You are not logged in.

#1 2024-09-29 23:31:09

gerwazy
Member
Registered: 2024-03-09
Posts: 5

TensorFlow and NumPy are not compatible?

I have installed from repository :
- python-numpy 2.1.1-1
- python-tensorflow 2.17.0-2
- tensorflow 2.17.0-2

When I try to import TensorFlow, I get the following error:

A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.1.1 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.

If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.

Traceback (most recent call last):  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/usr/lib/python3.12/site-packages/ipykernel_launcher.py", line 18, in <module>
    app.launch_new_instance()
  File "/usr/lib/python3.12/site-packages/traitlets/config/application.py", line 1075, in launch_instance
    app.start()
  File "/usr/lib/python3.12/site-packages/ipykernel/kernelapp.py", line 739, in start
    self.io_loop.start()
  File "/usr/lib/python3.12/site-packages/tornado/platform/asyncio.py", line 205, in start
    self.asyncio_loop.run_forever()
  File "/usr/lib/python3.12/asyncio/base_events.py", line 641, in run_forever
    self._run_once()
  File "/usr/lib/python3.12/asyncio/base_events.py", line 1986, in _run_once
    handle._run()
  File "/usr/lib/python3.12/asyncio/events.py", line 88, in _run
    self._context.run(self._callback, *self._args)
  File "/usr/lib/python3.12/site-packages/ipykernel/kernelbase.py", line 545, in dispatch_queue
    await self.process_one()
  File "/usr/lib/python3.12/site-packages/ipykernel/kernelbase.py", line 534, in process_one
    await dispatch(*args)
  File "/usr/lib/python3.12/site-packages/ipykernel/kernelbase.py", line 437, in dispatch_shell
    await result
  File "/usr/lib/python3.12/site-packages/ipykernel/ipkernel.py", line 362, in execute_request
    await super().execute_request(stream, ident, parent)
  File "/usr/lib/python3.12/site-packages/ipykernel/kernelbase.py", line 778, in execute_request
    reply_content = await reply_content
  File "/usr/lib/python3.12/site-packages/ipykernel/ipkernel.py", line 449, in do_execute
    res = shell.run_cell(
  File "/usr/lib/python3.12/site-packages/ipykernel/zmqshell.py", line 549, in run_cell
    return super().run_cell(*args, **kwargs)
  File "/usr/lib/python3.12/site-packages/IPython/core/interactiveshell.py", line 3075, in run_cell
    result = self._run_cell(
  File "/usr/lib/python3.12/site-packages/IPython/core/interactiveshell.py", line 3130, in _run_cell
    result = runner(coro)
  File "/usr/lib/python3.12/site-packages/IPython/core/async_helpers.py", line 128, in _pseudo_sync_runner
    coro.send(None)
  File "/usr/lib/python3.12/site-packages/IPython/core/interactiveshell.py", line 3334, in run_cell_async
    has_raised = await self.run_ast_nodes(code_ast.body, cell_name,
  File "/usr/lib/python3.12/site-packages/IPython/core/interactiveshell.py", line 3517, in run_ast_nodes
    if await self.run_code(code, result, async_=asy):
  File "/usr/lib/python3.12/site-packages/IPython/core/interactiveshell.py", line 3577, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "/tmp/ipykernel_51366/4197461287.py", line 7, in <module>
    import tensorflow as tf
  File "/usr/lib/python3.12/site-packages/tensorflow/__init__.py", line 46, in <module>
    from tensorflow._api.v2 import __internal__
  File "/usr/lib/python3.12/site-packages/tensorflow/_api/v2/__internal__/__init__.py", line 11, in <module>
    from tensorflow._api.v2.__internal__ import distribute
  File "/usr/lib/python3.12/site-packages/tensorflow/_api/v2/__internal__/distribute/__init__.py", line 8, in <module>
    from tensorflow._api.v2.__internal__.distribute import combinations
  File "/usr/lib/python3.12/site-packages/tensorflow/_api/v2/__internal__/distribute/combinations/__init__.py", line 8, in <module>
    from tensorflow.python.distribute.combinations import env # line: 456
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/combinations.py", line 33, in <module>
    from tensorflow.python.distribute import collective_all_reduce_strategy
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/collective_all_reduce_strategy.py", line 25, in <module>
    from tensorflow.python.distribute import cross_device_ops as cross_device_ops_lib
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/cross_device_ops.py", line 28, in <module>
    from tensorflow.python.distribute import cross_device_utils
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/cross_device_utils.py", line 22, in <module>
    from tensorflow.python.distribute import values as value_lib
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/values.py", line 23, in <module>
    from tensorflow.python.distribute import distribute_lib
  File "/usr/lib/python3.12/site-packages/tensorflow/python/distribute/distribute_lib.py", line 205, in <module>
    from tensorflow.python.data.ops import dataset_ops
  File "/usr/lib/python3.12/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 34, in <module>
    from tensorflow.python.data.ops import iterator_ops
  File "/usr/lib/python3.12/site-packages/tensorflow/python/data/ops/iterator_ops.py", line 45, in <module>
    from tensorflow.python.training.saver import BaseSaverBuilder
  File "/usr/lib/python3.12/site-packages/tensorflow/python/training/saver.py", line 50, in <module>
    from tensorflow.python.training import py_checkpoint_reader
  File "/usr/lib/python3.12/site-packages/tensorflow/python/training/py_checkpoint_reader.py", line 19, in <module>
    from tensorflow.python.util._pywrap_checkpoint_reader import CheckpointReader

How can I solve this error?

Offline

#2 2024-09-30 01:28:50

realasking
Member
Registered: 2023-08-18
Posts: 4

Re: TensorFlow and NumPy are not compatible?

same problem here...
I'm trying to use virtual envs now.

Offline

#3 2024-09-30 06:02:16

mountaintrek
Member
Registered: 2024-02-01
Posts: 29

Re: TensorFlow and NumPy are not compatible?

There is an issue for it: https://gitlab.archlinux.org/archlinux/ … /issues/11.
There was a discussion at  tensorflow. I'm not a user, but was curious. The content of the issue might have more meaning to you.

Just a suggestion. There is a train of thought in the python world --- leave the system python environment alone and do application development in a virtual environment.  This reduces the chances of negatively impacting the running system by isolating the projects' python environment. It also gives you the ability to easily roll-back or install a specific version of a module. In this case, the temporary fix seems to install a specific version of numpy, which can be done in a virtual environment.

A simple example of a Python virtual environment

# Make your project directory and change into it

   cd $HOME
   mkdir -p Projects/MyNewProject && cd Projects/MyNewProject

# Create virtual environment.
# Afterwards there will be a hidden directory, Projects/MyNewProject/.venv, where everything is installed.
# This will also be the first directory in $PATH. 

    python -m venv .venv

# activate/deactivate this specific python environment

    source .venv/bin/activate

# Add modules to "requirements.txt". 

    pip install numypy==1.26.4
    pip install tensorflow
    pip install pip-review
    # add as needed

# Install

    pip install -r requirements.txt

NOTE: When I installed "tensorflow" using "pip" in a virtual environment, it uninstalled numpy 2.1.1 and installed numpy 1.26.4 automatically.

Links
- venv — Creation of virtual environments
- Python package search
- Python Virtual Environments: A Primer

Offline

Board footer

Powered by FluxBB