This repository contains practical Python labs focused on machine learning, deep learning, NLP, computer vision, and AI API integrations.
Built a machine learning model for car price prediction using pandas, scikit-learn, Pipeline, ColumnTransformer, preprocessing, model evaluation, and cross-validation.
Trained CNN models for handwritten digit recognition using MNIST dataset. Compared base and improved models and saved trained models.
Compared Linear and Convolutional neural networks on different MNIST training set sizes. Measured accuracy, training time, and confusion matrices.
Used OpenAI API to analyze customer reviews, detect sentiment, extract key topics, and find improvement suggestions.
Used Gemini API to test sentiment analysis on difficult texts with sarcasm, irony, slang, and mixed context.
Explored Hugging Face models for NLP and computer vision tasks.
- Python
- Pandas
- Matplotlib
- Scikit-learn
- TensorFlow / Keras
- PyTorch
- OpenAI API
- Gemini API
- Hugging Face Transformers
- Google Colab