Welcome to the Fruit Classification project! This repository contains code and instructions to implement a Convolutional Neural Network (CNN) for classifying images of fruits into ten different classes.
The dataset consists of images of various fruits, categorized into ten classes. You can access the dataset here.
In this project, you will:
-
Data Handling
- Download and preprocess the dataset.
- Visualize the dataset as per the provided template.
-
Neural Network Design
- Design a neural network with a maximum of 12 layers.
- Utilize CNN, dense, and pooling layers for network design.
-
Training the Neural Network
- Train the neural network using the training dataset for at least 100 epochs.
- Report the training and test set accuracy.
-
Prediction
- Predict the labels for all samples in the "predict" set.
- Display and report the predicted labels.
-
Code Submission
- Share the Colab link to your project along with the report.
- Avoid copy-pasting code snippets; it will result in zero marks.
- Do not copy code from the internet or other sources. You can use web resources for understanding purposes only.
- Your report should include observations, analysis, and results regarding the experiments performed in Task-3. It may also include the outputs of the code (properly captioned), but not the code snippets. Report on Tasks 4 and 5 is mandatory.
- In Task-3, besides normal training, additional experiments include:
- Training the model with a learning rate scheduler.
- Changing the optimizer.
- For these additional experiments, report only the train and test accuracy.
To get started with the project, follow these steps:
- Clone this repository to your local machine:
git clone https://github.com/your-username/fruit-classification-cnn.git
- Download the dataset from the provided link and place it in the appropriate directory.
- Follow the instructions in the provided code template to complete the tasks.
- Once you've completed the tasks, share the Colab link to your project along with the report.