Edit model card

Visualize in Weights & Biases

my_awesome_food_model_v2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8053
  • Accuracy: 0.8083

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: 128
  • eval_batch_size: 128
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 512
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Accuracy
4.4448 0.9932 110 4.4236 0.0914
3.8312 1.9955 221 3.8007 0.4096
3.1568 2.9977 332 3.1221 0.5435
2.4967 4.0 443 2.4920 0.6308
2.0432 4.9932 553 2.0252 0.6825
1.6512 5.9955 664 1.6771 0.7184
1.388 6.9977 775 1.4464 0.7367
1.1677 8.0 886 1.2782 0.7533
1.0307 8.9932 996 1.1741 0.7625
0.9156 9.9955 1107 1.0900 0.7741
0.8283 10.9977 1218 1.0295 0.7771
0.8078 12.0 1329 0.9949 0.7776
0.7643 12.9932 1439 0.9656 0.7817
0.6578 13.9955 1550 0.9274 0.7868
0.611 14.9977 1661 0.9051 0.7921
0.6016 16.0 1772 0.9009 0.7912
0.5652 16.9932 1882 0.8772 0.7963
0.5492 17.9955 1993 0.8559 0.7992
0.5054 18.9977 2104 0.8734 0.7956
0.5351 20.0 2215 0.8617 0.7999
0.4949 20.9932 2325 0.8487 0.8013
0.4701 21.9955 2436 0.8437 0.8013
0.4576 22.9977 2547 0.8430 0.8008
0.4573 24.0 2658 0.8195 0.8071
0.4399 24.9932 2768 0.8206 0.8071
0.424 25.9955 2879 0.8212 0.8068
0.4031 26.9977 2990 0.8202 0.8069
0.4031 28.0 3101 0.8173 0.8080
0.407 28.9932 3211 0.8051 0.8069
0.4194 29.7968 3300 0.8053 0.8083

Framework versions

  • Transformers 4.42.3
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
85.9M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.