Zero to Deep Learning Masterclasses

Learn with us: www.zerotodeeplearning.com

Titles

Data Manipulation with Pandas

Learn to analyze real world data using Python & Pandas. Import data from multiple sources, clean, reshape, impute and visualize your data.

Machine Learning with Scikit Learn

Learn what machine learning is and use Python, Pandas and Scikit-Learn to build machine learning models on real world data.

Neural Networks with Keras

Learn what deep learning is and learn to use Python, Keras and Tensorflow to build deep learning models on real world data.

Pandas Matplotlib Seaborn

Recap of main functionality of Pandas, Matplotlib and Seaborn with lots of exercises.

Regression with Scikit Learn

Linear regression with single feature and multiple features using Scikit Learn. Model comparison on housing data.

Classification with Scikit Learn

Binary classification with Scikit Learn. Numerical and categorical features. Model comparison.

Clustering with Scikit Learn

Clustering with Scikit Learn on Iris flower dataset. KMeans, DBScan, Silhouette Score.

Feature Engineering with Scikit Learn

Feature engineering with Scikit Learn on Titanic Dataset. Use transformers to create new features and improve your model performance.

Model Evaluation and Dimensionality Reduction with Scikit Learn

Evaluating model performance using Pipelines, Cross-Validation, Learning curves and Dimensionality Reduction.

Introduction to Deep Learning with Keras

First steps with Neural Networks using Keras to solve classifiation and regression problems.

Keras Functional API and Model Inspection

Functional API, Callbacks, Inner Layer visualization and Tensorboard.

The Power of TensorFlow

Explore other APIs in TensorFlow. Dataset API, Feature Columns API, Layers and Model API.

Hyperparameter tuning

Tune hyperparameters using Grid Search, Random Search and Keras Autotuner.

Image Classification

Learn about several simple datasets for image classification and build a fully connected network that uses pixels as features

Convolutional Neural Networks

Build a convolutional neural network to classify images and optimize its architecture.

Pre-trained Models

Discover pre-trained models in Keras applications and Tensorflow Hub.

Transfer Learning

Adapt pre-trained models to your use case.

Data Augmentation

Flow data from a directory using data generators in Keras and Tensorflow

Image search

Retrieve images by similarity using bottleneck features.

Bag of Words Features for Text

Classify different languages using Scikit-Learn and Count Vectors

Natural Language Processing

Classify sentiment using bag of words with Sciki-Learn and NLTK

Word Embeddings

Classify sentiment using word embeddings and convolutional neural networks in Keras.

Time Series Forecasting

Forecast energy demand in Ontario using Fully connected and recurrent networks

Sentiment Classification with Recurrent Neural Networks

Classify sentiment using RNNs

Machine Translation with Recurrent Neural Networks

Train a neural network to perform character translation using RNNs

Model Serving with Flask

Serve Scikit Learn and Tensorflow models using Flask

Word Embeddings with Gensim

Visualize word embeddings with Gensim, Glove, Matplotlib and Plotly Express

Gradient Boosting with XGBoost and LightGBM

Exploration of binary classification with several boosting algorithms. Hyperparameter tuning, grid search and categorical features.

Real World ML Car Prices Regression

End-to-End Machine Learning exercise on messy dataset. Data cleaning, feature engineering, model selection with Scikit Learn, XGBoost and LightGBM.

Real World ML Ads Classification

End-to-End Machine Learning exercise on unbalanced text dataset. Data cleaning, model selection with Scikit Learn, XGBoost, LightGBM, FastText and Gensim.

zerotodeeplearning / ztdl-masterclasses