Improve packaging

-  Improve some packaging stuff
- Remove versioneer in favour of versiongit
- removed old setup.py
- setup conda environments
- Added install instructions.

.gitattributes

@@ -1 +1,2 @@
-mtenn/_version.py export-subst
+*/_version_git.py export-subst
+

.github/workflows/CI.yaml

@@ -23,12 +23,13 @@ jobs:
     name: Test on ${{ matrix.os }}, Python ${{ matrix.python-version }}
     runs-on: ${{ matrix.os }}
     strategy:
+      fail-fast: false
       matrix:
-        os: [macOS-latest, ubuntu-latest, windows-latest]
-        python-version: [3.7, 3.8, 3.9]
+        os: [macOS-latest, ubuntu-latest]
+        python-version: [3.8, 3.9, "3.10"]
 
     steps:
-    - uses: actions/checkout@v1
+    - uses: actions/checkout@v2
 
     - name: Additional info about the build
       shell: bash
@@ -39,18 +40,17 @@ jobs:
 
 
     # More info on options: https://github.com/conda-incubator/setup-miniconda
-    - uses: conda-incubator/setup-miniconda@v2
+    - name: Setup Conda
+      uses: conda-incubator/setup-miniconda@v2
       with:
         python-version: ${{ matrix.python-version }}
         environment-file: devtools/conda-envs/test_env.yaml
 
-        channels: conda-forge,defaults
-
         activate-environment: test
         auto-update-conda: false
         auto-activate-base: false
         show-channel-urls: true
-
+    
     - name: Install package
 
       # conda setup requires this special shell

README.md

@@ -83,6 +83,41 @@ print(f'Using delta Model: {delta_e_new.item():0.5f}')
 print(f'Using concat Model: {concat_e.item():0.5f}')
 ```
 
+### Installation
+
+Installation of pytorch is required. We provide a minimal environment file to install pytorch and conda dependencies for conda users in ```environment.yml```. You should create a new environment
+like so:
+
+```bash
+conda env create --file environment.yml
+conda activate mtenn
+```
+
+Training and inference is often faster using the GPU version of pytorch.
+We provide a minimal environment file to install GPU versions of pytorch and conda dependencies for conda users in ```environment-gpu.yml```. You should create a new environment like so:
+
+ ```bash
+conda env create --file environment-gpu.yml
+conda activate mtenn-gpu
+```
+
+If not using conda an installation of pytorch **MUST** be done before installing the package itself. See the pytorch documentation on how to best install pytorch for your system.
+
+ To install mtenn and its dependencies (excluding pytorch), run
+```bash
+pip install -e .
+```
+
+If not using conda some of the `mtenn` dependencies do not come with pre-built wheels for all platforms, so pip may need to build them from source. This requires a C++ compiler and may take a while.
+For advanced users you can directly install the dependency packages listed in the `requirements.txt` using manually specified wheels you can find on `https://data.pyg.org/whl/`.
+
+If compatibility is proving difficult you may need to purge your pip cache
+
+```bash
+pip cache purge
+```
+
+
 #### Acknowledgements
 
 Project based on the

devtools/conda-envs/test_env.yaml

@@ -1,20 +1,17 @@
 name: test
 channels:
-
   - conda-forge
-
-  - defaults
 dependencies:
-    # Base depends
-  - python
-  - pip
-
-    # Testing
+  - pytorch
+  - pytorch_geometric
+  - pytorch_cluster
+  - pytorch_scatter
+  - pytorch_sparse
+  - numpy
+  - h5py
+  - e3nn
+  # testing dependencies
   - pytest
   - pytest-cov
   - codecov
 
-    # Pip-only installs
-  #- pip:
-  #  - codecov
-

environment-gpu.yml

@@ -0,0 +1,13 @@
+name: mtenn-gpu
+channels:
+  - conda-forge
+dependencies:
+  - pytorch
+  - pytorch-gpu
+  - pytorch_geometric
+  - pytorch_cluster
+  - pytorch_scatter
+  - pytorch_sparse
+  - numpy
+  - h5py
+  - e3nn

environment.yml

@@ -0,0 +1,12 @@
+name: mtenn
+channels:
+  - conda-forge
+dependencies:
+  - pytorch
+  - pytorch_geometric
+  - pytorch_cluster
+  - pytorch_scatter
+  - pytorch_sparse
+  - numpy
+  - h5py
+  - e3nn

mtenn/__init__.py

@@ -4,10 +4,4 @@
 from .mtenn import *
 from . import conversion_utils
 
-# Handle versioneer
-from ._version import get_versions
-
-versions = get_versions()
-__version__ = versions["version"]
-__git_revision__ = versions["full-revisionid"]
-del get_versions, versions
+from ._version_git import __version__