feat: added generate_requirements.sh

generate_requirements.sh

@@ -0,0 +1,21 @@
+# Usage: ./generate_requirements.sh <ENV-NAME>  (e.g. ./generate_requirements.sh dev-3.10)
+
+# set env variable
+# to allow pip-compile-cross-platform to use pip with --find-links.
+# not entirely sure this is needed.
+export PIP_FIND_LINKS=https://data.pyg.org/whl/torch-2.1.2+cpu.html
+
+# compile the dependencies from .in files
+pip-compile-cross-platform requirements/$1.in --min-python-version 3.10 -o requirements/$1.txt
+
+# remove the hashes from the .txt files
+# this is slightly less safe in terms of reproducibility
+# (e.g. if a package was re-uploaded to PyPI with the same version)
+# but it is necessary to allow `pip install -r requirements` to use --find-links
+# in our case, without --find-links, torch-cluster often cannot find the
+# proper wheels and throws out an error `no torch module` when trying to build
+sed -i '/--hash=/d' requirements/$1.txt
+sed -i 's/\\//g' requirements/$1.txt
+
+# removes the nvidia requirements
+sed -i '/nvidia/d' requirements/$1.txt