Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bm/load_from_checkpoint #116

Merged
merged 95 commits into from
Jun 25, 2024
Merged

Bm/load_from_checkpoint #116

merged 95 commits into from
Jun 25, 2024

Conversation

BenjaminMidtvedt
Copy link
Collaborator

@BenjaminMidtvedt BenjaminMidtvedt commented Jun 20, 2024 •
edited
Loading

Allows loading deeplay applications from checkpoints. This makes integration with lightning much more convenient. Example:

net = ...
app = dl.Classifier(net).build()

app.fit(data, max_epochs=100)

# reload best model
path = app.trainer.checkpoint_callback.best_model_path
best_app = dl.Classifier.load_from_checkpoint(path)

BenjaminMidtvedt and others added 30 commits April 22, 2024 14:50
@BenjaminMidtvedt
implement test behavior with no test inputs
21d4246
@BenjaminMidtvedt
docs
e3e5089
@BenjaminMidtvedt
Refactor code in deeplay/activelearning/data.py, deeplay/activelearni…
12bd38e 
…ng/strategies/uncertainty.py, deeplay/applications/application.py, and deeplay/activelearning/strategies/strategy.py
@BenjaminMidtvedt
clear config before multi
2cf1cd9
@BenjaminMidtvedt
update to use absolute imports
f9dee3b
@BenjaminMidtvedt
Refactor import statements in test_selectors.py
b1e18f4
@BenjaminMidtvedt
Bm/fix-config-nested-new
3c04e1e 
Fixes an issue with multi-blocks not having their configuration correctly cleared if created multiple times
@BenjaminMidtvedt
remove test file
16758c5
@BenjaminMidtvedt
Fix issue with clearing configuration before creating multiple blocks
e30a6f7
@BenjaminMidtvedt
Fix issue with configuring upsample in Conv2dBlock
6714c65
@BenjaminMidtvedt
Re-enable test_strided_multi in test_conv.py
08ac96c
@BenjaminMidtvedt
add stubs
95cf49f
@BenjaminMidtvedt
Refactor residual function in Conv2dBlock to support flexible layer o…
2279397 
…rder
@BenjaminMidtvedt
Refactor available_styles method in DeeplayModule to use classmethod
5697e66
@BenjaminMidtvedt
Implement script to create stubs with style typing
e329512
@BenjaminMidtvedt
Refactor Conv2dBlock and related functions in conv2d.pyi
af7d358
@BenjaminMidtvedt
Remove publish script from package.json
2bea2ed
@BenjaminMidtvedt
Add .gitignore entry for package.json
21fb687
@BenjaminMidtvedt
add docs
139e7ef
@BenjaminMidtvedt
delete data
a449dd1
@BenjaminMidtvedt
Update documentation and style guide
6af3f43
@BenjaminMidtvedt
Update naming conventions, imports, documentation, and testing guidel…
2af03bf 
…ines
@BenjaminMidtvedt
Refactor Conv2dBlock and related functions in conv2d.pyi
dfa89e1
@BenjaminMidtvedt
Refactor Conv2dBlock and related functions in conv2d.pyi
f1a03ab
@BenjaminMidtvedt
Refactor Conv2dBlock and related functions in conv2d.pyi
7001e42
@BenjaminMidtvedt
Merge branch 'develop' into bm/docs
10a8fe8
@BenjaminMidtvedt
§
f10cb09
@BenjaminMidtvedt
u
1dee8a7
@BenjaminMidtvedt
Merge branch 'bm/docs' of https://github.com/softmatterlab/DeepTorch …
346ee21 
…into bm/docs
@BenjaminMidtvedt
baseline shape compute
28c938b
@BenjaminMidtvedt
Pickle using dill instead of pickle
33f5ce3
@BenjaminMidtvedt
chore: Update requirements.txt with dill package
a8f0bff
@BenjaminMidtvedt
feat: Update requirements.txt with numpy>=1.24.1
0962c3f
@BenjaminMidtvedt
chore: Update numpy version in requirements.txt
a834343
@BenjaminMidtvedt
add a cleanup method before create
dafe7e8
@BenjaminMidtvedt
add test case for double create bug
26847fe
@BenjaminMidtvedt
remove unused import
c48a79c
@BenjaminMidtvedt
Reduce use of deepcopy (for speed)
2f4dd36
@BenjaminMidtvedt
Reduce use of deepcopy (for speed)
56fd135
@BenjaminMidtvedt
chore: Remove print statement in layer.py
6fc049a
@BenjaminMidtvedt
refactor: Add non-class type arguments to External class initialization
da18c7e
@BenjaminMidtvedt
Refactor import statements in deeplay.blocks.ls.py and stylestubgen.py
18a3236
@BenjaminMidtvedt
refactor: Add stateful decorator for methods modifying object state
9136a13
@BenjaminMidtvedt
refactor: Add initialization steps for object attributes in meta.py
2f0ce64
@BenjaminMidtvedt
refactor new method to not use deepcopy
7be1b07
@BenjaminMidtvedt
refactor: Remove unused import and regex module from deeplay/module.py
561c96d
@BenjaminMidtvedt
Merge branch 'bm/load_from_checkpoint' of https://github.com/softmatt…
f34e043 
…erlab/DeepTorch into bm/load_from_checkpoint
@BenjaminMidtvedt
refactor: Update ExtendedConstructorMeta to use _module_state instead…
4d2949a 
… of _is_top_level
@BenjaminMidtvedt
refactor: Remove commented out code in deeplay/module.py
616f783
@BenjaminMidtvedt
refactor: Add typing_extensions.Self import to stylestubgen.py
79f279c
@BenjaminMidtvedt
Refactor test_encdec.py and test_gnn.py to use gnn.create() instead o…
9f5b19b 
…f gnn.build() for consistency
@BenjaminMidtvedt
add application tests
e535855
@BenjaminMidtvedt
refactor: Update log_metrics method to preprocess y_hat and y before …
4e8d470 
…passing them to metrics
@BenjaminMidtvedt
refactor: Remove unused imports in deeplay/tests/applications/base.py
daf2dd8
@BenjaminMidtvedt
del notebook
a004c7e
@BenjaminMidtvedt
Merge branch 'develop' into bm/load_from_checkpoint
b80dbc6
@BenjaminMidtvedt
del notebook
26c21eb
@BenjaminMidtvedt
refactor: Add _non_classtype_args attribute to External class
9e24c1b
@giovannivolpe giovannivolpe merged commit 9705dd3 into develop Jun 25, 2024
12 checks passed
@giovannivolpe giovannivolpe deleted the bm/load_from_checkpoint branch June 25, 2024 14:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@BenjaminMidtvedt @giovannivolpe