Publication date : 02/22/2024
Course : Deep Learning with PyTorch
Practical course - 4d
- 28h00 - Ref. DLT
|
Download in PDF format
Share this course by email
 | How to manipulate images and text with PyTorch |
| Set up neural network training from scratch or using transfer learning |
| PyTorch modules for loading data |
| Knowledge of distributed drives |
| Knowledge of new meta-architectures such as transformers and off-line curves |
Intended audience
Machine learning designers and developers, data scientists, AI engineers.
Prerequisites
Python and machine learning.
Course schedule
1 Getting started with PyTorch
- PyTorch and its fundamental principles.
- Install PyTorch and related components.
- Comparison between the Numpy and PyTorch libraries.
- PyTorch vs Tensorflow.
- Principles of distributed computing.
Hands-on work
Installing PyTorch. Tensor and matrix manipulation.
2 PyTorch submodules for training neural networks
- Presentation of Pytorch submodules for training neural networks.
- A reminder of forward propagation.
- A reminder of gradient backpropagation.
- Data loading.
- Define a convolution neural network with the torch.nn package, train the model, test it.
Hands-on work
Setting up a CNN network for image classification.
3 Transfer learning and the use of pre-trained networks
- The principle of transfer learning.
- Examples of transfer learning.
- Transfer learning steps in machine learning projects.
- Use of pre-trained networks.
Hands-on work
Repetition of previous exercises, to improve metrics with the implementation of transfer learning.
4 Meta-architectures for complex projects
- Introduction to meta-architectures.
- The problem of object detection.
- Image segmentation problems.
- UNet network architecture: encoder-decoder blocks and PyTorch.
Hands-on work
Creation of a simple UNet model for image segmentation. Comparison with transfer learning for UNet.
etourchariot
5 NLP with PyTorch and spaCy
- Automatic natural language processing (NLP).
- The benefits of PyTorch and spaCy.
- Pipeline principle.
- Text processing .
- Recurrent network drive / biLSTM.
- Using PyTorch and spaCy for NLP.
Hands-on work
Topic modelling on movie reviews. Sentiment analysis on tweets.
6 Transformers and attention mechanisms
- Transformers for automatic language processing.
- Detail of attention mechanisms.
- The attention mechanism applied to a sequence: self-attention.
- Transformer operation.
Hands-on work
Setting up a translation model.
Customer reviews
4,1 / 5
Customer reviews are based on end-of-course evaluations. The score is calculated from all evaluations within the past year. Only reviews with a textual comment are displayed.
CARLO MARIA Z.
07/10/25
4 / 5
The content is very dense and it's normal that you can't go into each part in depth. The teaching is good. The practical work did not seem to me to be sufficiently supervised and, after the event, corrected (although examples of working code were provided).
PIERRE G.
07/10/25
4 / 5
Good overview of pytorch and deep learning. I appreciated the frequent discussions and Q&A with the teacher, who was always very available, which for me was the main added value of this course. What I didn't like were the pre-filled practical exercises, which didn't really allow you to think for yourself, and the time spent reading the 'independent' practical exercises, which were sometimes too long.
AURÉLIEN B.
07/10/25
4 / 5
Overall satisfied, perhaps at my level on the subject a few chapters less for a little more practical code and the workflow pytorch itself. Otherwise content (code + presentation) very complete and functional therefore qualitative.
PARTICIPANTS
Machine learning designers and developers, data scientists, AI engineers.
PREREQUISITES
Python and machine learning.
TRAINER QUALIFICATIONS
The experts leading the training are specialists in the covered subjects. They have been approved by our instructional teams for both their professional knowledge and their teaching ability, for each course they teach. They have at least five to ten years of experience in their field and hold (or have held) decision-making positions in companies.
ASSESSMENT TERMS
The trainer evaluates each participant’s academic progress throughout the training using multiple choice, scenarios, hands-on work and more.
Participants also complete a placement test before and after the course to measure the skills they’ve developed.
TERMS AND DEADLINES
Registration must be completed 24 hours before the start of the training.
ACCESSIBILITY FOR PEOPLE WITH DISABILITIES
Do you need special accessibility accommodations? Contact Mrs. Fosse, Disability Manager, at psh-accueil@orsys.fr to review your request and its feasibility.
Machine learning designers and developers, data scientists, AI engineers.
PREREQUISITES
Python and machine learning.
TRAINER QUALIFICATIONS
The experts leading the training are specialists in the covered subjects. They have been approved by our instructional teams for both their professional knowledge and their teaching ability, for each course they teach. They have at least five to ten years of experience in their field and hold (or have held) decision-making positions in companies.
ASSESSMENT TERMS
The trainer evaluates each participant’s academic progress throughout the training using multiple choice, scenarios, hands-on work and more.
Participants also complete a placement test before and after the course to measure the skills they’ve developed.
TEACHING AIDS AND TECHNICAL RESOURCES
• The main teaching aids and instructional methods used in the training are audiovisual aids, documentation and course material, hands-on application exercises and corrected exercises for practical training courses, case studies and coverage of real cases for training seminars.
• At the end of each course or seminar, ORSYS provides participants with a course evaluation questionnaire that is analysed by our instructional teams.
• A check-in sheet for each half-day of attendance is provided at the end of the training, along with a course completion certificate if the trainee attended the entire session.
• The main teaching aids and instructional methods used in the training are audiovisual aids, documentation and course material, hands-on application exercises and corrected exercises for practical training courses, case studies and coverage of real cases for training seminars.
• At the end of each course or seminar, ORSYS provides participants with a course evaluation questionnaire that is analysed by our instructional teams.
• A check-in sheet for each half-day of attendance is provided at the end of the training, along with a course completion certificate if the trainee attended the entire session.
TERMS AND DEADLINES
Registration must be completed 24 hours before the start of the training.
ACCESSIBILITY FOR PEOPLE WITH DISABILITIES
Do you need special accessibility accommodations? Contact Mrs. Fosse, Disability Manager, at psh-accueil@orsys.fr to review your request and its feasibility.
Dates and locations
Select your location or opt for the remote class then choose your date.
Remote class
Last places available
Guaranteed date, in person or remotely
Guaranteed session