Update RTD configuration, and linear_operator requirement.

RTD is removing the "use system packages" feature on 29 Aug 2023.
This PR ensures that our docs will sill build.

Moreover, the linear_operator requirement needs to be updated for #2288.

.readthedocs.yml

@@ -10,8 +10,10 @@ build:
   tools:
     python: "3.8"
   jobs:
+    pre_install:  # Lock version of torch at 1.11
+      - pip install torch==1.11.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
     pre_build:
-      - python -m setuptools_scm
+      - python -m setuptools_scm  # Get correct version number
 
 # Build documentation in the docs/ directory with Sphinx
 sphinx:
@@ -25,4 +27,8 @@ sphinx:
 # Optionally set the version of Python and requirements required to build your docs
 python:
   install:
-    - requirements: docs/requirements.txt
+    - method: pip
+      path: .
+      extra_requirements:
+        - docs
+        - pyro

CONTRIBUTING.md

@@ -9,8 +9,7 @@ linting, testing, and building the documentation, run the following:
 ```bash
 git clone https://github.com/cornellius-gp/gpytorch.git
 cd gpytorch
-pip install -e .[dev,examples,pyro,keops,test]  # keops and pyro are optional
-pip install -r docs/requirements.txt
+pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
 pre-commit install
 ```
 

README.md

@@ -57,14 +57,7 @@ If you are contributing a pull request, it is best to perform a manual installat
 ```sh
 git clone https://github.com/cornellius-gp/gpytorch.git
 cd gpytorch
-pip install -e .[dev,examples,test,pyro,keops]
-```
-
-To generate the documentation locally, you will also need to run the following command
-from the linear_operator folder:
-
-```sh
-pip install -r docs/requirements.txt
+pip install -e .[dev,docs,examples,keops,pyro,test]  # keops and pyro are optional
 ```
 
 #### ArchLinux Package

setup.py

@@ -40,7 +40,8 @@ def find_version(*file_paths):
 torch_min = "1.11"
 install_requires = [
     "scikit-learn",
-    "linear_operator>=0.5.0",
+    "scipy",
+    "linear_operator>=0.5.2",
 ]
 # if recent dev version of PyTorch is installed, no need to install stable
 try:
@@ -72,10 +73,23 @@ def find_version(*file_paths):
     python_requires=">=3.8",
     install_requires=install_requires,
     extras_require={
-        "dev": ["ufmt", "twine", "pre-commit"],
+        "dev": ["pre-commit", "setuptools_scm", "twine", "ufmt"],
+        "docs": [
+            "ipykernel<=6.17.1",
+            "ipython<=8.6.0",
+            "m2r2<=0.3.3.post2",
+            "nbclient<=0.7.3",
+            "nbformat<=5.8.0",
+            "nbsphinx<=0.9.1",
+            "platformdirs<=3.2.0",
+            "setuptools_scm<=7.1.0",
+            "sphinx<=6.2.1",
+            "sphinx_autodoc_typehints<=1.23.0",
+            "sphinx_rtd_theme<0.5",
+        ],
         "examples": ["ipython", "jupyter", "matplotlib", "scipy", "torchvision", "tqdm"],
-        "pyro": ["pyro-ppl>=1.8"],
         "keops": ["pykeops>=1.1.1"],
+        "pyro": ["pyro-ppl>=1.8"],
         "test": ["flake8==4.0.1", "flake8-print==4.0.0", "pytest", "nbval"],
     },
     test_suite="test",