File size: 4,470 Bytes
4d80e96 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 |
---
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
|