Machine Learning Outline
Course Number: PYTH-126
Duration: 4 days (26 hours)
Format: Live, hands-on
Machine Learning with Python Training Overview
Accelebrate’s private, onsite or online Comprehensive Machine Learning with Python training course builds on our Comprehensive Data Science with Python class and teaches attendees how to write machine learning applications in Python.
Location and Pricing
Accelebrate offers instructor-led enterprise training for groups of 3 or more online or at your site. Most Accelebrate classes can be flexibly scheduled for your group, including delivery in half-day segments across a week or set of weeks. To receive a customized proposal and price quote for private corporate training on-site or online, please contact us.
Objectives
- Understand machine learning as a useful tool for predictive models
- Know when to reach for machine learning as a tool
- Implement data preprocessing for an ML workflow
- Understand the difference between supervised and unsupervised tasks
- Implement several classification algorithms
- Evaluate model performance using a variety of metrics
- Use labeled text to perform document classification
- Compare models across a workflow
- Implement regression algorithm variations
- Understand clustering approaches to data
- Interpret labels generated from clustering
- Transform unstructured text data into structured data
- Understand text-specific data preparation
- Visualize frequency data from text sources
- Perform topic modeling on a collection of documents
Prerequisites
All attendees should have completed the Comprehensive Data Science with Python class or have equivalent experience.
Outline
- Anaconda Computing Environment
- Importing and manipulating Data with Pandas
- Exploratory Data Analysis with Pandas and Seaborn
- Numpy ndarrays versus Pandas Dataframes
- Machine Learning Theory
- Data pre-processing
- Missing Data
- Dummy Coding
- Standardization
- Data Validation Strategies
- Supervised Versus Unsupervised Learning
- Linear Regression
- Penalized Linear Regression
- Stochastic Gradient Descent
- Decision Tree Regressor
- Random Forest Regression
- Gradient Boosting Regressor
- Scoring New Data Sets
- Cross Validation
- Variance-Bias Tradeoff
- Feature Importance
- Logistic Regression
- LASSO
- Support Vector Machine
- Random Forest
- Ensemble Methods
- Feature Importance
- Scoring New Data Sets
- Cross Validation
- Preparing Data for Ingestion
- K-Means Clustering
- Visualizing Clusters
- Comparison of Clustering Methods
- Agglomerative Clustering and DBSCAN
- Evaluating Cluster Performance with Silhouette Scores
- Scaling
- Mean Shift, Affinity Propagation and Birch
- Scaling Clustering with mini-batch approaches
- Understand average versus conditional treatment effects
- Estimating conditional average treatment effects for a sample
- Summarizing and Interpreting
- Intro to H20
- Launching the cluster, checking status
- Data Import, manipulation in H20
- Fitting models in H20
- Generalized Linear Models
- naïve bayes
- Random forest
- Gradient boosting machine (GBM)
- Ensemble model building
- automl
- data preparation
- leaderboards
- Methods for explaining modeling output
- Transforming Raw Text Data into a Corpus of Documents
- Identifying Methods for Representing Text Data
- Transformations of Text Data
- Summarizing a Corpus into a TF—IDF Matrix
- Visualizing Word Frequencies
- Installing And Accessing Sample Text Corpora
- Tokenizing Text
- Cleaning/Processing Tokens
- Segmentation
- Tagging And Categorizing Tokens
- Stopwords
- Vectorization Schemes for Representing Text
- Parts-of-speech (POS) Tagging
- Sentiment Analysis
- Topic Modeling with Latent Semantic Analysis
- Unsupervised Machine Learning and Text Data
- Topic Modeling via Clustering
- Supervised Machine Learning Applications in NLP
Training Materials
All Machine Learning with Python students receive courseware covering the topics in the class.
Software Requirements
- Windows, Mac, or Linux with at least 8 GB RAM
- A current version of Anaconda for Python 3.x
- Related lab files that Accelebrate will provide