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

Enable customizing training loss and val metric, add Time-LLM #570

Merged
merged 43 commits into from
Feb 19, 2025
Merged

Enable customizing training loss and val metric, add Time-LLM #570

merged 43 commits into from
Feb 19, 2025

Conversation

WenjieDu
Copy link
Owner

What does this PR do?

  1. fixing allow customized loss function for imputation/informer #432 and Allow users to customize training loss function and validation metric #525;
  2. fixing Add TimeLLM #564;

Before submitting

  • This PR is made to fix a typo or improve the docs (you can dismiss the other checks if this is the case).
  • Was this discussed/approved via a GitHub issue? Please add a link to it if that's the case.
  • I have commented my code, particularly in hard-to-understand areas.
  • I have written necessary tests and already run them locally.

WenjieDu added 30 commits May 8, 2024 14:58
@WenjieDu
feat: enable users to customize training loss func and val metric func;
41b00a3
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
06ba371
@WenjieDu
refactor: move pypots.utils.metrics to pypots.nn.functional;
620a95c
@WenjieDu
feat: add BaseLoss and BaseMetric;
c7dc6be
@WenjieDu
feat: add TimeLLM layers;
0649b5c
@WenjieDu
feat: enable to disable positional encoding in PatchTST embedding;
c748e5c
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
83879ff 
# Conflicts:
#	pypots/classification/base.py
#	pypots/classification/grud/core.py
#	pypots/classification/raindrop/core.py
#	pypots/clustering/base.py
#	pypots/clustering/crli/model.py
#	pypots/clustering/vader/core.py
#	pypots/clustering/vader/model.py
#	pypots/forecasting/base.py
#	pypots/forecasting/csdi/core.py
#	pypots/forecasting/csdi/model.py
#	pypots/forecasting/template/model.py
#	pypots/imputation/base.py
#	pypots/imputation/csdi/core.py
#	pypots/imputation/csdi/model.py
#	pypots/imputation/gpvae/model.py
#	pypots/imputation/nonstationary_transformer/core.py
#	pypots/imputation/saits/core.py
#	pypots/imputation/saits/model.py
#	pypots/imputation/template/model.py
#	pypots/imputation/usgan/core.py
#	pypots/nn/modules/csdi/backbone.py
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
6b5c449
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
92773f1
@WenjieDu
refactor: make FITS able to apply customized loss func;
f674a3c
@WenjieDu
refactor: replace arg training with self attribute in new added models;
40e9602
@WenjieDu
fix: globally replace importing from with ;
cc286ee
@WenjieDu
refactor: still keep pypots.utils.metrics for future compatibility;
ba4588c
@WenjieDu
refactor: do not expose by default;
86340d0
@WenjieDu
refactor: remove lingting issues;
b4b5b48
@WenjieDu
fix: rename labels in classification CSAI into y;
af3bf85
@WenjieDu
refactor: add customized loss and metric funcs for CSAI models;
054b3c9
@WenjieDu
docs: update doc strings;
9bc40f4
@WenjieDu
refactor: add customized loss and metric funcs for SegRNN;
2f5840a
@WenjieDu
refactor: simplify some parts;
566b859
@WenjieDu
refactor: disable clustering algos to customize training loss and val…
30efa4e 
…idation metric;
@WenjieDu
feat: add loss and metric classes;
6f3b36e
@WenjieDu
refactor: use classes to replace funcs in CLAS algos;
8705db6
@WenjieDu
refactor: simplify some parts;
5b58313
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
507cc63
@WenjieDu
Merge branch 'dev' into (feat)add_timellm
b67a9d5
@WenjieDu
Merge branch '(feat)customized_loss_and_val_func' into (feat)add_timellm
47c566d
@WenjieDu
feat: add TimeLLM core modules;
8a7d8bd
@WenjieDu
feat: implement TimeLLM for imputation;
a9f6014
@WenjieDu
depen: add transformers as a dependency;
6260deb
@WenjieDu
test: add unit testing cases for TimeLLM;
0dcc1f0
@WenjieDu
test: ignore TimeLLM testing cases when running CI because of too lar…
5e112cf 
…ge model size;
@WenjieDu
docs: update docs to add TimeLLM;
b72582d
@WenjieDu
Merge branch 'dev' into (feat)customized_loss_and_val_func
9dbc1b1 
# Conflicts:
#	pypots/utils/metrics/error.py
@WenjieDu
refactor: import from pypots.nn.functional instead of pypots.utils.me…
6ca6da8 
…trics;
@WenjieDu
Merge pull request #526 from WenjieDu/(feat)customized_loss_and_val_func
5eddfab 
Enable to customized loss and val funcs
@WenjieDu
Merge branch 'dev' into (feat)add_timellm
8ea7032
@WenjieDu
refactor: simplify some parts;
50f4967
@WenjieDu
fix: typo in TimeLLM task assertion;
eaee9de
@WenjieDu
test: add annotation to TimeLLM initialization progress;
15be8ed
@WenjieDu
Merge pull request #567 from WenjieDu/(feat)add_timellm
95ba443 
Implement TimeLLM as an imputation model
@WenjieDu
test: ignore tests/*/llms;
f6747d6
@WenjieDu
Merge pull request #569 from WenjieDu/temp_test_branch
00dbb50 
Make pytest ignore LLM-based testing cases
@coveralls
Copy link
Collaborator

coveralls commented Feb 19, 2025 •
edited
Loading

Pull Request Test Coverage Report for Build 13413452867

Details

  • 502 of 810 (61.98%) changed or added relevant lines in 113 files are covered.
  • 18 unchanged lines in 14 files lost coverage.
  • Overall coverage decreased (-1.4%) to 83.002%
Changes Missing Coverage Covered Lines Changed/Added Lines %
pypots/classification/base.py 17 18 94.44%
pypots/clustering/base.py 0 1 0.0%
pypots/clustering/vader/model.py 2 3 66.67%
pypots/forecasting/csdi/model.py 5 6 83.33%
pypots/imputation/base.py 10 11 90.91%
pypots/imputation/csdi/model.py 5 6 83.33%
pypots/imputation/gpvae/model.py 3 4 75.0%
pypots/imputation/timellm/data.py 6 7 85.71%
pypots/nn/modules/patchtst/layers.py 5 6 83.33%
pypots/forecasting/base.py 1 3 33.33%
Files with Coverage Reduction New Missed Lines %
pypots/classification/base.py 1 78.23%
pypots/cli/env.py 1 78.38%
pypots/clustering/base.py 1 38.83%
pypots/clustering/vader/model.py 1 85.43%
pypots/forecasting/base.py 1 36.94%
pypots/forecasting/csdi/model.py 1 86.43%
pypots/imputation/base.py 1 78.63%
pypots/imputation/csdi/model.py 1 86.03%
pypots/imputation/gpvae/model.py 1 86.21%
pypots/imputation/itransformer/model.py 1 94.94%
Totals Coverage Status
Change from base Build 13409453716: -1.4%
Covered Lines: 12330
Relevant Lines: 14855
💛 - Coveralls

@WenjieDu WenjieDu merged commit 872a296 into main Feb 19, 2025
8 of 10 checks passed
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
@WenjieDu @coveralls