Welcome to the official repository of my internship journey with PalSoft for AI & Technology Solutions, where I successfully completed the training program titled:

"Introduction to Deep Learning with PyTorch: Theoretical Knowledge and Practical Implementation"

🧠 This training equipped me with both a strong theoretical foundation and practical experience in deep learning using PyTorch. Below is a summary of the tasks, notebooks, and projects I completed during this journey.

Each folder in this organization represents a phase or topic within the training program. Here’s a brief overview:

• Topics Covered: Perceptrons, activation functions, loss functions, and forward/backward propagation.
• Highlights: Manual implementation of a basic neural network from scratch using NumPy and PyTorch.
• 📌 Learning Outcome: Gained a solid understanding of how neural networks learn via gradient descent.

• Topics Covered: Tensors, data loading, autograd, and model building.
• Highlights: Built and trained simple models using the PyTorch workflow: model, loss, optimizer, training loop.
• 📌 Learning Outcome: Became comfortable with PyTorch's API and tensor operations.

• Topics Covered: CNN architectures, convolution layers, pooling, and regularization.
• Projects:
  • MNIST Digit Classifier
  • FashionMNIST Classifier
• 📌 Learning Outcome: Learned how to use convolutional layers for feature extraction in image tasks.

• Topics Covered: Pre-trained models, fine-tuning, and feature extraction.
• Project:
  • Dog vs. Cat Classifier using ResNet
• 📌 Learning Outcome: Leveraged pre-trained models to improve performance on custom datasets with minimal training.

• Topics Covered: Accuracy, confusion matrix, precision/recall, overfitting prevention.
• Highlights: Visualization with Matplotlib and evaluation using validation/test splits.
• 📌 Learning Outcome: Understood how to assess model performance effectively and avoid overfitting.

• Project: Real-world image classification or object detection task.
• Techniques Used: Data augmentation, advanced training strategies, checkpointing.
• 📌 Learning Outcome: Applied everything learned in a capstone-style project.

• Instructor: Saeb Swaiaty – Thank you for your clear explanations, support, and mentorship.
• Company: PalSoft – Grateful for this opportunity to grow in the field of deep learning.

Feel free to explore the repositories, leave feedback, or reach out if you'd like to collaborate.

🔗 My LinkedIn
📧 Email: baraalsedih@gmail.com

“The best way to learn is by doing – and deep learning is no exception.”