ERROR: No matching distribution found for jaxlib==0.4.10

[YSYL-FSH]

Why is there a version incompatibility problem during installation, I've tried 3.9 and 3.10

[YSYL-FSH]

Collecting jax==0.4.10 (from cell2fate==0.1a0)
Using cached jax-0.4.10.tar.gz (1.3 MB)
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
INFO: pip is looking at multiple versions of cell2fate to determine which version is compatible with other requirements. This could take a while.
ERROR: Ignored the following yanked versions: 0.4.32
ERROR: Ignored the following versions that require a different python version: 0.12.0rc1 Requires-Python >=3.11; 1.6.2 Requires-Python >=3.7,<3.10; 1.6.3 Requires-Python >=3.7,<3.10; 1.7.0 Requires-Python >=3.7,<3.10; 1.7.1 Requires-Python >=3.7,<3.10
ERROR: Could not find a version that satisfies the requirement jaxlib==0.4.10 (from cell2fate) (from versions: 0.4.18, 0.4.19, 0.4.20, 0.4.21, 0.4.22, 0.4.23, 0.4.24, 0.4.25, 0.4.26, 0.4.27, 0.4.28, 0.4.29, 0.4.30, 0.4.31, 0.4.33, 0.4.34, 0.4.35, 0.4.36, 0.4.38, 0.5.0, 0.5.1, 0.5.3, 0.6.0)
ERROR: No matching distribution found for jaxlib==0.4.10
(cell2fate_env_s) [fsh123@cpu7 python]$ pip install jaxlib==0.4.10
ERROR: Ignored the following yanked versions: 0.4.32
ERROR: Could not find a version that satisfies the requirement jaxlib==0.4.10 (from versions: 0.4.18, 0.4.19, 0.4.20, 0.4.21, 0.4.22, 0.4.23, 0.4.24, 0.4.25, 0.4.26, 0.4.27, 0.4.28, 0.4.29, 0.4.30, 0.4.31, 0.4.33, 0.4.34, 0.4.35, 0.4.36, 0.4.38, 0.5.0, 0.5.1, 0.5.3, 0.6.0)
ERROR: No matching distribution found for jaxlib==0.4.10

[Windaswill]

Me too, I meet the same problem, did you sovle it?

[nynkekatinka]

I have the same issue. Given that JAX has not supported python 3.9 for over a year I thought that would solve it. They stopped supporting python 3.10 last month. However, when running the pip install with python==3.11 it runs into issues with scanpy and scipy.

Honestly, given that Cell2Fate relies on several packages that are well-known to run into dependency issues, it would've been so much easier if the developers would've just build a conda package so that the conda dependency solver could figure it out before breaking everything like pip tends to do.

Shame, guess I'll look for a different tool.

[Thapeachydude]

Hi, has anyone been able to solve this?

[Anna-E-F]

Hi, I am facing the same problem ; with py=3.9 or 3.10 it fails to find jaxlib, and with py=3.11 it fails to find scipy
A solution is to install jax and jaxlib before running pip install git+...cell2fate

[Anna-E-F]

Hi again, it seems to work now with the following conda environment. Please let me know if you see anything that could be improved (I'm quite new to conda environments etc.).

# create env and have compatible pip version
conda create -n cell2fate2 python=3.9 -y
conda activate cell2fate2

pip install "pip<24.1"

# install torch etc
mamba install -c conda-forge pytorch torchvision torchaudio -y

# installations via pip
pip install gseapy importlib_metadata==4.8
pip install leidenalg==0.10.1
pip install scvelo==0.2.4
pip install flatbuffers==2.0
pip install pytorch-lightning==1.5.10

# installations via mamba
mamba install -c conda-forge ipykernel -y

mamba install -c conda-forge \
  numpy=1.23.5 \
  scipy=1.8.0 \
  pandas=1.4.2 \
  matplotlib=3.5.2 \
  rich=12.3.0 \
  anndata=0.8.0 \
  scanpy=1.9.1 \
  scvi-tools=0.16.1 \
  pyro-ppl=1.8.* \
  einops \
  ml_dtypes=0.2.0 \
  -y

# install jax, --no-deps to keep vesions of packages installed before
pip install --no-deps  \
  jax==0.4.20 \
  jaxlib==0.4.20 \
  chex==0.1.86 \
  optax==0.2.2 \
  numpyro==0.13.2 \
  -f https://storage.googleapis.com/jax-releases/jax_releases.html

# install cell2fate also with --no-deps
pip install --no-deps git+https://github.com/BayraktarLab/cell2fate

At this step I had the error ImportError: cannot import name 'ShapedArray' from 'jax'
which seems to come again from an incompatibility of cell2fate with this jax version, but since jax 0.4 seems necessary, I had to modify the source code "from jax import ShapedArray" into

try:   
    from jax import ShapedArray   
except ImportError:   
    from jax.core import ShapedArray