Add models US-GAN and GP-VAE, update docs, refactor testing cases, add cal_internal_cluster_validation_metrics()

.github/workflows/testing_ci.yml

@@ -15,43 +15,61 @@ jobs:
         runs-on: ${{ matrix.os }}
         defaults:
             run:
-                shell: bash -l {0}
+                shell: bash {0}
         strategy:
             fail-fast: false
             matrix:
                 os: [ubuntu-latest, windows-latest, macOS-latest]
-                python-version: ["3.7", "3.8", "3.9", "3.10"]
+                python-version: ["3.7", "3.10"]
+                torch-version: ["1.13.1"]
 
         steps:
             - name: Check out the repo code
               uses: actions/checkout@v3
 
-            - name: Set up Conda
-              uses: conda-incubator/setup-miniconda@v2
+            - name: Determine the Python version
+              uses: haya14busa/action-cond@v1
+              id: condval
               with:
-                  activate-environment: pypots-test
-                  python-version: ${{ matrix.python-version }}
-                  environment-file: tests/environment_for_conda_test.yml
-                  auto-activate-base: false
+                cond: ${{ matrix.python-version == 3.7 && matrix.os == 'macOS-latest' }}
+                # Note: the latest 3.7 subversion 3.7.17 for MacOS has "ModuleNotFoundError: No module named '_bz2'"
+                if_true: "3.7.16"
+                if_false: ${{ matrix.python-version }}
+
+            - name: Set up Python
+              uses: actions/setup-python@v4
+              with:
+                python-version: ${{ steps.condval.outputs.value }}
+                check-latest: true
+                cache: pip
+                cache-dependency-path: |
+                  setup.cfg
+
+            - name: Install PyTorch ${{ matrix.torch-version }}+cpu
+              # we have to install torch in advance because torch_sparse needs it for compilation,
+              # refer to https://github.com/rusty1s/pytorch_sparse/issues/156#issuecomment-1304869772 for details
+              run: |
+                  which python
+                  which pip
+                  python -m pip install --upgrade pip
+                  pip install torch==${{ matrix.torch-version }} -f https://download.pytorch.org/whl/cpu
+                  python -c "import torch; print('PyTorch:', torch.__version__)"
+
+            - name: Install other dependencies
+              run: |
+                  pip install pypots
+                  pip install torch-geometric torch-scatter torch-sparse -f "https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html"
+                  pip install -e ".[dev]"
 
             - name: Fetch the test environment details
               run: |
                   which python
-                  conda info
-                  conda list
+                  pip list
 
             - name: Test with pytest
               run: |
-                  # run tests separately here due to Segmentation Fault in test_clustering when run all in
-                  # one command with `pytest` on MacOS. Bugs not caught, so this is a trade-off to avoid SF.
-                  python -m pytest -rA tests/test_classification.py -n auto --cov=pypots --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_imputation.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_clustering.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_forecasting.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_optim.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_data.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_utils.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
-                  python -m pytest -rA tests/test_cli.py -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  rm -rf tests/__pycache__
+                  python -m pytest -rA tests/*/* -n auto --cov=pypots --dist=loadgroup --cov-config=.coveragerc
 
             - name: Generate the LCOV report
               run: |
@@ -61,4 +79,4 @@ jobs:
               uses: coverallsapp/github-action@master
               with:
                   github-token: ${{ secrets.GITHUB_TOKEN }}
-                  path-to-lcov: 'coverage.lcov'
+                  path-to-lcov: "coverage.lcov"

.github/workflows/testing_daily.yml

@@ -10,61 +10,43 @@ jobs:
         runs-on: ${{ matrix.os }}
         defaults:
             run:
-                shell: bash {0}
+                shell: bash -l {0}
         strategy:
             fail-fast: false
             matrix:
                 os: [ubuntu-latest, windows-latest, macOS-latest]
-                python-version: ["3.7", "3.8", "3.9", "3.10"]
-                torch-version: ["1.13.1"]
+                python-version: ["3.7", "3.10"]
 
         steps:
             - name: Check out the repo code
               uses: actions/checkout@v3
 
-            - name: Determine the Python version
-              uses: haya14busa/action-cond@v1
-              id: condval
+            - name: Set up Conda
+              uses: conda-incubator/setup-miniconda@v2
               with:
-                cond: ${{ matrix.python-version == 3.7 && matrix.os == 'macOS-latest' }}
-                # Note: the latest 3.7 subversion 3.7.17 for MacOS has "ModuleNotFoundError: No module named '_bz2'"
-                if_true: "3.7.16"
-                if_false: ${{ matrix.python-version }}
-
-            - name: Set up Python
-              uses: actions/setup-python@v4
-              with:
-                python-version: ${{ steps.condval.outputs.value }}
-                check-latest: true
-                cache: pip
-                cache-dependency-path: |
-                  setup.cfg
-
-            - name: Install PyTorch ${{ matrix.torch-version }}+cpu
-              # we have to install torch in advance because torch_sparse needs it for compilation,
-              # refer to https://github.com/rusty1s/pytorch_sparse/issues/156#issuecomment-1304869772 for details
-              run: |
-                  which python
-                  which pip
-                  python -m pip install --upgrade pip
-                  pip install torch==${{ matrix.torch-version }} -f https://download.pytorch.org/whl/cpu
-                  python -c "import torch; print('PyTorch:', torch.__version__)"
-
-            - name: Install other dependencies
-              run: |
-                  pip install pypots
-                  pip install torch-geometric torch-scatter torch-sparse -f "https://data.pyg.org/whl/torch-${{ matrix.torch-version }}+cpu.html"
-                  pip install -e ".[dev]"
+                  activate-environment: pypots-test
+                  python-version: ${{ matrix.python-version }}
+                  environment-file: tests/environment_for_conda_test.yml
+                  auto-activate-base: false
 
             - name: Fetch the test environment details
               run: |
                   which python
-                  pip list
+                  conda info
+                  conda list
 
             - name: Test with pytest
               run: |
-                  coverage run --source=pypots -m pytest --ignore tests/test_training_on_multi_gpus.py
-                  # ignore the test_training_on_multi_gpus.py because it requires multiple GPUs which are not available on GitHub Actions
+                  # run tests separately here due to Segmentation Fault in test_clustering when run all in
+                  # one command with `pytest` on MacOS. Bugs not caught, so this is a trade-off to avoid SF.
+                  python -m pytest -rA tests/classification/* -n auto --cov=pypots --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/imputation/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/clustering/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/forecasting/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/optim/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/data/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/utils/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
+                  python -m pytest -rA tests/cli/* -n auto --cov=pypots --cov-append --dist=loadgroup --cov-config=.coveragerc
 
             - name: Generate the LCOV report
               run: |
@@ -74,4 +56,4 @@ jobs:
               uses: coverallsapp/github-action@master
               with:
                   github-token: ${{ secrets.GITHUB_TOKEN }}
-                  path-to-lcov: "coverage.lcov"
+                  path-to-lcov: 'coverage.lcov'

.gitignore

@@ -14,7 +14,8 @@ docs/_build
 .coverage
 .pytest_cache
 *__pycache__*
-*testing_results*
+*test*
 
 # ignore specific kinds of files like all PDFs
 *.pdf
+*.ipynb

README.md

@@ -2,7 +2,8 @@
     <img src="https://pypots.com/figs/pypots_logos/PyPOTS_logo_FFBG.svg?sanitize=true" width="200" align="right">
 </a>
 
-## <p align="center">Welcome to PyPOTS</p>
+<h2 align="center">Welcome to PyPOTS</h2>
+
 **<p align="center">A Python Toolbox for Data Mining on Partially-Observed Time Series</p>**
 
 <p align="center">
@@ -161,6 +162,8 @@ PyPOTS supports imputation, classification, clustering, and forecasting tasks on
 |        **Type**        |  **Abbr.**  |                                                                    **Full name of the algorithm/model/paper**                                                                     | **Year** |
 |       Neural Net       |    SAITS    |                                                               Self-Attention-based Imputation for Time Series [^1]                                                                |   2023   |
 |       Neural Net       | Transformer | Attention is All you Need [^2];<br>Self-Attention-based Imputation for Time Series [^1];<br><sub>Note: proposed in [^2], and re-implemented as an imputation model in [^1].</sub> |   2017   |
+|       Neural Net       |   US-GAN    |                                                 Generative Semi-supervised Learning for Multivariate Time Series Imputation [^10]                                                 |   2021   |
+|       Neural Net       |   GP-VAE    |                                                              GP-VAE: Deep Probabilistic Time Series Imputation [^11]                                                              |   2020   |
 |       Neural Net       |    BRITS    |                                                              Bidirectional Recurrent Imputation for Time Series [^3]                                                              |   2018   |
 |       Neural Net       |    M-RNN    |                                                                  Multi-directional Recurrent Neural Network [^9]                                                                  |   2019   |
 |         Naive          |    LOCF     |                                                                         Last Observation Carried Forward                                                                          |    -     |
@@ -253,7 +256,7 @@ We care about the feedback from our users, so we're building PyPOTS community on
 If you have any suggestions or want to contribute ideas or share time-series related papers, join us and tell.
 PyPOTS community is open, transparent, and surely friendly. Let's work together to build and improve PyPOTS!
 
-
+[//]: # (Use APA reference style below)
 [^1]: Du, W., Cote, D., & Liu, Y. (2023). [SAITS: Self-Attention-based Imputation for Time Series](https://doi.org/10.1016/j.eswa.2023.119619). *Expert systems with applications*.
 [^2]: Vaswani, A., Shazeer, N.M., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., & Polosukhin, I. (2017). [Attention is All you Need](https://papers.nips.cc/paper/2017/hash/3f5ee243547dee91fbd053c1c4a845aa-Abstract.html). *NeurIPS 2017*.
 [^3]: Cao, W., Wang, D., Li, J., Zhou, H., Li, L., & Li, Y. (2018). [BRITS: Bidirectional Recurrent Imputation for Time Series](https://papers.nips.cc/paper/2018/hash/734e6bfcd358e25ac1db0a4241b95651-Abstract.html). *NeurIPS 2018*.
@@ -263,12 +266,13 @@ PyPOTS community is open, transparent, and surely friendly. Let's work together
 [^7]: Jong, J.D., Emon, M.A., Wu, P., Karki, R., Sood, M., Godard, P., Ahmad, A., Vrooman, H.A., Hofmann-Apitius, M., & Fröhlich, H. (2019). [Deep learning for clustering of multivariate clinical patient trajectories with missing values](https://academic.oup.com/gigascience/article/8/11/giz134/5626377). *GigaScience*.
 [^8]: Chen, X., & Sun, L. (2021). [Bayesian Temporal Factorization for Multidimensional Time Series Prediction](https://arxiv.org/abs/1910.06366). *IEEE transactions on pattern analysis and machine intelligence*.
 [^9]: Yoon, J., Zame, W. R., & van der Schaar, M. (2019). [Estimating Missing Data in Temporal Data Streams Using Multi-Directional Recurrent Neural Networks](https://ieeexplore.ieee.org/document/8485748). *IEEE Transactions on Biomedical Engineering*.
-
+[^10]: Miao, X., Wu, Y., Wang, J., Gao, Y., Mao, X., & Yin, J. (2021). [Generative Semi-supervised Learning for Multivariate Time Series Imputation](https://ojs.aaai.org/index.php/AAAI/article/view/17086). *AAAI 2021*.
+[^11]: Fortuin, V., Baranchuk, D., Raetsch, G. & Mandt, S.. (2020). [GP-VAE: Deep Probabilistic Time Series Imputation](https://proceedings.mlr.press/v108/fortuin20a.html). *AISTATS 2020*.
 
 <details>
 <summary>🏠 Visits</summary>
 <a href="https://github.com/WenjieDu/PyPOTS">
     <img alt="PyPOTS visits" align="left" src="https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Fgithub.com%2FPyPOTS%2FPyPOTS&count_bg=%23009A0A&title_bg=%23555555&icon=&icon_color=%23E7E7E7&title=Visits%20since%20May%202022&edge_flat=false">
 </a>
 </details>
-<br>
+<br>

docs/about_us.rst

@@ -33,5 +33,5 @@ PyPOTS exists thanks to all the nice people (sorted by contribution time) who co
 
 .. raw:: html
 
-    <object data="https://pypots.com/figs/PyPOTS_contributors.svg">
+    <object data="https://pypots.com/figs/pypots_logos/PyPOTS_contributors.svg">
     </object>

docs/pypots.data.rst

@@ -10,6 +10,15 @@ pypots.data.base module
    :show-inheritance:
    :inherited-members:
 
+pypots.data.saving module
+-----------------------------
+
+.. automodule:: pypots.data.saving
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :inherited-members:
+
 pypots.data.generating module
 -----------------------------
 

docs/pypots.forecasting.rst

@@ -1,11 +1,31 @@
 pypots.forecasting package
 ==========================
 
+Subpackages
+-----------
 
-pypots.forecasting.bttf module
+.. toctree::
+   :maxdepth: 4
+
+   pypots.forecasting.bttf
+   pypots.forecasting.template
+
+Submodules
+----------
+
+pypots.forecasting.base module
 ------------------------------
 
-.. automodule:: pypots.forecasting.bttf
+.. automodule:: pypots.forecasting.base
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :inherited-members:
+
+Module contents
+---------------
+
+.. automodule:: pypots.forecasting
    :members:
    :undoc-members:
    :show-inheritance:

docs/pypots.imputation.rst

@@ -19,6 +19,24 @@ pypots.imputation.transformer module
    :show-inheritance:
    :inherited-members:
 
+pypots.imputation.usgan module
+------------------------------
+
+.. automodule:: pypots.imputation.usgan
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :inherited-members:
+
+pypots.imputation.gpvae module
+------------------------------
+
+.. automodule:: pypots.imputation.gpvae
+   :members:
+   :undoc-members:
+   :show-inheritance:
+   :inherited-members:
+
 pypots.imputation.brits module
 ------------------------------
 

pypots/base.py

@@ -96,7 +96,9 @@ def _setup_device(self, device: Union[None, str, torch.device, list]):
                 self.device = device
             elif isinstance(device, list):
                 if len(device) == 0:
-                    raise ValueError("The list of devices should have at least 1 device, but got 0.")
+                    raise ValueError(
+                        "The list of devices should have at least 1 device, but got 0."
+                    )
                 elif len(device) == 1:
                     return self._setup_device(device[0])
                 # parallely training on multiple CUDA devices
@@ -176,7 +178,6 @@ def _send_data_to_given_device(self, data):
         if isinstance(self.device, torch.device):  # single device
             data = map(lambda x: x.to(self.device), data)
         else:  # parallely training on multiple devices
-
             # randomly choose one device to balance the workload
             # device = np.random.choice(self.device)
 

pypots/classification/base.py

@@ -256,7 +256,6 @@ def _train_model(
         training_loader: DataLoader,
         val_loader: DataLoader = None,
     ) -> None:
-
         # each training starts from the very beginning, so reset the loss and model dict here
         self.best_loss = float("inf")
         self.best_model_dict = None

pypots/classification/grud/data.py

@@ -123,7 +123,7 @@ def _fetch_data_from_file(self, idx: int) -> Iterable:
         if self.file_handle is None:
             self.file_handle = self._open_file_handle()
 
-        X = torch.from_numpy(self.file_handle["X"][idx])
+        X = torch.from_numpy(self.file_handle["X"][idx]).to(torch.float32)
         missing_mask = (~torch.isnan(X)).to(torch.float32)
         X_filledLOCF = self.locf._locf_torch(X.unsqueeze(dim=0)).squeeze()
         X = torch.nan_to_num(X)

pypots/classification/raindrop/modules.py

@@ -174,7 +174,6 @@ def forward(
         edge_attr: OptTensor = None,
         return_attention_weights=None,
     ) -> Tuple[torch.Tensor, Any]:
-
         r"""
         Args:
             return_attention_weights (bool, optional): If set to :obj:`True`,

pypots/clustering/base.py

@@ -244,7 +244,6 @@ def _train_model(
         training_loader: DataLoader,
         val_loader: DataLoader = None,
     ) -> None:
-
         """
 
         Parameters

pypots/clustering/crli/model.py

@@ -226,7 +226,6 @@ def __init__(
         saving_path: Optional[str] = None,
         model_saving_strategy: Optional[str] = "best",
     ):
-
         super().__init__(
             n_clusters,
             batch_size,

pypots/clustering/vader/data.py

@@ -6,12 +6,12 @@
 # License: GLP-v3
 
 
-from typing import Union
+from typing import Union, Iterable
 
-from ..crli.data import DatasetForCRLI
+from ...data.base import BaseDataset
 
 
-class DatasetForVaDER(DatasetForCRLI):
+class DatasetForVaDER(BaseDataset):
     """Dataset class for model VaDER.
 
     Parameters
@@ -45,3 +45,9 @@ def __init__(
         file_type: str = "h5py",
     ):
         super().__init__(data, return_labels, file_type)
+
+    def _fetch_data_from_array(self, idx: int) -> Iterable:
+        return super()._fetch_data_from_array(idx)
+
+    def _fetch_data_from_file(self, idx: int) -> Iterable:
+        return super()._fetch_data_from_file(idx)