Applied Data Science and Artificial Intelligence (Live-Streaming)
Abstract This course will take you to a different level to pursue your career as a successful data Scientist. In this course you will learn how data scientists exactly work in completing an end-to-end project. Starting from the understanding of a business problem then, data visualization, data preparation and moving through the whole data pipeline […]
Abstract
This course will take you to a different level to pursue your career as a successful data Scientist. In this course you will learn how data scientists exactly work in completing an end-to-end project. Starting from the understanding of a business problem then, data visualization, data preparation and moving through the whole data pipeline till deployment and reporting the results in the most effective way. Not only that, but you will get to work on different cloud-based platforms like Google COLAB, Amazon SageMaker, IBM Watson and MS Azure ML.
Instructor
Eng. Amr Helal, Data Scientist, Markov
Duration
10 Weeks
Sessions
Sundays, 10:00 AM to 2:00PM
Location
Live Streaming (on-line)
Prerequisites
Basics of Machine Learning & Python
Module 1
Module 2
- Collect incidents data for one year
- Segment incidents based on type, address and location
- Python Folium maps
- Geographical visualization of crime distribution
- Intensity maps of crime rate
- Collect world immigration data for 20 years
- Python Choropleth maps
- Geographical features of non-spatial attributes
- GeoJSON files
- Geographic variation of immigrants’ density
Module 3
- Scrap neighbourhoods’ data of different cities
- Python Geopy client
- Locate neighbourhoods’ coordinates
- Python FourSquare API
- Explore particular venues of industries
- K-Means un-supervised algorithm
- Elbow curve evaluation
Module 4
- Exploratory data analysis
- Data pre-processing
- Training and prediction Pipeline
- Logistic Regression, KNN, SVM, Decision Tree, Random Forest
- Algorithms hyper parameters
- Grid Search
- Model evaluation
- Ensemble Learning
- Optimize best candidate algorithm
Module 5
- Images’ features extraction
- Transfer learning
- Python PyTorch
- Torch Vision models (VGG and DenseNet)
- Images pre-processing
- Neural Network model classification layer
- Build classifier with different model architectures
- Python Argument Parser
- Control user arguments to the model
- Different computing platforms (CPU - GPU)
Module 6
- Text detection and localization
- Optical character recognition
- Compute bounding boxes
- Tesseract OCR
- Python OpenCV
- Python Pytesseract
- Confidence of detected text
- Command line arguments
Module 7
- Pre-processing users’ data
- Pre-processing items’ data
- Content-Based filtering
- Collaborative filtering
- Pearson correlation function
- Similarity score
- Predict best match items
Be the first to add a review.
