Demo Video 👇
video.mp4
Demo Image 👇
- This Repo Represents "Indian Food Classification"
- with the help of this project we can easily identity Indian Foods
-
I use Transfer learning (InceptionV3 Model) to create this model and got a 92% accuracy and I Perfrom some EDA to check each classes and Check number of training examples of each food.
-
This project will help people from other countries to classify Indian food easily.
-
I also Perform Data Augmentation for increasing the number of training examples , Data Augmentation can also reduce Overfitting.
-
I use callbacks can helps you prevent overfitting, visualize training progress, debug your code, save checkpoints, generate logs etc.
-
I use InceptionV3 Pre-Trained and got Training Accuracy 93% , you can also use other Pre-Trained Model like ResNet50 , Vgg19 etc
-
I use Flask to create a Web app and make an easy or Simple UI
- Download the Dataset here 👇
- https://www.kaggle.com/theeyeschico/indian-food-classification
- scikit-learn==0.24.1
- pandas==1.1.3
- numpy==1.18.5
- seaborn==0.11.1
- missingno==0.4.2
- imblearn==0.8.0
- tensorflow==2.5.0
- matlotlib==3.4.2 etc
To install the required packages and libraries, run this command in the project directory after cloning the repository:
pip install -r requirements.txt
What to do when you are at home due to this pandemic situation? I started to learn Machine Learning and Deep learning in last 8 months . I came to know mathematics behind all supervised models. Finally it is important to work on application (real world application) to actually make a difference.
If you find any bug or other problem regarding this project so just contact me on handles given below or also create an issue in issue section