This repository brings together all practical deep learning tasks. Each task explores foundational to intermediate concepts in deep learning, using libraries such as NumPy, TensorFlow, and PyTorch.
assignment-1/: Tasks 1–10 implemented using TensorFlow, NumPy, and Scikit-learn.assignment-2/: Extended or re-implemented tasks using additional frameworks like PyTorch and more advanced techniques.requirements.txt: Contains all required packages for both assignments.data/: Add datasets and images used across tasks.
| Task | Concept | Assignment Branch |
|---|---|---|
| 01 | Linear Regression Implementation | Assignment 1 & 2 |
| 02 | Feedforward Neural Network on MNIST | Assignment 1 & 2 |
| 03 | Bias-Variance with Polynomial Regression | Assignment 1 & 2 |
| 04 | PCA on Iris Dataset | Assignment 1 & 2 |
| 05 | MLP: NumPy vs Framework (CIFAR-10) | Assignment 1 & 2 |
| 06 | Backpropagation from Scratch | Assignment 1 & 2 |
| 07 | Function Approximation (Sine) | Assignment 1 & 2 |
| 08 | Raw vs Engineered Features (Titanic) | Assignment 1 & 2 |
| 09 | L2 Regularization in NN | Assignment 1 & 2 |
| 10 | Hyperparameter Tuning (Keras Tuner / Others) | Assignment 1 & 2 |
| Dataset | Description |
|---|---|
| Boston Housing | House price prediction |
| MNIST & Fashion MNIST | Digit & fashion item classification |
| Synthetic Regression | Bias-variance exploration |
| Iris Dataset | Flower species classification |
| CIFAR-10 | Image classification |
| Sine Function Data | Function approximation |
| Titanic Dataset | Survival prediction |
For more info on datasets, see
assignment-1/README.mdordata/README.md.
numpy
pandas
matplotlib
scikit-learn
tensorflow
keras-tuner
seaborn
torch- Clone the repository:
git clone https://github.com/yourusername/deep-learning-tasks-2024.git
cd deep-learning-tasks-2024- Install dependencies:
pip install -r requirements.txt- Navigate into either assignment folder and run tasks:
cd assignment-1
python task01_linear_regression.py- Each assignment is maintained in a separate Git branch:
assignment-1,assignment-2. - Refer to respective README files in each branch for detailed task descriptions.
Happy Learning! 🚀