parham.saremi
/
region-detection-using-food-features
Watch 1
Star 0
Fork 0
Code Issues 0 Pull Requests 0 Releases 0 Wiki Activity
BSc project of Parham Saremi. The goal of the project was to detect the geographical region of the food using textual and visual features extracted from recipes and ingredients of the food.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
2 Commits
1 Branch
Branch: master
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'master'
${ noResults }
region-detection-using-food.../best_config.yml

best_config.yml 842B
Raw Permalink Blame History

12345678910111213141516171819202122232425262728293031 
  1. optim:

  2.     epochs: 50

  3.     batch_size: 256

  4.     max_lr: 0.001

  5.     weight_decay: 0.0005

  6.     device: "cuda:0"

  7.     num_workers: 4

  8.     sam_rho: 0.1

  9. text_model: "bert-base-uncased"

  10. image_model: "efficientnet_b0"

  11. image_features_path: "/home/dml/food/CuisineAdaptation/IngredientsEncoding/image-features-full"

  12. text_features_path: "/home/dml/food/CuisineAdaptation/IngredientsEncoding/text-features"

  13. use_recipe_text: True

  14. use_image_ingredients: True

  15. use_text_ingredients: True

  16. model:

  17.     ingredient_feature_extractor:

  18.         layers: "TTTTT"

  19.         H: 384

  20.         transformer:

  21.             L: 4

  22.             n_heads: 4

  23.         final_ingredient_feature_size: 200

  24.     image_feature_size: 200

  25.     text_feature_size: 200

  26.     final_classes: -1

  27. data:

  28.     embedding_size: 100

  29.     dataset_path: "Data"

  30.     fasttext_path: "trained_1m.model"

  31.     target: "region"