Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
Y
yapt
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
llvm
yapt
Repository graph
Repository graph
You can move around the graph by using the arrow keys.
main
Select Git revision
Branches
2
main
default
old_main
2 results
Begin with the selected commit
Created with Raphaël 2.2.0
13
Mar
9
28
Feb
22
20
19
3
16
Dec
14
Oct
9
8
7
13
Sep
14
Aug
13
12
22
Jul
1
30
Jun
29
21
Apr
20
19
18
Add deepcopy / use implicit device
main
main
Better configuration dumping
Fixed train/eval mode confusion
Fixed channel/filter confusion: channel -> filter
Fixed pytorch-lightning version requirements
Fixed an issue with example_input_array
Removed unused files, outdated req, and updated readme.md
Some reformatting
Improved StructuredPruningCallback
Added EagleEye pruning
Reorganized structure pruning directory
Added functionality to prepare for more pruning methods
Added test for all image classification net architectures in torchvision
Organize supported operators better
Fixed failed tests due to pytorch version change
Improved CFG visual and type annotation
Fixed issues on last pruning epoch
Slightly extended DefaultShapeHandler
Supports loading directly from pruned checkpoint
Added lr rewinding and checkpoint reloading
Improved checkpointing and accuracy logging
_run_pruning -> prune_module_ (so it can be used directly)
Fixed a problem about onnx_output_names
Fixed misuse of pl_module.result
Implemented pruning on ConvTranspose2D
Fixed pruning not going all the way till last channel
Fixed sample_input on custom-collated dataloader
Fixed last prune step not saved
Added capability to prune the conv layer before linear layer
Fixed a bug when dealing with layers that prunes 0 channels
Added switch on ONNX model export
Remove a portion of filters from every layer
Fixed a problem
Improved pruning logging
Added protection ratio for layer to prevent all channels being pruned
Reformatted files
Added test for callback (end to end) and fixed some problems
Fixed and improved setup.cfg
This only works with lightning 1.3.x
Updated env requirement
Loading