Update dependencies: modernize PyTorch versions and separate dev depe… #12
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ndencies This commit addresses outdated dependencies and improves dependency management: Changes: - Update PyTorch from 1.6.0 (2020) to latest stable (2.0.0+) - Update torchvision from 0.7.0 (2020) to latest compatible version (0.15.0+) - Add minimum version constraints for devito (>=4.8.0) - Separate runtime and development dependencies: * requirements.txt: Core runtime deps (devito, torch) * requirements-dev.txt: Test-only deps (pytest, torchvision, numpy) - Update CI workflow to use modern PyTorch CPU builds - Update Dockerfile to use latest PyTorch versions - Update pip cache key to reflect new requirements structure Benefits: - Removes 5+ year old security-outdated packages - Clarifies which dependencies are needed for runtime vs testing - Allows users to install Joey without test-only bloat - Maintains compatibility with modern Python ecosystems - No known security vulnerabilities (verified with pip-audit) Torchvision moved to dev dependencies as it's only used in test_lenet.py for MNIST dataset loading.
This notebook provides a detailed performance benchmark comparing Joey and PyTorch implementations of LeNet-5 on the MNIST dataset. Features: - Complete LeNet-5 implementation in both Joey and PyTorch - Performance benchmarks for: * Forward pass (with statistical analysis) * Backward pass (gradient computation) * Complete training loop - Visualization of comparative results - Numerical correctness verification - Detailed Portuguese documentation The notebook demonstrates: - How to build CNNs with Joey - Fair performance comparison with identical initial weights - Statistical analysis with mean and standard deviation - Visual comparisons through matplotlib charts This provides a practical example for users to understand Joey's performance characteristics compared to the industry-standard PyTorch.
Added a new section (0. Instalação) with: - Automatic installation script for Joey and dependencies - Multiple installation options (GitHub, local clone, manual) - Dependency checking for torchvision and matplotlib - Clear instructions in Portuguese for users - Handles missing Joey package gracefully This makes the notebook standalone and easier to use for new users.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
…ndencies
This commit addresses outdated dependencies and improves dependency management:
Changes:
Benefits:
Torchvision moved to dev dependencies as it's only used in test_lenet.py for MNIST dataset loading.