Real World Auto Machine Learning Bootcamp: Build 14 Projects

Explore real-world heart disease detection using automated machine learning techniques on a tabular dataset, with a python-based pipeline that automates model building and evaluation.

Develop a heart attack risk prediction app by performing data analysis, standardisation, modelling, and predictions, then apply automated machine learning with essential libraries in Google Colab using drive.

Analyze a medical dataset by examining shape (303 rows, 11 columns), nulls, correlations, and distributions; explore categorical features, identify heart attack risk, and prepare data for ml using standardization.

Explore data preparation and model building in real world auto machine learning, including standard scaling, train-test split, and comparing logistic regression, decision trees, random forest, and SVM.

Explore model building by comparing decision-tree classifiers with logistic regression, tuning hyperparameters to maximize accuracy. Assess results using confusion matrices and accuracy scores to identify the best performing model.

Explore AutoML techniques using open-source Python libraries to automate model selection, preprocessing, and evaluation, building and comparing pipelines, and saving the best model for predictions.

Build a credit card fraud detection project using anomaly detection on transaction data, exploring isolation forest and outlier methods, and automating model deployment with autoML.

Explore ID card fraud detection using principal component analysis for dimensionality reduction, preparing a numerical dataset with 28 components for automated machine learning model building and prediction.

Explore data analysis for fraud detection: inspect dataset shape, check for nulls, visualize normal versus fraud cases, address imbalance, and prepare independent features and the target for modeling.

Build models for anomaly detection using isolation forest; compare with local outlier factor and svm, evaluate accuracy with the confusion matrix, and explore automated machine learning approaches.

Explore automated machine learning with PyCaret to compare and train classification, anomaly detection, and other models, set up data, evaluate, and deploy the best model.

Explore predicting flight prices with machine learning and AutoML, using a flight price dataset to perform data analysis, feature engineering, model building, and automated hyperparameter tuning.

Import libraries and dataset, mount drive in Colab, and combine train/test data to prepare a 13,354-row dataset for predicting price using airline, route, departure and arrival times, stops, and class.

Analyze the relationship between price and airline features using data analysis, box plots, and exploration of sources, destinations, durations, and stops, with emphasis on encoding categorical variables.

Learn practical feature engineering by converting mixed data types into numeric and categorical features, extracting date and time, encoding travel attributes like stops and class, and cleaning irrelevant fields.

Split arrival and departure times into hours and minutes, encode routes and airlines, and prepare data with one-hot and label encoding for modeling.

Explore feature selection by splitting data into X and y, applying lasso via select from model, and visualizing feature importance to identify key predictors.

Build a model with random forest regression, tune hyperparameters via randomized search, and evaluate performance using distribution plots, training versus test predictions, and the r-squared score.

Explore automated machine learning with auto-sklearn to test multiple models, optimize hyperparameters, preprocess data, and identify the best model via mean square error and root mean square error.

Discover how to forecast petrol prices with deep learning and auto ml, comparing lstm and arima models, exploring data preprocessing and econometric models.

Learn to import libraries and data for petrol price forecasting with TensorFlow and auto machine learning, mount Google Drive, upload datasets, and begin data analysis, preprocessing, and model building.

Analyze the dataset, check for nulls, clean data, rename the patrol column, explore distributions with a scatter plot, apply min-max scaling, and perform a sequential 65/35 train-test split for forecasting.

This lecture explains data preprocessing for time-series, converting raw data into input features using time steps, creating F1, F2, F3 features and an output, and reshaping data for training.

Build and import an LSD sequential model, configure layers, train with mean squared error on training data with validation data, monitor loss and predictions, and plot results.

Explore building and forecasting time series with auto keras and auto ml libraries. Learn data loading, preprocessing, model training, evaluation, and exporting for practical automation.

Explore how auto machine learning and neural networks, including sequential models, predict bank customer churn, using data preprocessing, feature engineering, and model tuning to build a churn forecast.

Import essential libraries, mount Google Drive, and load the dataset. Begin data analysis, drop irrelevant columns, handle missing values and outliers, and build with auto machine learning.

Analyze data to reveal how credit score, geography, age, balance, and products influence customer exit. Present findings with a pie chart and box plots comparing exited versus retained customers.

Explore feature engineering to boost model accuracy by creating new features like salary balance ratio, applying preprocessing, encoding (label and one-hot), and scaling with min-max and standard scalers.

Build and predict with artificial neural networks using a sequential model, configure multiple layers and activation functions, and optimize loss, then assess accuracy with a confusion matrix.

Explore automated machine learning with the AutoML tool. Install and configure the library, prepare data, train multiple models, compare the leaderboard, and select the best model for predictions.

Develop an air quality index predictor for Bangalore city using auto machine learning, trained with 2013–2020 data, and deployed via Flask API on Heroku.

Develop an air quality index predictor by importing libraries and datasets from 2013–2018, using Bangalore data with temperature, humidity, wind, and PM levels, and explore autoML and MLP deployment.

Analyze air quality dataset by exploring distributions, handling missing values, and describing continuous variables; visualize with seaborn distribution plots and heatmaps to reveal correlations and insights.

Assess data normality, apply scaling and standardization, identify feature importance, and prepare data with a train/test split for modeling.

Explore model building in real world auto machine learning bootcamp by implementing linear regression, gradient boosting, and random forest, with hyperparameter tuning and evaluation using R-squared and RMSE.

Build a random forest regression model and tune key hyperparameters. Validate with five-fold cross-validation to reach about 76% accuracy, comparing with linear and boosted models.

Explore automated machine learning with the TPOT library, building and evaluating regression models, optimizing pipelines via genetic programming, and preparing deployment via a Flask API.

Learn to deploy a machine learning model with a Flask API, load a pickle model, and build a simple web app and Heroku deployment workflow.

Launch an end-to-end auto machine learning project to predict rain, build a deployable model with the bike library, and expose it via a Flask API on Heroku.

Explore building a rain prediction web app using machine learning and auto machine learning. Learn data analysis, missing value imputation, encoding categorical variables, feature scaling, training multiple models, evaluating accuracy, and deploying the model with a cloud api and Heroku.

Analyze a large dataset to identify missing values, categorize features into numerical (continuous or discrete) and categorical, and distinguish discrete from continuous using a twenty-five unique values threshold; plan imputation.

Identify data types and distributions, apply random sample imputation for missing values, and use mean imputation for numerical features, then explore correlations and prepare for label encoding, standardization, and modeling.

Learn feature engineering for real-world auto machine learning, including one-hot and level encoding, handling location and date, imputation, scaling, and train-test splitting to prep data for models.

Compare and evaluate random forest, gradient boosting, and nearest neighbor classifiers on a project dataset, reporting accuracies around 80–85% and preparing deployment-ready models.

Explore automated machine learning with pycaret to prepare data, train models, and deploy predictions quickly. Compare models to identify best performing classifiers.

Learn to deploy a machine learning model with a Flask API, load the pickle model, create a web interface, and prepare deployment to Heroku for real-world predictions.

Deploy a machine learning model to Heroku by creating a repository, uploading files, installing dependencies, connecting the repo, and launching the prediction app.

Explore building a pizza price prediction model with AutoML, data analysis, and feature engineering, using Evil ML to compare models like XGBoost and logistic regression.

Import libraries and the Kaggle pizza dataset in a Google Colab workflow to build a price model using machine learning and AutoML, including one hot encoding and train-test split.

Analyze data by checking for null values, converting types, renaming and removing columns, then visualize distributions with pie charts, histograms, and distribution plots across price, diameter, toppings, and size.

Use one hot encoding and label encoding to convert categorical attributes into readable features. Create dummy variables with category encoders, split data into X and y, prepare Xtrain and ytrain.

Build and compare machine learning models using XGBoost regressor and support vector regression, tune with grid search cv and randomized search cv, and evaluate with r squared.

Automate model building, hyper tuning, and evaluation with AutoML and Evil ML for regression, compare pipelines, and select using R square to save predictions of pizza prices.

Build a model to predict the final score of an IPL match using automated machine learning, leveraging ball-by-ball data, feature encoding, and genetic programming pipelines.

Explore importing libraries and datasets, cleaning and analyzing data, and building a model with an automated machine learning library on a real-world IPL dataset.

Perform data analysis and cleaning by exploring the dataset, dropping noncontributory columns, selecting consistent teams, removing the first five overs, and converting date to datetime for subsequent modeling.

Explore data preprocessing techniques, including label encoding and one-hot encoding, to prepare categorical features for modeling, then standardize and split data for training and testing.

Explore building regression models with machine learning algorithms, including lasso and random forest, and tune hyperparameters with grid search and randomized cross-validation to optimize accuracy.

Explore automated machine learning with the tpot auto ml library, using genetic programming to optimize regression models, perform cross-validation, and evaluate with negative mean absolute error.

Explore building regression models with the TPOT AutoML library, identifying the best pipeline, evaluating with negative mean absolute error, and validating predictions through distribution plots and cross-validation.