |
| 1.2774        | 4.0   | 200  | 1.0377          | 0.3370 | 0.1809 | 0.3253 | 0.3273    |
| 1.2395        | 5.0   | 250  | 0.8060          | 0.3688 | 0.2162 | 0.3561 | 0.3589    |
| 1.6069        | 6.0   | 300  | 0.6646          | 0.3898 | 0.2558 | 0.3752 | 0.3774    |
| 1.2469        | 7.0   | 350  | 0.5189          | 0.4014 | 0.2612 | 0.3875 | 0.3896    |
| 0.664         | 8.0   | 400  | 0.4060          | 0.4506 | 0.3294 | 0.4380 | 0.4402    |
| 1.8862        | 9.0   | 450  | 0.3131          | 0.4657 | 0.3734 | 0.4600 | 0.4623    |
| 0.3403        | 10.0  | 500  | 0.2534          | 0.4662 | 0.3662 | 0.4602 | 0.4622    |
| 0.5261        | 11.0  | 550  | 0.1996          | 0.4984 | 0.4243 | 0.4951 | 0.4967    |
| 1.0378        | 12.0  | 600  | 0.1467          | 0.5047 | 0.4431 | 0.5035 | 0.5055    |
| 0.218         | 13.0  | 650  | 0.1176          | 0.4912 | 0.4286 | 0.4883 | 0.4924    |
| 0.5121        | 14.0  | 700  | 0.0843          | 0.5418 | 0.4951 | 0.5416 | 0.5427    |
| 0.4996        | 15.0  | 750  | 0.0679          | 0.5645 | 0.5231 | 0.5612 | 0.5645    |
| 0.2511        | 16.0  | 800  | 0.0521          | 0.5426 | 0.4922 | 0.5399 | 0.5434    |
| 0.3803        | 17.0  | 850  | 0.0395          | 0.5607 | 0.5232 | 0.5604 | 0.5616    |
| 0.0706        | 18.0  | 900  | 0.0261          | 0.5816 | 0.5460 | 0.5824 | 0.5838    |
| 0.1628        | 19.0  | 950  | 0.0193          | 0.5833 | 0.5541 | 0.5843 | 0.5848    |
| 0.1777        | 20.0  | 1000 | 0.0141          | 0.5833 | 0.5524 | 0.5831 | 0.5853    |
| 0.1254        | 21.0  | 1050 | 0.0122          | 0.5849 | 0.5554 | 0.5842 | 0.5866    |
| 0.2481        | 22.0  | 1100 | 0.0109          | 0.5916 | 0.5616 | 0.5915 | 0.5932    |
| 0.0604        | 23.0  | 1150 | 0.0066          | 0.5912 | 0.5623 | 0.5909 | 0.5926    |
| 0.1083        | 24.0  | 1200 | 0.0048          | 0.5902 | 0.5611 | 0.5904 | 0.5916    |
| 0.2921        | 25.0  | 1250 | 0.0051          | 0.5890 | 0.5602 | 0.5891 | 0.5904    |
| 0.1846        | 26.0  | 1300 | 0.0037          | 0.5902 | 0.5618 | 0.5907 | 0.5920    |
| 0.1952        | 27.0  | 1350 | 0.0029          | 0.5907 | 0.5623 | 0.5912 | 0.5922    |
| 0.0419        | 28.0  | 1400 | 0.0026          | 0.5897 | 0.5611 | 0.5898 | 0.5914    |
| 0.1353        | 29.0  | 1450 | 0.0025          | 0.5906 | 0.5618 | 0.5902 | 0.5916    |
| 0.0072        | 30.0  | 1500 | 0.0024          | 0.5907 | 0.5623 | 0.5912 | 0.5922    |


### Framework versions

- Transformers 4.45.2
- Pytorch 2.5.0+cu121
- Datasets 3.0.2
- Tokenizers 0.20.1