Hands-On Deep Learning for NLP
Start building deep learning models from scratch! Learn the tools and the techniques for solving real world problems.
Deep Learning is becoming an increasingly popular method for solving problems relating to text classification, word representations, text summarization and machine translation. While we have ample access to theoretical information on Deep Learning, there is very limited guidance on how to actually use these techniques to solve practical problems relating to text data. This hands-on course will not only teach you how to build effective Deep Learning models for text related problems, but will also teach you how to leverage this models in cost effective ways to serve production traffic.
Who will benefit from this course?
- Software Engineers
- Data Scientists
- Data Analysts
- Electrical/Mechanical/Biomedical Engineers
- Research Scientists
What you will learn?
By the end of this course, students will be able to:
- Choose the right models for text classification problems
- Perform appropriate feature engineering for text prediction tasks
- Build a text-classifier from scratch
- Evaluate the performance of text classifiers adequately
- Improve the accuracy of text classifiers
- Generalize text classification models for production use cases
- Hands-on with a mix of theory
- Basic knowledge in Machine Learning and Data Science
- Programming knowledge in Python
- Duration: 2-3 Full Days
- Can be conducted at your premises or as Live Online Training
- Can be customized to meet your specific needs
Subscribe to view our syllabus and get latest articles and updates.
- How Deep Learning works?
- How is Deep Learning different from traditional machine learning?
- Why is Deep Learning excellent for text and natural language data?
- Applications of Deep Learning for NLP
- What are word representations?
- Leverage pre-trained models for NLP tasks
- Word2Vec, GloVe and fastText
- Building and using your first Word2Vec model
- Building a model using fastText
- What is sentence representation?
- Building a sentence representation model with Sent2Vec
ANN for Text Classification
- Introduction to ANN
- Building your first ANN model for document categorization
RNN + LSTM for Text Classification & Generation
- Introduction to RNN and LSTM
- Build an LSTM model for document categorization
- Build an LSTM model for text generation
Day 3 (optional)
- Leveraging word embeddings to improve performance
- How to detect and handle overfitting?
- Model training, testing and deployment strategies
- Introduction to project option
- Guide to sourcing your data
- Guide to building your model
- Grading criteria
- How to submit your project?
- Receive a certification upon successful completion of project