pkg_resources.ContextualVersionConflict: numpy 1.14.0

[ugurtigli]

Environment

• Operating System: Darwin Kernel Version 17.5.0: Mon Mar 5 22:24:32 PST 2018; root:xnu-4570.51.1~1/RELEASE_X86_64 x86_64
• Python Version: Python 3.6.7
• Python Bitness: 64
• Python runtime: python is /opt/venv/bin/python
  python is /usr/bin/python
• Python alias: alias python | grep python
• Python packages: pip freeze
  tensorflow==2.0.0
  tensorflow-estimator==2.0.1
• How did you install Stock Analysis Engine: pip
  • Are you using a virtualenv? yes
  • Are you using a pipenv? no
  • Are you using anaconda? no
• Are you running outside docker or inside containers?
  inside docker
  Now that you know a little about me, let me tell you about the issue I am having:

Description of Issue

numpy version conflicts with what is required by tensorflow vs. mains stack.

• What did you expect to happen?
• What happened instead?
  | --> fetch -t SPY
  Traceback (most recent call last):
  File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 583, in _build_master
  ws.require(requires)
  File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 900, in require
  needed = self.resolve(parse_requirements(requirements))
  File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 791, in resolve
  raise VersionConflict(dist, req).with_context(dependent_req)
  pkg_resources.ContextualVersionConflict: (numpy 1.14.0 (/opt/venv/lib/python3.6/site-packages), Requirement.parse('numpy<2.0,>=1.16.0'), {'tensorflow'})

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/opt/venv/bin/fetch", line 6, in
from pkg_resources import load_entry_point
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 3251, in
@_call_aside
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 3235, in _call_aside
f(*args, **kwargs)
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 3264, in _initialize_master_working_set
working_set = WorkingSet._build_master()
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 585, in _build_master
return cls._build_from_requirements(requires)
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 598, in _build_from_requirements
dists = ws.resolve(reqs, Environment())
File "/opt/venv/lib/python3.6/site-packages/pkg_resources/init.py", line 791, in resolve
raise VersionConflict(dist, req).with_context(dependent_req)
pkg_resources.ContextualVersionConflict: (numpy 1.14.0 (/opt/venv/lib/python3.6/site-packages), Requirement.parse('numpy<2.0,>=1.16.0'), {'tensorflow'})
Here is how you can reproduce this issue on your machine:

What steps have you taken to resolve this already?

Tried to upgrade numpy however, then main engine stops working since it require version lower (i.e. 1.14)
...

Anything else?

...

[vmajor]

I have the same issue.

Parser demands numpy<=1.14 and numpy<2.0,>=1.14.5 simultaneously.

This seems unresolvable. I have 1.18.0 which is the latest pip3 installable version on Ubuntu 18.04

[vmajor]

SOLVED:

It is a cludge, but it works to get me to the next error (KeyError: 'No metadata except PKG-INFO is available'):

1. Uninstall numpy from both the venv and user that you are using to run sa (in my case root due to the instructions to run it from /opt/sa)

2. install numpy 1.13.3 either with pip or pip3. In my case: pip3 install 'numpy==1.13.3'

3. If you were blessed with multiple python installations just because, symlink from python3.6 to python3:

ln -s /usr/local/lib/python3.6/dist-packages/numpy  /usr/lib/python3/dist-packages/numpy
ln -s /usr/local/lib/python3.6/dist-packages/numpy-1.13.3.dist-info  /usr/lib/python3/dist-packages/numpy-1.13.3.dist-info

[atulhm]

Any progress on this issue? I am having the same problem

[jay-johnson]

Sorry yea the requirements needed to be updated.

I got tests passing with yesterday's release (pypi's latest should be the same as well):

pip list | grep -E "stock-analysis-engine|numpy|pandas|tensorflow"
numpy                         1.19.1
pandas                        1.0.5
stock-analysis-engine         1.9.15      /opt/sa
tensorflow                    2.2.0

python --version
Python 3.8.0

Hopefully this new build resolves this issue.