File size: 2,188 Bytes
d216ec3
ae72845
 
 
 
 
 
d216ec3
ae72845
ce31277
22b3b86
d216ec3
 
ae72845
d216ec3
ae72845
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
22b3b86
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
---
title: Fruit & Veg Classifier / Meyve ve Sebze Sınıflandırıcı
emoji: 🍎
colorFrom: green
colorTo: yellow
sdk: streamlit
app_file: app.py
pinned: false
short_description: CNN model to identify 36 fruit and vegetable species.
license: apache-2.0
sdk_version: 1.56.0
---

## 🍎 Fruit & Veg Classifier / Meyve ve Sebze Sınıflandırıcı

### 📝 Project Overview / Proje Özeti
**EN:** This project is a Deep Learning application that classifies **36 different fruits and vegetables**. It features a custom-built **5-layer Convolutional Neural Network (CNN)**. The project demonstrates the full pipeline from data preprocessing (EDA) to model deployment.

**TR:** Bu proje, **36 farklı meyve ve sebzeyi** sınıflandıran bir Derin Öğrenme uygulamasıdır. **5 katmanlı özel bir Evrişimli Sinir Ağı (CNN)** mimarisine sahiptir. Proje, veri ön işlemeden (EDA) modelin canlıya alınmasına kadar tüm süreçleri kapsar.

---

### 🚀 How to Use / Nasıl Kullanılır?
**EN:**
1. **Upload:** Drop a fruit or vegetable photo (JPG/PNG).
2. **Predict:** Click the **"Predict / Tahmin Et"** button.
3. **Result:** See the English and Turkish names of the species with the confidence level.

**TR:**
1. **Yükle:** Bir meyve veya sebze fotoğrafı yükleyin (JPG/PNG).
2. **Tahmin:** **"Predict / Tahmin Et"** butonuna tıklayın.
3. **Sonuç:** Türün İngilizce ve Türkçe isimlerini güven oranıyla birlikte görün.

---

### 📊 Model Architecture / Model Mimarisi
**EN:** * **CNN Layers:** 5 Convolutional layers for feature extraction.
* **Regularization:** Batch Normalization and Dropout (0.5) to prevent overfitting.
* **Input Size:** 128x128 pixels.
* **Optimizer:** Adam.

**TR:**
* **CNN Katmanları:** Öznitelik çıkarımı için 5 Evrişim katmanı.
* **Düzenlileştirme:** Aşırı öğrenmeyi önlemek için Batch Normalization ve Dropout (0.5).
* **Giriş Boyutu:** 128x128 piksel.
* **Optimizasyon:** Adam.

---

### 📂 Files / Dosyalar
* `app.py`: Streamlit interface code / Arayüz kodu.
* `cnn_model.h5`: Trained CNN model weights / Eğitilmiş model ağırlıkları.
* `requirements.txt`: Necessary libraries / Gerekli kütüphaneler.

---