bug#60545: Keras h5py Version Mismatch

bug#60545: Keras h5py Version Mismatch

[Zacchaeus Scheffer]

[-- Attachment #1: Type: text/plain, Size: 1210 bytes --]

Hi Guix!


It would seem that the current keras version expects an earlier h5py version for loading models.  Specifically, running:

from tensorflow.keras.models import load_model
model = load_model("model.h5")

fails with:

  File "/gnu/store/idfn4l2kxil9id21v1yci4jcw6103fah-profile/lib/python3.9/site-packages/tensorflow/python/keras/engine/saving.py", line 228, in load_model
    model_config = json.loads(model_config.decode('utf-8'))
AttributeError: 'str' object has no attribute 'decode'

The above works fine in my local pip environment (see below).  A quick search online indicates that this can be fixed by downgrading h5py to version 2.10.  The current h5py version in guix is 3.6.

My assumption is that this could also be fixed with an upgrade to keras, but upgrading keras sounds much more involved.  For a short-term fix, would it be possible to add back a package definition for an earlier h5py in addition to the current version?  I'm sure an appropriate version exists somewhere in guix's git history, though my git-foo is too weak for me to check.


Thanks,
Zacchae


P.S.
working (pip): keras 2.10.0, h5py 3.7.0
not working (current guix): keras 2.2.4, h5py 3.6.0

[-- Attachment #2: Type: text/html, Size: 5052 bytes --]

bug#60545: Keras h5py Version Mismatch

[zimoun]

Hi,

On Wed, 04 Jan 2023 at 01:37, Zacchaeus Scheffer <Zacchaeus.Scheffer@synaptics.com> wrote:

> working (pip): keras 2.10.0, h5py 3.7.0
> not working (current guix): keras 2.2.4, h5py 3.6.0

Indeed, it is collateral to ea403bf290f4bf8c7bae314e32dbb90653f92996

--8<---------------cut here---------------start------------->8---
gnu: python-h5py: Update to 3.6.0.

* gnu/packages/python-xyz.scm (python-h5py): Update to 3.6.0.
[arguments]: Set HDF5_DIR variable instead of patching files.
--8<---------------cut here---------------end--------------->8---

Well, the update of Keras to 2.11.0 (the current latest) requires some
work.  First, fetch from Github instead of PyPI because the sources are
not available after v2.6.0; e.g.,

    https://pypi.org/project/keras/2.11.0/#files

Then, second adapt the recipe.  Do you want to give a try?


Cheers,
simon

bug#60545: Keras h5py Version Mismatch

[Zacchaeus Scheffer]

[-- Attachment #1: Type: text/plain, Size: 1820 bytes --]

Hi,

I should probably mention that I am not looking into this.  I want to learn more guix, but keras and the additional dependencies required for later versions is kind of a monster to me at the moment.

Thanks,
Zacchae
________________________________
From: zimoun <zimon.toutoune@gmail.com>
Sent: Friday, January 6, 2023 6:40 AM
To: Zacchaeus Scheffer <Zacchaeus.Scheffer@synaptics.com>; 60545@debbugs.gnu.org <60545@debbugs.gnu.org>
Subject: Re: bug#60545: Keras h5py Version Mismatch

CAUTION: Email originated externally, do not click links or open attachments unless you recognize the sender and know the content is safe.


Hi,

On Wed, 04 Jan 2023 at 01:37, Zacchaeus Scheffer <Zacchaeus.Scheffer@synaptics.com> wrote:

> working (pip): keras 2.10.0, h5py 3.7.0
> not working (current guix): keras 2.2.4, h5py 3.6.0

Indeed, it is collateral to ea403bf290f4bf8c7bae314e32dbb90653f92996

--8<---------------cut here---------------start------------->8---
gnu: python-h5py: Update to 3.6.0.

* gnu/packages/python-xyz.scm (python-h5py): Update to 3.6.0.
[arguments]: Set HDF5_DIR variable instead of patching files.
--8<---------------cut here---------------end--------------->8---

Well, the update of Keras to 2.11.0 (the current latest) requires some
work.  First, fetch from Github instead of PyPI because the sources are
not available after v2.6.0; e.g.,

    https://urldefense.proofpoint.com/v2/url?u=https-3A__pypi.org_project_keras_2.11.0_-23files&d=DwIBAg&c=7dfBJ8cXbWjhc0BhImu8wVIoUFmBzj1s88r8EGyM0UY&r=1moyBSYvgSFmdTYHNEHEwakZGFkPl3pixLezH1qWWXFZPpSXOvpoeGr4XIg6G8V0&m=NqD4egq46utBpoyT-8Gr6A3lESGvdg2j7CIMrB0r_y4I-pFRV2Gxn2-Lb7s7tsyD&s=Ro23h-Sd5x7VKOXeIYl4j5KGqDMAyZprruKUgdacbCo&e=

Then, second adapt the recipe.  Do you want to give a try?


Cheers,
simon


[-- Attachment #2: Type: text/html, Size: 4052 bytes --]