Waste Sorting Project - Environment Setup

This guide provides step-by-step instructions to set up the environment for training the waste classification model.

---

Download the Dataset

Google Drive link 🔗 Dataset
Extract the folder named "datasets" to the same level as modelT.ipynb
The Dataset is already preprossed You dont have to Run the preprocessing code again!!.

🚀 Step 1: Install Python 3.12.3

First, install Python 3.12.3 from the official website:
🔗 Download Python

After installation, verify the installation by running:

python --version

---

🛠 Step 2: Install Miniconda

Download and install Miniconda from the official website:
🔗 Download Miniconda

After installation, verify the installation by running:

conda --version

---

⚡ Step 3: Install CUDA & CuDNN (For GPU Acceleration)

Inside Conda Command Prompt, run:

conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0

This installs CUDA 11.2 and cuDNN 8.1.0, which are required for TensorFlow GPU support.

---

📂 Step 4: Create Conda Environment

Navigate to your project folder and create a new folder named envs inside it.

Then, create a Conda environment:

conda create --name waste_sorting python=3.9 -y

Activate the environment:

conda activate waste_sorting

---

🔧 Step 5: Install TensorFlow

Install TensorFlow 2.10 (latest version that supports GPU acceleration):

python -m pip install "tensorflow<2.11"

Verify if GPU is being used:

python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

---

📦 Step 6: Install Required Libraries

Install all necessary libraries:

pip install numpy==1.23.5
pip install matplotlib
pip install keras-efficientnet
pip install keras opencv-python pandas scikit-learn tqdm pillow

---

🔍 Step 7: Verify Installed Packages

To list all installed packages inside the Conda environment:

conda list

This should display all the installed dependencies.

---

✅ Environment Setup Complete!

You are now ready to proceed with the project. 🚀