---
library_name: transformers
tags:
- generated_from_trainer
model-index:
- name: ScratchCNN-FacesMTL-EXP1
  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. -->

# ScratchCNN-FacesMTL-EXP1

This model has been pretrained on [[thethinkmachine/faces-mtl](huggingface.co/thethinkmachine/faces-mtl)](https://huggingface.co/datasets/thethinkmachine/faces-mtl) dataset.
It achieves the following results on the evaluation set:
- Gender Accuracy: 0.7496
- Gender F1: 0.4285
- Age Mae: 11.5140
- Age Rmse: 14.6737
- Loss: 215.8605

## 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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Gender Accuracy | Gender F1 | Age Mae | Age Rmse | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|:--------:|:---------------:|
| 461.5582      | 0.1728 | 150  | 0.7502          | 0.4286    | 16.4067 | 22.4096  | 502.7608        |
| 247.5034      | 0.3456 | 300  | 0.7502          | 0.4286    | 12.8211 | 17.7391  | 315.2488        |
| 277.9706      | 0.5184 | 450  | 0.7502          | 0.4286    | 12.0017 | 15.9622  | 255.3581        |
| 262.517       | 0.6912 | 600  | 0.7502          | 0.4286    | 12.5625 | 15.3095  | 234.9443        |
| 272.0832      | 0.8641 | 750  | 0.7502          | 0.4286    | 12.4385 | 15.1538  | 230.1988        |
| 276.318       | 1.0369 | 900  | 0.7502          | 0.4286    | 12.2044 | 15.1683  | 230.6356        |
| 225.0257      | 1.2097 | 1050 | 0.7502          | 0.4286    | 12.6894 | 15.6222  | 244.6085        |
| 222.1815      | 1.3825 | 1200 | 0.7502          | 0.4286    | 11.6216 | 14.8317  | 220.5305        |
| 284.4039      | 1.5553 | 1350 | 0.7502          | 0.4286    | 12.0362 | 14.8581  | 221.3155        |
| 273.2046      | 1.7281 | 1500 | 0.7502          | 0.4286    | 12.0341 | 14.7725  | 218.7777        |
| 212.378       | 1.9009 | 1650 | 0.7502          | 0.4286    | 11.6115 | 15.3262  | 235.4443        |
| 229.5636      | 2.0737 | 1800 | 0.7502          | 0.4286    | 11.5028 | 14.6914  | 216.3843        |
| 247.4141      | 2.2465 | 1950 | 0.7502          | 0.4286    | 11.5840 | 15.4565  | 239.4500        |
| 219.0596      | 2.4194 | 2100 | 0.7502          | 0.4286    | 11.6788 | 14.6211  | 214.3237        |
| 218.1279      | 2.5922 | 2250 | 0.7502          | 0.4286    | 11.3897 | 14.7950  | 219.4363        |
| 215.6589      | 2.7650 | 2400 | 0.7502          | 0.4286    | 11.4476 | 14.6256  | 214.4511        |
| 278.3962      | 2.9378 | 2550 | 0.7502          | 0.4286    | 12.0106 | 14.7823  | 219.0586        |
| 237.293       | 3.1106 | 2700 | 0.7502          | 0.4286    | 11.3412 | 14.4551  | 209.4899        |
| 199.9965      | 3.2834 | 2850 | 0.7502          | 0.4286    | 11.4439 | 14.3701  | 207.0407        |
| 252.8384      | 3.4562 | 3000 | 0.7505          | 0.4299    | 11.3330 | 14.3047  | 205.1660        |
| 215.6976      | 3.6290 | 3150 | 0.7502          | 0.4286    | 11.2316 | 14.3694  | 207.0205        |
| 231.9552      | 3.8018 | 3300 | 0.7502          | 0.4286    | 11.3182 | 14.2583  | 203.8388        |
| 265.6475      | 3.9747 | 3450 | 0.7502          | 0.4286    | 11.2547 | 14.5174  | 211.2962        |
| 217.5101      | 4.1475 | 3600 | 0.7502          | 0.4286    | 11.1854 | 14.2591  | 203.8592        |
| 211.7694      | 4.3203 | 3750 | 0.7502          | 0.4286    | 11.2458 | 14.2679  | 204.1106        |
| 256.7229      | 4.4931 | 3900 | 0.7502          | 0.4286    | 11.1988 | 14.2688  | 204.1362        |
| 184.4238      | 4.6659 | 4050 | 0.7502          | 0.4286    | 11.1603 | 14.2800  | 204.4571        |
| 211.074       | 4.8387 | 4200 | 0.7502          | 0.4286    | 11.1595 | 14.2715  | 204.2160        |


### Framework versions

- Transformers 4.57.1
- Pytorch 2.9.0+cu130
- Datasets 4.4.1
- Tokenizers 0.22.1