Open pro55b opened 3 years ago
If you install it in a conda environment through miniforge you will have access to SciPy on M1.
Does anyone know when it will be available through other python installation pathways (e.g. pyenv)?
you can use conda
If you need help setting up conda, refer to the 'Miniforge' instructions on my issue: https://github.com/apple/tensorflow_macos/issues/153
PS: I do not believe Keras works (well) on M1 so your milage may vary there.
scipy/scipy#13409
I got Keras working with Conda. Though I still had to build pillow with home-brew because I also use it and there was also some issue with h5py which I had to solve.
And as eager mode did not work I inserted the following in my code:
import tensorflow as tf tf.compat.v1.disable_eager_execution()
gpu is working and it is fast !!
Hi, I wonder how you installed keras? When I run conda install keras
or conda install -c conda-forge kera
here is what I got
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: |
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed
UnsatisfiableError: The following specifications were found to be incompatible with each other:
Output in format: Requested package -> Available versions
I'm using a MacBook Air M1 , Minforge3 with Conda 4.9.2 python 3.9.1
I've tried conda config --set channel_priority flexible
& conda update conda
but didn't work.
here is my tensorflow environment detail:
absl-py 0.11.0 pyhd8ed1ab_0 conda-forge
appnope 0.1.2 pypi_0 pypi
argon2-cffi 20.1.0 pypi_0 pypi
astunparse 1.6.3 pyhd8ed1ab_0 conda-forge
async-generator 1.10 pypi_0 pypi
attrs 20.3.0 pypi_0 pypi
backcall 0.2.0 pypi_0 pypi
bleach 3.3.0 pypi_0 pypi
ca-certificates 2020.12.5 h4653dfc_0 conda-forge
cachetools 4.2.1 pypi_0 pypi
certifi 2020.12.5 py38h10201cd_1 conda-forge
cffi 1.14.5 pypi_0 pypi
chardet 4.0.0 pypi_0 pypi
cycler 0.10.0 py_2 conda-forge
decorator 4.4.2 pypi_0 pypi
defusedxml 0.6.0 pypi_0 pypi
entrypoints 0.3 pypi_0 pypi
flatbuffers 1.12 pypi_0 pypi
freetype 2.10.4 h17b34a0_1 conda-forge
gast 0.4.0 pyh9f0ad1d_0 conda-forge
google-auth 1.27.0 pypi_0 pypi
google-auth-oauthlib 0.4.2 pypi_0 pypi
google-pasta 0.2.0 pypi_0 pypi
grpcio 1.33.2 pypi_0 pypi
h5py 2.10.0 pypi_0 pypi
idna 2.10 pypi_0 pypi
ipykernel 5.4.3 pypi_0 pypi
ipython 7.20.0 pypi_0 pypi
ipython-genutils 0.2.0 pypi_0 pypi
ipywidgets 7.6.3 pypi_0 pypi
jedi 0.18.0 pypi_0 pypi
jinja2 2.11.3 pypi_0 pypi
jpeg 9d h27ca646_0 conda-forge
jsonschema 3.2.0 pypi_0 pypi
jupyter 1.0.0 pypi_0 pypi
jupyter-client 6.1.11 pypi_0 pypi
jupyter-console 6.2.0 pypi_0 pypi
jupyter-core 4.7.1 pypi_0 pypi
jupyterlab-pygments 0.1.2 pypi_0 pypi
jupyterlab-widgets 1.0.0 pypi_0 pypi
keras-preprocessing 1.1.2 pypi_0 pypi
kiwisolver 1.3.1 py38h12a6f45_1 conda-forge
lcms2 2.12 had6a04f_0 conda-forge
libblas 3.9.0 8_openblas conda-forge
libcblas 3.9.0 8_openblas conda-forge
libcxx 11.0.1 h168391b_0 conda-forge
libffi 3.3 h9f76cd9_2 conda-forge
libgfortran 5.0.0.dev0 11_0_0_hdc626ea_18 conda-forge
libgfortran5 11.0.0.dev0 hdc626ea_18 conda-forge
liblapack 3.9.0 8_openblas conda-forge
libopenblas 0.3.12 openmp_h2ecc587_1 conda-forge
libpng 1.6.37 hf7e6567_2 conda-forge
libtiff 4.2.0 h70663a0_0 conda-forge
libwebp-base 1.2.0 h27ca646_0 conda-forge
llvm-openmp 11.0.1 hb3022d6_0 conda-forge
lz4-c 1.9.3 h9f76cd9_0 conda-forge
markdown 3.3.3 pypi_0 pypi
markupsafe 1.1.1 pypi_0 pypi
matplotlib 3.3.4 py38h150bfb4_0 conda-forge
matplotlib-base 3.3.4 py38ha029820_0 conda-forge
mistune 0.8.4 pypi_0 pypi
nbclient 0.5.2 pypi_0 pypi
nbconvert 6.0.7 pypi_0 pypi
nbformat 5.1.2 pypi_0 pypi
ncurses 6.2 h9aa5885_4 conda-forge
nest-asyncio 1.5.1 pypi_0 pypi
notebook 6.2.0 pypi_0 pypi
numpy 1.20.1 py38h9e6c65a_0 conda-forge
oauthlib 3.1.0 pypi_0 pypi
olefile 0.46 pyh9f0ad1d_1 conda-forge
openssl 1.1.1j h27ca646_0 conda-forge
opt_einsum 3.3.0 py_0 conda-forge
packaging 20.9 pypi_0 pypi
pandocfilters 1.4.3 pypi_0 pypi
parso 0.8.1 pypi_0 pypi
pexpect 4.8.0 pypi_0 pypi
pickleshare 0.7.5 pypi_0 pypi
pillow 8.1.0 py38h5cebd47_2 conda-forge
pip 21.0.1 pyhd8ed1ab_0 conda-forge
prometheus-client 0.9.0 pypi_0 pypi
prompt-toolkit 3.0.16 pypi_0 pypi
protobuf 3.15.0 pypi_0 pypi
ptyprocess 0.7.0 pypi_0 pypi
pyasn1 0.4.8 pypi_0 pypi
pyasn1-modules 0.2.8 pypi_0 pypi
pycparser 2.20 pypi_0 pypi
pygments 2.8.0 pypi_0 pypi
pyparsing 2.4.7 pyh9f0ad1d_0 conda-forge
pyrsistent 0.17.3 pypi_0 pypi
python 3.8.6 h12cc5a1_5_cpython conda-forge
python-dateutil 2.8.1 py_0 conda-forge
python_abi 3.8 1_cp38 conda-forge
pyzmq 22.0.3 pypi_0 pypi
qtconsole 5.0.2 pypi_0 pypi
qtpy 1.9.0 pypi_0 pypi
readline 8.0 hc8eb9b7_2 conda-forge
requests 2.25.1 pypi_0 pypi
requests-oauthlib 1.3.0 pypi_0 pypi
rsa 4.7.1 pypi_0 pypi
scipy 1.6.0 py38hdf044fb_0 conda-forge
send2trash 1.5.0 pypi_0 pypi
setuptools 49.6.0 py38h10201cd_3 conda-forge
six 1.15.0 pyh9f0ad1d_0 conda-forge
sqlite 3.34.0 h6d56c25_0 conda-forge
tensorboard 2.4.1 pypi_0 pypi
tensorboard-plugin-wit 1.8.0 pypi_0 pypi
tensorflow 2.4.0rc0 pypi_0 pypi
tensorflow-addons 0.11.2 pypi_0 pypi
tensorflow-estimator 2.4.0 pypi_0 pypi
termcolor 1.1.0 py_2 conda-forge
terminado 0.9.2 pypi_0 pypi
testpath 0.4.4 pypi_0 pypi
tk 8.6.10 hf7e6567_1 conda-forge
tornado 6.1 py38h30f7421_1 conda-forge
traitlets 5.0.5 pypi_0 pypi
typeguard 2.11.1 pyhd8ed1ab_0 conda-forge
typing_extensions 3.7.4.3 py_0 conda-forge
urllib3 1.26.3 pypi_0 pypi
wcwidth 0.2.5 pypi_0 pypi
webencodings 0.5.1 pypi_0 pypi
werkzeug 1.0.1 pypi_0 pypi
wheel 0.36.2 pyhd3deb0d_0 conda-forge
widgetsnbextension 3.5.1 pypi_0 pypi
wrapt 1.12.1 pypi_0 pypi
xz 5.2.5 h642e427_1 conda-forge
zlib 1.2.11 h31e879b_1009 conda-forge
zstd 1.4.8 h5b28eab_1 conda-forge
After installing the Conda environment as explained in #153 https://github.com/apple/tensorflow_macos/issues/153 I activated the environment and used "pip install keras”
This worked for me
On 21 Feb 2021, at 19:28, Phasheen notifications@github.com wrote:
Hi, I wonder how you installed keras? When I run conda install keras or conda install -c conda-forge kera here is what I got
Collecting package metadata (current_repodata.json): done Solving environment: failed with initial frozen solve. Retrying with flexible solve. Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source. Collecting package metadata (repodata.json): done Solving environment: failed with initial frozen solve. Retrying with flexible solve. Solving environment: | Found conflicts! Looking for incompatible packages. This can take several minutes. Press CTRL-C to abort. failed
UnsatisfiableError: The following specifications were found to be incompatible with each other:
Output in format: Requested package -> Available versions I'm using a MacBook Air M1 , Minforge3 with Conda 4.9.2 python 3.9.1
here is my tensorflow environment detail:
absl-py 0.11.0 pyhd8ed1ab_0 conda-forge appnope 0.1.2 pypi_0 pypi argon2-cffi 20.1.0 pypi_0 pypi astunparse 1.6.3 pyhd8ed1ab_0 conda-forge async-generator 1.10 pypi_0 pypi attrs 20.3.0 pypi_0 pypi backcall 0.2.0 pypi_0 pypi bleach 3.3.0 pypi_0 pypi ca-certificates 2020.12.5 h4653dfc_0 conda-forge cachetools 4.2.1 pypi_0 pypi certifi 2020.12.5 py38h10201cd_1 conda-forge cffi 1.14.5 pypi_0 pypi chardet 4.0.0 pypi_0 pypi cycler 0.10.0 py_2 conda-forge decorator 4.4.2 pypi_0 pypi defusedxml 0.6.0 pypi_0 pypi entrypoints 0.3 pypi_0 pypi flatbuffers 1.12 pypi_0 pypi freetype 2.10.4 h17b34a0_1 conda-forge gast 0.4.0 pyh9f0ad1d_0 conda-forge google-auth 1.27.0 pypi_0 pypi google-auth-oauthlib 0.4.2 pypi_0 pypi google-pasta 0.2.0 pypi_0 pypi grpcio 1.33.2 pypi_0 pypi h5py 2.10.0 pypi_0 pypi idna 2.10 pypi_0 pypi ipykernel 5.4.3 pypi_0 pypi ipython 7.20.0 pypi_0 pypi ipython-genutils 0.2.0 pypi_0 pypi ipywidgets 7.6.3 pypi_0 pypi jedi 0.18.0 pypi_0 pypi jinja2 2.11.3 pypi_0 pypi jpeg 9d h27ca646_0 conda-forge jsonschema 3.2.0 pypi_0 pypi jupyter 1.0.0 pypi_0 pypi jupyter-client 6.1.11 pypi_0 pypi jupyter-console 6.2.0 pypi_0 pypi jupyter-core 4.7.1 pypi_0 pypi jupyterlab-pygments 0.1.2 pypi_0 pypi jupyterlab-widgets 1.0.0 pypi_0 pypi keras-preprocessing 1.1.2 pypi_0 pypi kiwisolver 1.3.1 py38h12a6f45_1 conda-forge lcms2 2.12 had6a04f_0 conda-forge libblas 3.9.0 8_openblas conda-forge libcblas 3.9.0 8_openblas conda-forge libcxx 11.0.1 h168391b_0 conda-forge libffi 3.3 h9f76cd9_2 conda-forge libgfortran 5.0.0.dev0 11_0_0_hdc626ea_18 conda-forge libgfortran5 11.0.0.dev0 hdc626ea_18 conda-forge liblapack 3.9.0 8_openblas conda-forge libopenblas 0.3.12 openmp_h2ecc587_1 conda-forge libpng 1.6.37 hf7e6567_2 conda-forge libtiff 4.2.0 h70663a0_0 conda-forge libwebp-base 1.2.0 h27ca646_0 conda-forge llvm-openmp 11.0.1 hb3022d6_0 conda-forge lz4-c 1.9.3 h9f76cd9_0 conda-forge markdown 3.3.3 pypi_0 pypi markupsafe 1.1.1 pypi_0 pypi matplotlib 3.3.4 py38h150bfb4_0 conda-forge matplotlib-base 3.3.4 py38ha029820_0 conda-forge mistune 0.8.4 pypi_0 pypi nbclient 0.5.2 pypi_0 pypi nbconvert 6.0.7 pypi_0 pypi nbformat 5.1.2 pypi_0 pypi ncurses 6.2 h9aa5885_4 conda-forge nest-asyncio 1.5.1 pypi_0 pypi notebook 6.2.0 pypi_0 pypi numpy 1.20.1 py38h9e6c65a_0 conda-forge oauthlib 3.1.0 pypi_0 pypi olefile 0.46 pyh9f0ad1d_1 conda-forge openssl 1.1.1j h27ca646_0 conda-forge opt_einsum 3.3.0 py_0 conda-forge packaging 20.9 pypi_0 pypi pandocfilters 1.4.3 pypi_0 pypi parso 0.8.1 pypi_0 pypi pexpect 4.8.0 pypi_0 pypi pickleshare 0.7.5 pypi_0 pypi pillow 8.1.0 py38h5cebd47_2 conda-forge pip 21.0.1 pyhd8ed1ab_0 conda-forge prometheus-client 0.9.0 pypi_0 pypi prompt-toolkit 3.0.16 pypi_0 pypi protobuf 3.15.0 pypi_0 pypi ptyprocess 0.7.0 pypi_0 pypi pyasn1 0.4.8 pypi_0 pypi pyasn1-modules 0.2.8 pypi_0 pypi pycparser 2.20 pypi_0 pypi pygments 2.8.0 pypi_0 pypi pyparsing 2.4.7 pyh9f0ad1d_0 conda-forge pyrsistent 0.17.3 pypi_0 pypi python 3.8.6 h12cc5a1_5_cpython conda-forge python-dateutil 2.8.1 py_0 conda-forge python_abi 3.8 1_cp38 conda-forge pyzmq 22.0.3 pypi_0 pypi qtconsole 5.0.2 pypi_0 pypi qtpy 1.9.0 pypi_0 pypi readline 8.0 hc8eb9b7_2 conda-forge requests 2.25.1 pypi_0 pypi requests-oauthlib 1.3.0 pypi_0 pypi rsa 4.7.1 pypi_0 pypi scipy 1.6.0 py38hdf044fb_0 conda-forge send2trash 1.5.0 pypi_0 pypi setuptools 49.6.0 py38h10201cd_3 conda-forge six 1.15.0 pyh9f0ad1d_0 conda-forge sqlite 3.34.0 h6d56c25_0 conda-forge tensorboard 2.4.1 pypi_0 pypi tensorboard-plugin-wit 1.8.0 pypi_0 pypi tensorflow 2.4.0rc0 pypi_0 pypi tensorflow-addons 0.11.2 pypi_0 pypi tensorflow-estimator 2.4.0 pypi_0 pypi termcolor 1.1.0 py_2 conda-forge terminado 0.9.2 pypi_0 pypi testpath 0.4.4 pypi_0 pypi tk 8.6.10 hf7e6567_1 conda-forge tornado 6.1 py38h30f7421_1 conda-forge traitlets 5.0.5 pypi_0 pypi typeguard 2.11.1 pyhd8ed1ab_0 conda-forge typing_extensions 3.7.4.3 py_0 conda-forge urllib3 1.26.3 pypi_0 pypi wcwidth 0.2.5 pypi_0 pypi webencodings 0.5.1 pypi_0 pypi werkzeug 1.0.1 pypi_0 pypi wheel 0.36.2 pyhd3deb0d_0 conda-forge widgetsnbextension 3.5.1 pypi_0 pypi wrapt 1.12.1 pypi_0 pypi xz 5.2.5 h642e427_1 conda-forge zlib 1.2.11 h31e879b_1009 conda-forge zstd 1.4.8 h5b28eab_1 conda-forge — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/apple/tensorflow_macos/issues/161#issuecomment-782902916, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIWR3K6M3HYYGBSP7HVT57LTAFGEFANCNFSM4XJKT5QQ.
This is my environment detail:
absl-py 0.11.0 pypi_0 pypi
appnope 0.1.2 py38h10201cd_1 conda-forge
astunparse 1.6.3 pypi_0 pypi
backcall 0.2.0 pyh9f0ad1d_0 conda-forge
backports 1.0 py_2 conda-forge
backports.functools_lru_cache 1.6.1 py_0 conda-forge
c-ares 1.17.1 h27ca646_0 conda-forge
ca-certificates 2020.12.5 h4653dfc_0 conda-forge
cached-property 1.5.1 py_0 conda-forge
cachetools 4.2.1 pypi_0 pypi
certifi 2020.6.20 pyhd3eb1b0_3
chardet 4.0.0 pypi_0 pypi
cycler 0.10.0 pypi_0 pypi
decorator 4.4.2 py_0 conda-forge
flatbuffers 1.12 pypi_0 pypi
freetype 2.10.4 h17b34a0_1 conda-forge
gast 0.4.0 pypi_0 pypi
google-auth 1.26.1 pypi_0 pypi
google-auth-oauthlib 0.4.2 pypi_0 pypi
google-pasta 0.2.0 pypi_0 pypi
grpcio 1.35.0 py38h446835f_0 conda-forge
h5py 3.1.0 nompi_py38h032b01a_100 conda-forge
hdf5 1.10.6 nompi_h0fc092c_1114 conda-forge
idna 2.10 pypi_0 pypi
install 1.3.4 pypi_0 pypi
ipython 7.20.0 py38h2cb4d76_2 conda-forge
ipython_genutils 0.2.0 py_1 conda-forge
jedi 0.18.0 py38h10201cd_2 conda-forge
jpeg 9d h27ca646_0 conda-forge
keras 2.4.3 pypi_0 pypi
keras-preprocessing 1.1.2 pypi_0 pypi
kiwisolver 1.3.1 pypi_0 pypi
krb5 1.17.2 h17618d6_0 conda-forge
lcms2 2.12 had6a04f_0 conda-forge
libblas 3.9.0 8_openblas conda-forge
libcblas 3.9.0 8_openblas conda-forge
libcurl 7.71.1 hd2aec06_8 conda-forge
libcxx 11.0.1 h168391b_0 conda-forge
libedit 3.1.20191231 hc8eb9b7_2 conda-forge
libev 4.33 h642e427_1 conda-forge
libffi 3.3 h9f76cd9_2 conda-forge
libgfortran 5.0.0.dev0 11_0_0_hdc626ea_18 conda-forge
libgfortran5 11.0.0.dev0 hdc626ea_18 conda-forge
liblapack 3.9.0 8_openblas conda-forge
libnghttp2 1.43.0 hf3018f0_0 conda-forge
libopenblas 0.3.12 openmp_h2ecc587_1 conda-forge
libpng 1.6.37 hf7e6567_2 conda-forge
libssh2 1.9.0 h1c49ba1_5 conda-forge
libtiff 4.2.0 h70663a0_0 conda-forge
libwebp-base 1.2.0 h27ca646_0 conda-forge
llvm-openmp 11.0.1 hb3022d6_0 conda-forge
lz4-c 1.9.3 h9f76cd9_0 conda-forge
markdown 3.3.3 pypi_0 pypi
matplotlib 3.3.4 pypi_0 pypi
ncurses 6.2 h9aa5885_4 conda-forge
numpy 1.20.1 py38h9e6c65a_0 conda-forge
oauthlib 3.1.0 pypi_0 pypi
olefile 0.46 py_0
openssl 1.1.1j h27ca646_0 conda-forge
opt-einsum 3.3.0 pypi_0 pypi
parso 0.8.1 pyhd8ed1ab_0 conda-forge
pexpect 4.8.0 pyh9f0ad1d_2 conda-forge
pickleshare 0.7.5 py_1003 conda-forge
pillow 8.1.0 py38h5cebd47_2 conda-forge
pip 20.2.4 py_0 conda-forge
prompt-toolkit 3.0.16 pyha770c72_0 conda-forge
protobuf 3.14.0 pypi_0 pypi
ptyprocess 0.7.0 pyhd3deb0d_0 conda-forge
pyasn1 0.4.8 pypi_0 pypi
pyasn1-modules 0.2.8 pypi_0 pypi
pygments 2.7.4 pyhd8ed1ab_0 conda-forge
pyparsing 2.4.7 pypi_0 pypi
python 3.8.6 h12cc5a1_5_cpython conda-forge
python-dateutil 2.8.1 pypi_0 pypi
python_abi 3.8 1_cp38 conda-forge
pyyaml 5.4.1 pypi_0 pypi
readline 8.0 hc8eb9b7_2 conda-forge
requests 2.25.1 pypi_0 pypi
requests-oauthlib 1.3.0 pypi_0 pypi
rsa 4.7 pypi_0 pypi
scipy 1.6.0 py38hdf044fb_0 conda-forge
setuptools 49.6.0 py38h10201cd_3 conda-forge
six 1.15.0 pyh9f0ad1d_0 conda-forge
sqlite 3.34.0 h6d56c25_0 conda-forge
tensorboard 2.4.1 pypi_0 pypi
tensorboard-plugin-wit 1.8.0 pypi_0 pypi
tensorflow 2.4.0rc0 pypi_0 pypi
tensorflow-addons 0.11.2 pypi_0 pypi
tensorflow-estimator 2.4.0 pypi_0 pypi
termcolor 1.1.0 py_2 conda-forge
tk 8.6.10 hf7e6567_1 conda-forge
traitlets 5.0.5 py_0 conda-forge
typeguard 2.9.1 py_0 conda-forge
typing-extensions 3.7.4.3 pypi_0 pypi
urllib3 1.26.3 pypi_0 pypi
wcwidth 0.2.5 pyh9f0ad1d_2 conda-forge
werkzeug 1.0.1 pypi_0 pypi
wheel 0.36.2 pyhd3deb0d_0 conda-forge
wrapt 1.12.1 pypi_0 pypi
xz 5.2.5 h642e427_1 conda-forge
zlib 1.2.11 h31e879b_1009 conda-forge
zstd 1.4.8 h5b28eab_1 conda-forge
oh my god, I tried pip install before and didn't work, but it worked just now. I figured that was because I didn't install scipy before, and pip install keras will fail to install scipy. Later I installed Scipy with conda, but never tried pip again. Thank you!
SciPy is installed with the environment.yml mentioned in the article https://github.com/apple/tensorflow_macos/issues/153
Does anyone know when it will be available through other python installation pathways (e.g. pyenv)?
It can be manually installed using pip.
$ NPY_DISTUTILS_APPEND_FLAGS=1 CFLAGS="-falign-functions=8 ${CFLAGS}" pip install scipy
You need gcc (and probably some other utilities) installed via Homebrew. The option "NPY_DISTUTILS_APPEND_FLAGS=1" is required when compiling for Python 3.8 (not necessary for 3.9).
A fix for gcc
has been reflected in the latest Homebrew a few days ago, and the workaround of -falign-functions=8
becomes unnecessary anymore. Add the following lines to ~/.numpy-site.cfg
to use Homebrew's OpenBLAS:
[openblas]
libraries = openblas
library_dirs = /opt/homebrew/opt/openblas/lib
include_dirs = /opt/homebrew/opt/openblas/include
And then, you can install SciPy in the usual way.
$ NPY_DISTUTILS_APPEND_FLAGS=1 pip install scipy
I want to Use Keras but as there is a SciPy dependency this is not yet possible.
Any idea when SciPy will be available for M1.