Task 1 - Decision Tree based ID3 Algorithm
Understanding Basic Terminology
Understand ID3
Implement ID3 for ID3
Decision Trees and Terminology
A decision tree is a powerful tool used in machine learning for classification and regression tasks. It mimics the human decision-making process by breaking down data into smaller and smaller subsets while at the same time an associated decision tree is incrementally developed.
ID3 Algorithm
ID3 (Iterative Dichotomiser 3) is a simple decision tree algorithm introduced by Ross Quinlan in 1986. It uses entropy and information gain to construct a decision tree by recursively partitioning the dataset.
Here is my Task ID3
Task 2 - Naive Bayesian Classifier
Understand Naive Bayesian Classifier, watch it in action using sklearn
Implement Naive Bayesian Classifier for text classification and other applicable datasets
The Naive Bayesian Classifier is a probabilistic machine learning model used for classification tasks. It is based on Bayes' Theorem and assumes that the features are conditionally independent given the class. Despite this "naive" assumption, the classifier often performs surprisingly well in practice, especially for text classification tasks such as spam detection and sentiment analysis.
Here is my Task Naive Bayesian Classifier
Task 3 - Ensemble techniques
What are ensemble techniques??
Apply the ensemble techniques on the Titanic Dataset
Ensemble techniques in machine learning involve combining multiple base models to produce a single, powerful model. The idea is to leverage the strengths of various models to achieve better performance than any single model alone. Ensemble methods can reduce overfitting, improve accuracy, and provide more robust predictions.
Here is my Task Ensemble techniques
Task 4 - Random Forest, GBM and Xgboost
Random forest: Understand & Implement
GBM: Understand & Implement
Xboost: Understand & Implement
Random Forest is an ensemble learning method used for classification and regression tasks. It combines multiple decision trees (hence "forest") to create a more robust model that reduces overfitting and improves accuracy. It uses two key concepts: bagging and feature randomness.
Here is my Task Random Forest, GBM and Xgboost
Task 5 - Hyperparameter Tuning
Understanding
Pick a suitable problem (and dataset) and train a model to fit the problem
Tune the hyperparameters of the model to increase accuracy
Hyperparameter tuning is the process of optimizing the hyperparameters of a machine learning model to achieve the best performance on a given dataset. Hyperparameters are the parameters that are not learned during the training process and must be set before training the model, such as learning rate, number of trees, depth of trees, etc.
Here is my task Hyperparameter Tuning
Task 6 : Image Classification using KMeans Clustering
Understanding K Means Clustering:
Classify a given set of images into a given number of categories using KMeans Clustering using MNIST dataset
KMeans Clustering is an unsupervised machine learning algorithm used to classify data into k clusters. In image classification, KMeans can group similar images based on pixel intensity, color, or any other feature. Although KMeans isn't typically used for traditional image classification tasks like Convolutional Neural Networks (CNNs), it can be an excellent way to cluster images based on similarity when labels are unavailable.
Here is my task Image Classification using KMeans Clustering
Task 7: Anomaly Detection
Anomaly detection is a way to detect erroneous data points in a stream, by looking at statistical differences. Anomaly detection can be done through unsupervised or supervised learning methods.
Anomaly Detection involves identifying data points that deviate significantly from the general trend or pattern within a dataset. These data points, known as outliers or anomalies, may indicate fraudulent activities, errors, or novel events.
Here is my task Anomaly Detection
Task 8: Generative AI Task Using GAN
Develop a generative adversarial network (GAN) model to generate realistic images of a specific category, such as faces, animals, or landscapes. Customize the GAN architecture and train it on a dataset relevant to the chosen category to produce high-quality and diverse synthetic images.
1. Outcome: Implementation and training of GAN model tailored to a specific image category.
2. Generating diverse and realistic synthetic images using the trained GAN.
3. Demonstrating understanding of GAN architecture and its applications in generative tasks
Generative Adversarial Networks (GANs) are a class of machine learning frameworks where two neural networks, the generator and the discriminator, compete against each other. The generator tries to create realistic synthetic data, while the discriminator attempts to distinguish between real and generated data. Through this adversarial process, the generator learns to produce increasingly realistic images.
Here is my task Generative AI Task Using GAN
Task 9: PDF Query Using LangChain
Utilize LangChain, a natural language processing framework, to extract relevant information from PDF documents based on user queries. Develop a system that can interpret user queries, process PDF documents, and retrieve relevant sections or excerpts using language understanding techniques.
Task Outcomes:
1. Development of a PDF query system using LangChain.
2. Implementation of PDF parsing and text extraction functionality.
3. Integration of natural language processing techniques for query interpretation.
4. Testing and validation of the system with various PDF documents and queries.
5. Documentation of system architecture, functionality, and usage guidelines.
Here is my task PDF Query Using LangChain
Task 10: Table Analysis Using PaddleOCR
Employ PaddleOCR, an Optical Character Recognition (OCR) toolkit, to extract and analyze tabular data from images or scanned documents. Develop a pipeline that can accurately detect tables, extract data, and perform analysis such as statistical computations or data visualization.
Task Outcomes:
1. Implementation of a table detection and extraction pipeline using PaddleOCR.
2. Development of algorithms for tabular data analysis, including statistical computations.
3. Integration of data visualization techniques to represent extracted data.
4. Evaluation of pipeline accuracy and performance on various image datasets.
5. Documentation of the process, including code, methodologies, and results.
Here is my task Table Analysis Using PaddleOCR