Description
Duration
- This specific Code.Learn program lasts 3 weeks (3 Fridays, 3 Saturdays) with 35 hours of lectures and hands-on exercise on a real life project.
Key Objectives – Curriculum (High Level):
The core perspectives of this program will be to present, explore and adequately cover with extended real-life business case studies & industry scenarios the following thematic units:
Day 1
- Python & Numpy
Day 2
- Data analysis with Pandas
- Data visualization with matplotlib
- Hands-on example (data analysis)
- Machine Learning
- Assignment 1 (Data Analysis)
Day 3
- Popular ML algorithms
- Supervised:
- k Nearest Neighbors
- Naive Bayes
- Decision Trees
- Neural Networks
- Unsupervised:
- k means.
- Hierarchical clustering.
- Intro to scikit-learn:
- Library structure
- Defining and training models.
- Evaluating the results
- Hyperparameter optimization
Day 4
- Advanced ML concepts
- Bias/variance tradeoff
- Regularization
- Typical holdout strategy
- Cross-validation
- Handling missing data
- Feature encoding
- Feature scaling
- Curse of dimensionality
- Feature selection
- Feature extraction
- Class imbalance
- Confusion matrix metrics
- Over/Under sampling
- Implementing advanced pre-processing/evaluation techniques in python. Complete supervised ML workflow
- Hands-on example (Classification)
- Assignment 2 (Classification)
Day 5
- Natural Language Processing
- Intro to unstructured data
- String operations
- Tokenization
- Stopword removal
- Stemming/Lemmatization
- Vector Space Models
- The Bag-of-words model
- TF-IDF
- Hands-on example (text classification):
- Load a text dataset.
- Process the documents through NLP techniques
- Convert it to a structured ML problem (tf-idf)
- Train and evaluate a classification model
- Examine the impact of the most important parameters to the result
Day 6
- Clustering
- – Intro to clustering
- – Step-by-step k-means
- – Clusteing solution evaluation
- Clustering in scikit-learn
- Hands-on example (text clustering)
- Load a Movie recommendation dataset
- Process it using MLP techniques
- Convert ti to a structured ML problem
- Train a clustring model
- Use evaluation metrics to optimize the above model
- Empirically evaluate solution
- Closing remarks
- General tips for ML
- Tips for clustering
- Tips for NLP
- How to approach a real life ML problem
- Real life applications and ideas
- Questions/Discussion
- Work in class
- Assignment 3 (Reuters dataset clustering)
Target Audience
Data analysts, data scientists, data engineers, BI engineers/developers, information-visualization-analytics professionals as well as computer scientists, software engineers, developers and consultants who vision their future in data-relevant career paths are welcome to participate to this code.learn program and unlock the full potentiality of the topics taught by upskilling their future career.
