TurkishFoods-15 / README.md
yunusserhat's picture
Update README.md
7ed346c verified
metadata
dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': biber_dolmasi
            '1': borek
            '2': cig_kofte
            '3': enginar
            '4': hamsi
            '5': hunkar_begendi
            '6': icli_kofte
            '7': ispanak
            '8': kebap
            '9': kisir
            '10': kuru_fasulye
            '11': lokum
            '12': manti
            '13': simit
            '14': yaprak_sarma
  splits:
    - name: train
      num_bytes: 60093194.203592494
      num_examples: 5928
    - name: eval
      num_bytes: 7529941.352449062
      num_examples: 741
    - name: test
      num_bytes: 7577916.82795844
      num_examples: 742
  download_size: 75019089
  dataset_size: 75201052.38399999
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: eval
        path: data/eval-*
      - split: test
        path: data/test-*
license: apache-2.0
task_categories:
  - image-classification
language:
  - tr
size_categories:
  - 1K<n<10K

⚠️ Important: English version is available below.

TürkSofrası-15 (TurkishFoods-15) Veri Seti

TürkSofrası-15, 15 farklı geleneksel Türk yemeğine ait toplam 7.411 görsel içeren ve yemek tanıma/sınıflandırma amaçlı hazırlanmış bir görüntü veri setidir. Görseller .jpg formatında olup her sınıf için ayrı klasörlerde yer almaktadır.

Veri seti, Hugging Face datasets kütüphanesi biçimindedir ve image (görsel) ile label (etiket) olmak üzere iki özelliğe sahiptir. Etiketler class_label biçiminde tanımlanmış olup aşağıdaki gibi numaralanmıştır:

Etiket Sınıf Adı
0 Biber Dolması
1 Börek
2 Çiğ Köfte
3 Enginar
4 Hamsi
5 Hünkâr Beğendi
6 İçli Köfte
7 Ispanak
8 Kebap
9 Kısır
10 Kuru Fasulye
11 Lokum
12 Mantı
13 Simit
14 Yaprak Sarma

Veri Bölünmeleri

Bölüm Örnek Sayısı Boyut (Bayt)
Eğitim (train) 5,928 ~60 MB
Doğrulama (eval) 741 ~7.5 MB
Test (test) 742 ~7.6 MB
Toplam 7,411 ~75 MB

Klasör Yapısı

TurkishFoods-15/
  ├── biber_dolmasi/
  ├── borek/
  ├── cig_kofte/
  ├── enginar/
  ├── hamsi/
  ├── hunkar_begendi/
  ├── icli_kofte/
  ├── ispanak/
  ├── kebap/
  ├── kisir/
  ├── kuru_fasulye/
  ├── lokum/
  ├── manti/
  ├── simit/
  └── yaprak_sarma/

Her klasör ilgili yemeğe ait .jpg uzantılı görselleri içerir.

Atıf

Önemli: Bu veri seti bana ait değildir. Lütfen araştırmalarınızda bu veri setini kullanırsanız orijinal yazarları ve aşağıdaki yayını kaynak gösteriniz.

@inproceedings{gungor2017turkish,
  title={Turkish cuisine: A benchmark dataset with Turkish meals for food recognition},
  author={G{\"u}ng{\"o}r, Cem and Baltac{\i}, Fatih and Erdem, Aykut and Erdem, Erkut},
  booktitle={2017 25th Signal Processing and Communications Applications Conference (SIU)},
  pages={1--4},
  year={2017},
  organization={IEEE}
}

TurkishFoods-15 (TürkSofrası-15) Dataset

TurkishFoods-15 is a benchmark dataset consisting of 7,411 images of 15 traditional Turkish dishes, intended for food recognition and classification tasks. All images are in .jpg format and organized into folders per class.

The dataset follows the Hugging Face datasets format with two features: image and label. The label is defined as a class_label with the following mapping:

Label Class Name
0 Biber Dolması
1 Börek
2 Çiğ Köfte
3 Enginar
4 Hamsi
5 Hünkâr Beğendi
6 İçli Köfte
7 Ispanak
8 Kebap
9 Kısır
10 Kuru Fasulye
11 Lokum
12 Mantı
13 Simit
14 Yaprak Sarma

Dataset Splits

Split Number of Examples Size (Bytes)
Train 5,928 ~60 MB
Eval 741 ~7.5 MB
Test 742 ~7.6 MB
Total 7,411 ~75 MB

Directory Structure

TurkishFoods-15/
  ├── biber_dolmasi/
  ├── borek/
  ├── cig_kofte/
  ├── enginar/
  ├── hamsi/
  ├── hunkar_begendi/
  ├── icli_kofte/
  ├── ispanak/
  ├── kebap/
  ├── kisir/
  ├── kuru_fasulye/
  ├── lokum/
  ├── manti/
  ├── simit/
  └── yaprak_sarma/

Each folder contains .jpg images corresponding to the dish name.

Citation

If you use this dataset in your research, please cite the following publication:

@inproceedings{gungor2017turkish,
  title={Turkish cuisine: A benchmark dataset with Turkish meals for food recognition},
  author={G{\"u}ng{\"o}r, Cem and Baltac{\i}, Fatih and Erdem, Aykut and Erdem, Erkut},
  booktitle={2017 25th Signal Processing and Communications Applications Conference (SIU)},
  pages={1--4},
  year={2017},
  organization={IEEE}
}