Skip to content

problem install scPhere using "python setup.py install" #23

@schloegl

Description

@schloegl

I tried to install the software on a GNU/Linux system (Debian 11), that had also various python versions available.
For installing scPhere I tried to using this command:

python setup.py install

Python 3.6 it fails with

if options.cache_dir else None
  File ".../python/3.6.15/lib/python3.6/posixpath.py", line 80, in join
    a = os.fspath(a)
TypeError: expected str, bytes or os.PathLike object, not int

Under python 3.7, it fails with this error.


Running matplotlib-3.7.1/setup.py -q bdist_egg --dist-dir /tmp/easy_install-fh_ykc9c/matplotlib-3.7.1/egg-dist-tmp-xnxvdzb6
error: Setup script exited with
Beginning with Matplotlib 3.6, Python 3.8 or above is required.
You are using Python 3.7.12.

Under python 3.9, installation fails with


No local packages or working download links found for tensorflow==1.14.0
error: Could not find suitable distribution for Requirement.parse('tensorflow==1.14.0')

When trying to relay the requirements in setup.py like this:

--- a/setup.py
+++ b/setup.py
@@ -13,11 +13,12 @@ setup(
     license='BSD 3-clause',
     url="https://github.com/klarman-cell-observatory/scPhere",
     install_requires=['numpy >= 1.16.4',
-                      'tensorflow == 1.14.0',
+                      'tensorflow',
                       'scipy >= 1.3.0',
                       'pandas >= 0.21.0',
                       'matplotlib >= 3.1.0',
                       'tensorflow_probability == 0.7.0',
+                      'urllib3 < 2.0',
                       ],
     packages=find_packages(),
     python_requires='>=3.6',

the installation finishes. But at runtime, it fails with this error:

user@myhost:~$ python $PREFIX/scPhere/20230622/scPhere/example/script/demo.py
2023-06-26 08:17:58.909399: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
Traceback (most recent call last):
  File "$PREFIX/scPhere/20230622/scPhere/example/script/demo.py", line 5, in <module>
    from scphere.util.util import read_mtx
  File "<frozen zipimport>", line 259, in load_module
  File "$PREFIXscPhere/20230622/lib/python3.9/site-packages/scPhere-0.1.0-py3.9.egg/scphere/util/util.py", line 3, in <module>
  File "$PREFIX/scPhere/20230622/lib/python3.9/site-packages/tensorflow-2.13.0rc1-py3.9-linux-x86_64.egg/tensorflow/__init__.py", line 38, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "$PREFIX/scPhere/20230622/lib/python3.9/site-packages/tensorflow-2.13.0rc1-py3.9-linux-x86_64.egg/tensorflow/python/__init__.py", line 37, in <module>
    from tensorflow.python.eager import context
  File "$PREFIX/scPhere/20230622/lib/python3.9/site-packages/tensorflow-2.13.0rc1-py3.9-linux-x86_64.egg/tensorflow/python/eager/context.py", line 29, in <module>
    from tensorflow.core.framework import function_pb2
  File "/$PREFIX/scPhere/20230622/lib/python3.9/site-packages/tensorflow-2.13.0rc1-py3.9-linux-x86_64.egg/tensorflow/core/framework/function_pb2.py", line 5, in <module>
    from google.protobuf.internal import builder as _builder
ModuleNotFoundError: No module named 'google.protobuf' 

Do you have any hints how to install scPhere on recent system ?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions