---
license: apache-2.0
base_model: google/flan-t5-small
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: flan_t5_small_patent
  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_small_patent

This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9641
- Accuracy: 0.6728
- F1 Macro: 0.6106
- F1 Micro: 0.6728

## 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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
| 1.5134        | 0.06  | 50   | 1.3919          | 0.516    | 0.3552   | 0.516    |
| 1.2208        | 0.13  | 100  | 1.2316          | 0.5706   | 0.4262   | 0.5706   |
| 1.2041        | 0.19  | 150  | 1.1814          | 0.5968   | 0.4508   | 0.5968   |
| 1.07          | 0.26  | 200  | 1.1794          | 0.5916   | 0.4729   | 0.5916   |
| 1.0635        | 0.32  | 250  | 1.1234          | 0.6146   | 0.4798   | 0.6146   |
| 1.2169        | 0.38  | 300  | 1.0960          | 0.6178   | 0.5243   | 0.6178   |
| 1.1206        | 0.45  | 350  | 1.0888          | 0.626    | 0.5557   | 0.626    |
| 1.0177        | 0.51  | 400  | 1.1059          | 0.6218   | 0.5212   | 0.6218   |
| 1.1345        | 0.58  | 450  | 1.0609          | 0.6324   | 0.5442   | 0.6324   |
| 1.0781        | 0.64  | 500  | 1.0528          | 0.6262   | 0.5170   | 0.6262   |
| 0.998         | 0.7   | 550  | 1.0378          | 0.641    | 0.5555   | 0.641    |
| 1.0518        | 0.77  | 600  | 1.0547          | 0.6412   | 0.5692   | 0.6412   |
| 1.0501        | 0.83  | 650  | 1.0224          | 0.6482   | 0.5811   | 0.6482   |
| 1.0372        | 0.9   | 700  | 1.0320          | 0.635    | 0.5654   | 0.635    |
| 0.9985        | 0.96  | 750  | 1.0189          | 0.6534   | 0.5790   | 0.6534   |
| 0.8571        | 1.02  | 800  | 1.0161          | 0.648    | 0.5822   | 0.648    |
| 0.9036        | 1.09  | 850  | 1.0358          | 0.6466   | 0.5687   | 0.6466   |
| 0.9211        | 1.15  | 900  | 1.0129          | 0.6482   | 0.5764   | 0.6482   |
| 1.0038        | 1.21  | 950  | 1.0135          | 0.6502   | 0.5775   | 0.6502   |
| 0.9907        | 1.28  | 1000 | 1.0086          | 0.6556   | 0.5864   | 0.6556   |
| 0.7771        | 1.34  | 1050 | 1.0048          | 0.663    | 0.5913   | 0.663    |
| 0.8815        | 1.41  | 1100 | 1.0003          | 0.6584   | 0.5864   | 0.6584   |
| 0.9128        | 1.47  | 1150 | 0.9995          | 0.6534   | 0.5956   | 0.6534   |
| 0.8203        | 1.53  | 1200 | 0.9912          | 0.6738   | 0.6054   | 0.6738   |
| 0.9036        | 1.6   | 1250 | 0.9848          | 0.6538   | 0.5950   | 0.6538   |
| 0.8475        | 1.66  | 1300 | 0.9964          | 0.667    | 0.6080   | 0.667    |
| 0.9845        | 1.73  | 1350 | 0.9751          | 0.6696   | 0.5983   | 0.6696   |
| 0.8423        | 1.79  | 1400 | 0.9785          | 0.6698   | 0.6125   | 0.6698   |
| 0.8428        | 1.85  | 1450 | 0.9641          | 0.6728   | 0.6106   | 0.6728   |
| 1.0668        | 1.92  | 1500 | 0.9672          | 0.6646   | 0.5979   | 0.6646   |
| 0.8673        | 1.98  | 1550 | 0.9696          | 0.6642   | 0.6094   | 0.6642   |
| 0.6981        | 2.05  | 1600 | 0.9956          | 0.6714   | 0.6004   | 0.6714   |
| 0.6865        | 2.11  | 1650 | 0.9826          | 0.671    | 0.6184   | 0.671    |
| 0.8338        | 2.17  | 1700 | 0.9957          | 0.6744   | 0.6101   | 0.6744   |
| 0.7192        | 2.24  | 1750 | 0.9904          | 0.672    | 0.6148   | 0.672    |
| 0.7433        | 2.3   | 1800 | 1.0078          | 0.6698   | 0.6085   | 0.6698   |
| 0.7905        | 2.37  | 1850 | 1.0001          | 0.6692   | 0.6071   | 0.6692   |
| 0.7354        | 2.43  | 1900 | 1.0259          | 0.6642   | 0.6044   | 0.6642   |
| 0.6919        | 2.49  | 1950 | 1.0109          | 0.67     | 0.6083   | 0.67     |
| 0.7576        | 2.56  | 2000 | 1.0073          | 0.6716   | 0.6058   | 0.6716   |
| 0.7766        | 2.62  | 2050 | 1.0038          | 0.6704   | 0.6032   | 0.6704   |
| 0.6799        | 2.69  | 2100 | 1.0070          | 0.6684   | 0.6011   | 0.6684   |
| 0.7073        | 2.75  | 2150 | 1.0022          | 0.6674   | 0.6043   | 0.6674   |
| 0.7153        | 2.81  | 2200 | 0.9952          | 0.6732   | 0.6093   | 0.6732   |
| 0.8588        | 2.88  | 2250 | 0.9965          | 0.6712   | 0.6069   | 0.6712   |
| 0.7789        | 2.94  | 2300 | 0.9939          | 0.6708   | 0.6051   | 0.6708   |


### Framework versions

- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2