Add resource estimates requirements as extras_require.

setup.py

@@ -12,17 +12,14 @@
 import io
 import os
 
-from setuptools import setup, find_packages
-
+from setuptools import find_packages, setup
 
 # This reads the __version__ variable from openfermion/_version.py
 __version__ = ''
 exec(open('src/openfermion/_version.py').read())
 
 # Readme file as long_description:
-long_description = ('===========\n' +
-                    'OpenFermion\n' +
-                    '===========\n')
+long_description = '===========\n' + 'OpenFermion\n' + '===========\n'
 stream = io.open('README.rst', encoding='utf-8')
 stream.readline()
 long_description += stream.read()
@@ -31,15 +28,19 @@
 requirements = open('dev_tools/requirements/deps/runtime.txt').readlines()
 requirements = [r.strip() for r in requirements]
 requirements = [r for r in requirements if not r.startswith('#')]
+# Resource estimates requirements.
+resource_requirements = open('dev_tools/requirements/deps/resource_estimates.txt').readlines()
+resource_requirements = [r.strip() for r in resource_requirements]
+resource_requirements = [r for r in resource_requirements if not r.startswith('#')]
 
 docs_files_gen = os.walk('docs')
 docs_data_files_tuples = []
 for cwd, subdirs, files in list(docs_files_gen)[1:]:
     if 'ipynb_checkpoints' in cwd:
         continue
     docs_data_files_tuples.append(
-        (os.path.join('openfermion',
-                      cwd), [os.path.join(cwd, file) for file in files]))
+        (os.path.join('openfermion', cwd), [os.path.join(cwd, file) for file in files])
+    )
 
 setup(
     name='openfermion',
@@ -58,7 +59,8 @@
         '': [
             os.path.join('src', 'openfermion', 'testing', '*.npy'),
             os.path.join('src', 'openfermion', 'testing', '*.hdf5'),
-        ],
+        ]
     },
+    extras_require={'resources': resource_requirements},
     data_files=docs_data_files_tuples,
 )

src/openfermion/resource_estimates/README.md

@@ -2,13 +2,10 @@
 
 Code is system tested on Debian GNU/Linux with Python 3.8.5. All the code comes with tests (use `pytest`), but not unit tested with GitHub worflows.
 
-Since the FT costing is closely tied to manipulation of molecular integrals (localization, active space selection, benchmarking against CCSD(T), ...) the code depends on [PySCF](https://pyscf.org/). Since we do not want to burden all OpenFermion users with this dependency, testing is disabled in the GitHub workflow. Moreover, the `resource_estimates` functionality requires the dependencies
+Since the FT costing is closely tied to manipulation of molecular integrals (localization, active space selection, benchmarking against CCSD(T), ...) the code depends on [PySCF](https://pyscf.org/) and is not installed by default. To install it do:
 
 ```
-pyscf
-h5py~=3.3.0
-jax
-jaxlib
+pip install openfermion[resources]
 ```
 
 For THC factorization, it also requires [BTAS](https://github.com/ValeevGroup/BTAS) and the [PyBTAS](https://github.com/ncrubin/pybtas) wrapper, which require their own installation + depends.

src/openfermion/resource_estimates/pbc/README.md

@@ -76,11 +76,10 @@ Further details are provided in a [tutorial](./notebooks/resource_estimates.ipyn
 
 Similar to the case of molecular resource estimation, we do not wish to burden all OpenFermion users with these dependencies, and testing with GitHub workflows is disabled. Currently we only check if pyscf is available. If it is then pytest will pick up the pbc module and run the tests. Note the tests can be quite slow due to the cost associated with building the integrals.
 
-## Requirements
-Requirements can be found in [resource_estimates.txt](../../../../dev_tools/requirements/deps/resource_estimates.txt)
+## Installation
+
+The additional requirements required can be installed with openfermion by doing
+
 ```
-pyscf
-jax
-jaxlib
-ase
+pip install openfermion[resources]
 ```