Merge branch 'master' into plot_black

.github/workflows/ci.yml

@@ -32,7 +32,7 @@ jobs:
     - name: test if api works
       run: poetry run python examples/api/jigen_dann_transformer.py
     - name: Generate coverage report
-      run: rm -r zoutput && poetry run pytest --cov=domainlab tests/ --cov-report=xml
+      run: rm -rf zoutput && poetry run pytest --maxfail=1 -vvv --tb=short --cov=domainlab tests/ --cov-report=xml
     - name: Upload coverage to Codecov
       uses: codecov/codecov-action@v1
       with:
@@ -42,8 +42,14 @@ jobs:
       env:
         CODECOV_TOKEN: 02ecb3ac-b7ce-4ea4-90a2-961c3d1a1030
     - name: check if readme yaml works
-      run: rm -r zoutput && python main_out.py -c ./examples/conf/vlcs_diva_mldg_dial.yaml
+      run: rm -rf zoutput && python main_out.py -c ./examples/conf/vlcs_diva_mldg_dial.yaml
     - name: test if examples in markdown works
-      run: bash -x -v ci_run_examples.sh
+      run: bash -x -v scripts/ci_run_examples.sh
     - name: test if benchmark works
-      run: pip install snakemake==7.32.0 && pip install pulp==2.7.0 && sed -i '1s/^/#!\/bin\/bash -x -v\n/' run_benchmark_standalone.sh && bash -x -v run_benchmark_standalone.sh examples/benchmark/demo_shared_hyper_grid.yaml && cat zoutput/benchmarks/mnist_benchmark_grid/hyperparameters.csv && cat zoutput/benchmarks/mnist_benchmark_grid/results.csv
+      run: | 
+          pip install snakemake==7.32.0 && pip install pulp==2.7.0
+          echo "insert a shebang line (#!/bin/bash -x -v) at the beginning of the bash script"
+          sed -i '1s/^/#!\/bin\/bash -x -v\n/' run_benchmark_standalone.sh
+          bash -x -v run_benchmark_standalone.sh examples/benchmark/demo_shared_hyper_grid.yaml
+          cat zoutput/benchmarks/mnist_benchmark_grid*/hyperparameters.csv
+          cat zoutput/benchmarks/mnist_benchmark_grid*/results.csv

.gitignore

@@ -1,7 +1,13 @@
 .ropeproject
-./zdpath
-./zoutput
+/zdpath
+/zoutput
 tests/__pycache__/
 *.pyc
 .vscode/
-data/pacs
+domainlab/zdata/pacs
+/data/
+/.snakemake/
+/dist 
+/domainlab.egg-info
+/runs
+/slurm_errors.txt

README.md

@@ -8,24 +8,37 @@
 
 ## Distribution shifts, domain generalization and DomainLab
 
-Neural networks trained using data from a specific distribution (domain) usually fails to generalize to novel distributions (domains). Domain generalization aims at learning domain invariant features by utilizing data from multiple domains (data sites, corhorts, batches, vendors) so the learned feature can generalize to new unseen domains (distributions).
+Neural networks trained using data from a specific distribution (domain) usually fail to generalize to novel distributions (domains). Domain generalization aims at learning domain invariant features by utilizing data from multiple domains (data sites, cohorts, batches, vendors) so the learned feature can be generalized to new unseen domains (distributions).
 
-DomainLab is a software platform with state-of-the-art domain generalization algorithms implemented, designed by maximal decoupling of different software components thus enhances maximal code reuse.
+DomainLab is a software platform with state-of-the-art domain generalization algorithms implemented and designed by maximal decoupling of different software components thus enhancing maximal code reuse.
 
 ### DomainLab
 DomainLab decouples the following concepts or objects:
-- task $M$: In DomainLab, a task is a container for datasets from different domains. (e.g. from distribution $D_1$ and $D_2$). Task offer a static protocol to evaluate the generalization performance of a neural network: which dataset(s) is used for training, wich dataset(s) used for testing.  
-- neural network: a map $\phi$ from the input data to the feature space and a map $\varphi$ from feature space to output $\hat{y}$ (e.g. decision variable).
+- task $M$: In DomainLab, a task is a container for datasets from different domains. (e.g. from distribution $D_1$ and $D_2$). The task offers a static protocol to evaluate the generalization performance of a neural network: which dataset(s) is used for training, and which dataset(s) is used for testing.  
+- neural network: a map $\phi$ from the input data to the feature space and a map $\varphi$ from the feature space to output $\hat{y}$ (e.g. decision variable).
 - model: structural risk in the form of $\ell() + \mu R()$  where
-  - $\ell(Y, \hat{y}=\varphi(\phi(X)))$ is the task specific empirical loss (e.g. cross entropy for classification task).
+  - $\ell(Y, \hat{y}=\varphi(\phi(X)))$ is the task-specific empirical loss (e.g. cross entropy for classification task).
   - $R(\phi(X))$ is the penalty loss to boost domain invariant feature extraction using $\phi$.
   - $\mu$ is the corresponding multiplier to each penalty function factor.
 - trainer:  an object that guides the data flow to model and append further domain invariant losses
 like inter-domain feature alignment.
 
-We offer detailed documentation on how these models and trainers work in our documentation page: https://marrlab.github.io/DomainLab/
+We offer detailed documentation on how these models and trainers work on our documentation page: https://marrlab.github.io/DomainLab/
 
-DomainLab makes it possible to combine models with models, trainers with models, and trainers with trainers in a decorator pattern like line of code `Trainer A(Trainer B(Model C(Model D(network E), network E, network F)))` which correspond to $\ell() + \mu_a R_a() + \mu_b R_b + \mu_c R_c() + \mu_d R_d()$, where Model C and Model D share neural network E, but Model C has an extra neural network F. All models share the same neural network for feature extraction, but can have different auxilliary networks for $R()$.
+DomainLab makes it possible to combine models with models, trainers with models, and trainers with trainers in a decorator pattern like the line of code below
+
+```
+Trainer A(
+		  Trainer B(Model C(
+						    Model D(network E),
+							network E,
+							network F
+						   )
+				   )
+	     )
+```
+
+which correspond to $\ell() + \mu_a R_a() + \mu_b R_b + \mu_c R_c() + \mu_d R_d()$, where Model C and Model D share neural network E, but Model C has an extra neural network F. All models share the same neural network for feature extraction, but can have different auxiliary networks for $R()$.
 
 <div style="align: center; text-align:center;">
 <figure>  
@@ -36,7 +49,7 @@ DomainLab makes it possible to combine models with models, trainers with models,
 ## Getting started
 
 ### Installation
-For development version in Github, see [Installation and Dependencies handling](./docs/doc_install.md)
+For the development version in Github, see [Installation and Dependencies handling](./docs/doc_install.md)
 
 We also offer a PyPI version here https://pypi.org/project/domainlab/  which one could install via `pip install domainlab` and it is recommended to create a virtual environment for it.
 
@@ -45,10 +58,28 @@ We offer various ways for the user to specify a scenario to evaluate the general
 [Task Specification](./docs/doc_tasks.md) 
 
 ### Example and usage
+#### Available arguments for commandline
 
-#### Either clone this repo and use command line
+The following command tells which arguments/hyperparameters/multipliers are available to be set by the user and which model they are associated with
+
+```shell
+python main_out.py --help
+```
+
+or 
+
+```
+domainlab --help
+```
+
+#### Command line configuration file
+
+`domainlab -c ./examples/conf/vlcs_diva_mldg_dial.yaml` (if you install via pip)
+
+or if you clone this the code repository for DomainLab
 
 `python main_out.py -c ./examples/conf/vlcs_diva_mldg_dial.yaml`
+
 where the configuration file below can be downloaded [here](https://raw.githubusercontent.com/marrlab/DomainLab/master/examples/conf/vlcs_diva_mldg_dial.yaml)
 ```
 te_d: caltech                       # domain name of test domain
@@ -77,7 +108,7 @@ One could simply run
 `bash run_benchmark_slurm.sh your_benchmark_configuration.yaml` to launch different experiments with specified configuraiton.
 
 
-For example,  the following result (without any augmentation like flip) is for PACS dataset using ResNet. 
+For example,  the following result (without any augmentation like flip) is for PACS dataset using ResNet. The reader should note that using different neural network, whether pre-trained or not, what kind of preprocessinga and augmentation to use can lead to very different result distributions, which is one of the features DomainLab provide: the above factors get decoupled in DomainLab. 
 
 <div style="align: center; text-align:center;">
 <figure>  
@@ -89,15 +120,15 @@ For example,  the following result (without any augmentation like flip) is for P
 </div>
 
 
-### Temporary citation
+### Citation
+
+Source: https://arxiv.org/pdf/2403.14356.pdf
 
 ```bibtex
-@manual{domainlab,
-  title={{DomainLab: modular python package for training domain invariant neural networks}},
-  author={{Xudong Sun, et.al.}},
-  organization={{Institute of AI for Health}},
-  year={2023},
-  url={https://github.com/marrlab/DomainLab},
-  note={temporary citation for domainlab}
+@misc{sun2024domainlab,
+  title={DomainLab: A modular Python package for domain generalization in deep learning},
+  author={Sun, Xudong and Feistner, Carla and Gossmann, Alexej and Schwarz, George and Umer, Rao Muhammad and Beer, Lisa and Rockenschaub, Patrick and Shrestha, Rahul Babu and Gruber, Armin and Chen, Nutan and others},
+  journal={https://arxiv.org/pdf/2403.14356.pdf},
+  year={2024}
 }
 ```