Description
Training objective: Equip participants with advanced Python skills in machine learning, network analysis, deep learning, and natural language processing (NLP). By the end of the course, participants will be able to apply machine learning algorithms for data modeling, perform network analysis, construct deep learning models, and understand NLP foundations.
Objectifs pédagogiques
- Build and evaluate machine learning models using Scikit-learn.
- Conduct network analysis and visualize network structures with NetworkX and PyVis.
- Construct and train neural networks using TensorFlow and Keras.
- Implement NLP pipelines for text processing and work with advanced models using Spacy and HuggingFace Transformers.
Public concerné
Data professionals and developers with strong foundational Python skills who wish to expand their expertise in machine learning, network analysis, deep learning, and NLP.
Prérequis
Completion of Python - Intermediate 2 or equivalent knowledge in data analysis, including Pandas, NumPy, and basic statistics and data visualization.
Déroulé du programme
1
Introduction to machine learning concepts (1 hour)
1h
- Overview of machine learning types (supervised, unsupervised)
- Key concepts: training, testing, validation, overfitting, underfitting
2
Building models with Scikit-learn (2 hours)
2h
- Setting up data for model training and testing
- Implementing linear regression, logistic regression, and decision trees
- Practical exercise: Training and evaluating models on a sample dataset
3
Model evaluation and validation techniques (2 hours)
2h
- Understanding model evaluation metrics (accuracy, precision, recall, F1-score)
- Cross-validation and hyperparameter tuning
- Practical exercise: Applying cross-validation and tuning to improve model performance
4
Clustering and dimensionality reduction (1 hour)
1h
- Implementing clustering algorithms: K-Means, hierarchical clustering
- Dimensionality reduction with PCA (Principal Component Analysis)
- Practical exercise: Applying clustering techniques and PCA on datasets for data insight
5
Network analysis with NetworkX and PyVis (2.5 hours)
5h
- Introduction to network structures and graph theory basics
- Creating and analyzing networks with NetworkX
- Visualizing networks with PyVis
- Practical exercise: Building and visualizing a network, identifying key nodes and communities
6
Introduction to deep learning with TensorFlow and Keras (3.5 hours)
5h
- Overview of neural networks and deep learning concepts
- Building and training a neural network using Keras
- Understanding activation functions, optimizers, and loss functions
- Practical exercise: Developing a simple neural network model to classify data, with focus on model architecture and optimization
7
Deep learning for advanced applications (2 hours)
2h
- Working with convolutional neural networks (CNNs) for image processing
- Introduction to recurrent neural networks (RNNs) for sequential data
- Practical exercise: Building a CNN for image classification or an RNN for text sequence prediction
8
Natural Language Processing (NLP) fundamentals (4 hours)
4h
- Overview of NLP: tokenization, stemming, lemmatization
- Implementing basic NLP with NLTK and Scikit-learn
- Advanced NLP with Spacy and HuggingFace Transformers for model fine-tuning and text generation
- Practical exercise: Building an NLP pipeline for text processing, sentiment analysis, or text generation using Spacy and Transformers
Informations
Durée
3 jour(s)
18h
Tarif
Sur demande
