Python310 support (#241)

* updating README and docs [skip ci]

* chore: attempt 3.10 support

* ci: include 3.10 workflow

* ci: disable clone protection

* ci: mitigate mypy failure

* chore: start fixing mypy errors

.github/workflows/pypi.yaml

@@ -16,6 +16,7 @@ jobs:
       max-parallel: 3
       matrix:
         python-version:
+          - 3.10
           - 3.8
           - 3.7
         os:

.github/workflows/tests.yaml

@@ -19,6 +19,8 @@ jobs:
     defaults:
       run:
         shell: bash -l {0} # for conda command
+    env:
+      GIT_CLONE_PROTECTION_ACTIVE: false
     steps:
       - uses: actions/checkout@v2
       - uses: conda-incubator/setup-miniconda@v2
@@ -48,7 +50,7 @@ jobs:
       - name: Check mypy
         run: |
           conda activate gt4sd
-          python -m mypy src/gt4sd
+          python -m mypy src/gt4sd --show-traceback
       - name: Run pytests
         run: |
           conda activate gt4sd

conda_cpu_linux.yml

@@ -5,8 +5,9 @@ channels:
   - https://conda.anaconda.org/pyg
   - https://conda.anaconda.org/conda-forge
 dependencies:
-  - python>=3.7,<3.9
-  - pip>=19.1,<20.3
+  - setuptools==69.5.1
+  - python>=3.7,<3.11
+  - pip=24.0
   - pytorch>=1.0,<=1.12.1
   - cpuonly
   - pytorch-scatter<=2.0.9=*cu102*

conda_cpu_mac.yml

@@ -5,8 +5,9 @@ channels:
   - https://conda.anaconda.org/pyg
   - https://conda.anaconda.org/conda-forge
 dependencies:
-  - python>=3.7,<3.9
-  - pip>=19.1,<20.3
+  - setuptools==69.5.1
+  - python>=3.7,<3.11
+  - pip=24.0
   - pytorch>=1.0,<=1.12.1
   - cpuonly
   - pytorch-scatter<=2.0.9

conda_gpu.yml

@@ -5,8 +5,9 @@ channels:
   - https://conda.anaconda.org/pyg
   - https://conda.anaconda.org/conda-forge
 dependencies:
-  - python>=3.7,<3.9
-  - pip>=19.1,<20.3
+  - setuptools==69.5.1
+  - python>=3.7,<3.11
+  - pip=24.0
   - pytorch>=1.0,<=1.12.1=*cu*
   - pytorch-scatter<=2.0.9=*cu102*
   - torchvision<=0.13.1=*cu*

dev_requirements.txt

@@ -6,11 +6,11 @@ flask==1.1.2
 flask_login==0.5.0
 # isort==5.7.0
 licenseheaders==0.8.8
-mypy==0.950
-myst-parser==0.13.3
-pytest==6.1.1
+mypy>=1.0.0
+myst-parser==1.0.0
+pytest==6.2.5
 pytest-cov==2.10.1
-sphinx==3.4.3
+sphinx>=5
 sphinx-autodoc-typehints==1.11.1
 jinja2<3.1.0
 sphinx_rtd_theme==0.5.1

requirements.txt

@@ -1,4 +1,5 @@
 # pypi requirements
+setuptools==69.5.1
 accelerate>=0.12,<0.20.0
 datasets>=1.11.0
 diffusers<=0.6.0
@@ -11,7 +12,7 @@ gt4sd-molformer>=0.1.1
 gt4sd-trainer-hf-pl>=0.0.2
 keras>=2.3.1,<2.11.0
 keybert>=0.7.0
-markdown-it-py<3.0.0,>=2.2.0
+markdown-it-py
 minio==7.0.1
 modlamp>=4.0.0
 molgx>=0.22.0a1
@@ -25,6 +26,7 @@ PyTDC==0.3.7
 pytorch_lightning<=1.7.7
 pyyaml>=5.4.1
 rdkit>=2022.3.5
+rdkit-stubs>=0.7
 regex>=2.5.91
 reinvent-chemistry==0.0.38
 sacremoses>=0.0.41

setup.cfg

@@ -284,4 +284,7 @@ ignore_missing_imports = True
 ignore_missing_imports = True
 
 [mypy-xgboost.*]
-ignore_missing_imports = True
+ignore_missing_imports = True
+
+[mypy-rdkit-stubs.*]
+ignore_errors = True

src/gt4sd/algorithms/core.py

@@ -25,7 +25,7 @@
 
 from __future__ import annotations
 
-import collections
+from collections.abc import Hashable
 import logging
 import os
 import shutil
@@ -233,7 +233,7 @@ def sample(self, number_of_items: int = 100) -> Iterator[S]:
                 try:
                     valid_item = self.configuration.validate_item(item)
                     # check if sample is hashable
-                    if not isinstance(item, collections.Hashable):
+                    if not isinstance(item, Hashable):
                         yield valid_item
                         item_set.add(str(index))
                     else:
@@ -623,9 +623,11 @@ def save_version_from_training_pipeline_arguments(
                 target_version,
             )
             filepaths_mapping = {
-                filename: source_filepath
-                if os.path.exists(source_filepath)
-                else os.path.join(source_missing_path, filename)
+                filename: (
+                    source_filepath
+                    if os.path.exists(source_filepath)
+                    else os.path.join(source_missing_path, filename)
+                )
                 for filename, source_filepath in filepaths_mapping.items()
             }
             logger.info(f"Saving artifacts into {target_path}...")
@@ -713,9 +715,11 @@ def upload_version_from_training_pipeline_arguments(
 
             # mapping between filenames and paths for a version.
             filepaths_mapping = {
-                filename: source_filepath
-                if os.path.exists(source_filepath)
-                else os.path.join(source_missing_path, filename)
+                filename: (
+                    source_filepath
+                    if os.path.exists(source_filepath)
+                    else os.path.join(source_missing_path, filename)
+                )
                 for filename, source_filepath in filepaths_mapping.items()
             }
 

src/gt4sd/frameworks/cgcnn/data.py

@@ -31,7 +31,7 @@
 import logging
 import os
 import random
-from typing import Any, Callable, List, Tuple, Union
+from typing import Any, Callable, List, Tuple, Union, Optional
 
 import numpy as np
 import torch
@@ -49,7 +49,7 @@ def get_train_val_test_loader(
     dataset: torch.utils.data.Dataset,
     collate_fn: Callable[[List[Any]], Any] = default_collate,
     batch_size: int = 64,
-    train_ratio: float = None,
+    train_ratio: Optional[float] = None,
     val_ratio: float = 0.1,
     test_ratio: float = 0.1,
     return_test: bool = False,
@@ -241,9 +241,7 @@ def expand(self, distances: np.ndarray) -> np.ndarray:
               Expanded distance matrix with the last dimension of length
               len(self.filter).
         """
-        return np.exp(
-            -((distances[..., np.newaxis] - self.filter) ** 2) / self.var**2
-        )
+        return np.exp(-((distances[..., np.newaxis] - self.filter) ** 2) / self.var**2)
 
 
 class AtomInitializer: