This README outlines the different types of machine learning tasks demonstrated by the example projects in this repository:
Classification Tasks:
- Sentiment Analysis: Predict whether a text (like a product review) expresses a positive or negative sentiment. This demonstrates a binary classification task. 👍👎
- GitHub Issue Classification: Categorize GitHub issues into different labels based on their description. This demonstrates a multiclass classification task. 🏷️➡️
Regression Tasks:
- Price Predictor: Predict a numerical value, such as the price of a product, based on its features. This demonstrates a regression task. 📈
Clustering Tasks:
- Iris Clustering: Group data points (in this case, Iris flower measurements) into clusters based on their similarities, without predefined labels. This demonstrates a clustering task. 🌸➡️
Recommendation Systems:
- Recommendation: Generate personalized recommendations for movies based on users' past ratings. 🎬⭐
Deep Learning Integration:
- Image Classification: Customize an existing image recognition model (TensorFlow) to classify images into new, specific categories. This demonstrates transfer learning. 🖼️➡️
- Detect Objects in Images: Identify and locate specific objects within images using a pre-trained ONNX model. 🎯🔍
Working with Pre-trained Models:
- Classify Sentiment of Movie Reviews (TensorFlow): Utilize a pre-trained TensorFlow model to determine the sentiment (positive or negative) expressed in movie reviews. 🎥➡️
Anomaly Detection:
- Anomaly Detection: Identify unusual or unexpected data points in product sales data. This demonstrates an anomaly detection task.
⚠️ 📊