I am facing 4 problems when I tried to install TensorFlow on Apple M1:
Conda has supported M1 since 2022.05.06 but most of articles I googled talk about using Miniforge, e.g. So I feel they are all kind of outdated.
How To Install TensorFlow on M1 Mac (The Easy Way)
AI - Apple Silicon Mac M1 natively supports TensorFlow 2.8 GPU acceleration
How to Setup TensorFlow on Apple M1 Pro and M1 Max (works for M1 too)
How To Install TensorFlow 2.7 on MacBook Pro M1 Pro With Ease
I used the latest conda 4.13 to setup my python environment(3.8, 3.9 and 3.10) successfully but when I tried to install tensorflow I got the error "No matching distribution found for tensorflow" (all failed).
ERROR: Could not find a version that satisfies the requirement tensorflow (from versions: none)
ERROR: No matching distribution found for tensorflow
The answers in Could not find a version that satisfies the requirement tensorflow didn't help. I can't find useful information on https://www.tensorflow.org/ too, actually https://www.tensorflow.org/install just said pip install tensorflow.
I tried to run pip install tensorflow-macos and it succeeded.
I read from the above "works for M1 too" article mentioned "Apple's fork of TensorFlow is called tensorflow-macos" although I can't find much information about that. For example, https://www.tensorflow.org/ does not mention that. I also found from https://developer.apple.com/forums/thread/686926 that someone hit that "ERROR: No matching distribution found for tensorflow-macos" (but I didn't).
All the articles I googled, including above 4 articles and this Tensorflow on macOS Apple M1, all say I also need to run the following 2 commands
conda install -c apple tensorflow-deps
pip install tensorflow-metal
But do I really need to that? I can't find this information from https://www.tensorflow.org/.
What are these 2 packages tensorflow-deps and tensorflow-metal ?
Distilling the official directions from Apple (as of 13 July 2022), one would create an environment using the following YAML:
tf-metal-arm64.yaml
name: tf-metal
channels:
- apple
- conda-forge
dependencies:
- python=3.9 ## specify desired version
- pip
- tensorflow-deps
## uncomment for use with Jupyter
## - ipykernel
## PyPI packages
- pip:
- tensorflow-macos
- tensorflow-metal ## optional, but recommended
Edit to include additional packages.
Creating environment
Before creating the environment we need to know what the base architecture is. Check this with conda config --show subdir.
Native (osx-arm64) base
If you have installed a native osx-arm64 Miniforge variant (I recommend Mambaforge), then you can create with:
mamba env create -n my_tf_env -f tf-metal-arm64.yaml
Note: If you don't have Mamba, then substitute conda for mamba; or install it for much faster solving: conda install -n base mamba.
Emulated (osx-64) base
If you do not have a native base, then you will need to override the subdir setting:
## create env
CONDA_SUBDIR=osx-arm64 mamba env create -n my_tf_env -f tf-metal-arm64.yaml
## activate
mamba activate my_tf_env
## permanently set the subdir
conda config --env --set subdir osx-arm64
Be sure to always activate the environment before installing or updating packages.
I battled with this for hours. The current instructions at https://developer.apple.com/metal/tensorflow-plugin/ specify using Miniconda and can be summarized as:
conda create -y --name cv python
conda activate cv
conda install -y -c apple tensorflow-deps
python -m pip install tensorflow-macos tensorflow-metal
As of Jan 2023, these instructions are riddled with issues:
Symptom: you ran conda install -c apple tensorflow-deps expecting to get the current version (2.10.0) , but conda list tensorflow-deps shows tensorflow-deps 2.9.0
Reason: Apple's tensorflow-deps package v2.10.0 depends on numpy >=1.23.2,<1.23.3. There is no such version of numpy in Anaconda (only conda-forge). Anaconda's dependency resolution silently falls back to an older version of tensorflow-deps. This will cause more problems as you continue with the instructions.
Symptom: you ran conda install -c apple tensorflow-deps==2.10.0 and got UnsatisfiableError: The following specifications were found to be incompatible with each other
Reason: Same as above, but at least Anaconda has told you about it. It doesn't mention what the incompatibility is, because that would be helpful.
Symptom: "RuntimeError: module compiled against API version 0x10 but this version of numpy is 0xf"
Reason: If you install tensorflow-deps 2.9.0 (or Anaconda installs it for you due to the above) you will get numpy 1.22.3. When you pip install tensorflow-macos tensorflow-metal, you will get tensorflow-macos 2.11.0 and tensorflow-metal 0.7.0, one or both of which is binary-incompatible with numpy 1.22.3.
Symptom: 2023-01-22 15:16:23.209301: W tensorflow/core/framework/op_kernel.cc:1830] OP_REQUIRES failed at xla_ops.cc:418 : NOT_FOUND: could not find registered platform with id
Problem: TensorFlow 2.11 has introduced an incompatibility between optimizers and pluggable architectures. You can have TF 2.11, but you'll need to use a legacy optimizer.
Solutions
I have found 3 options for a working GPU-accelerated TF install on Apple Silicon using Anaconda.
It will help your debugging to get a minimal test script like the one from https://developer.apple.com/metal/tensorflow-plugin/ which is:
import tensorflow as tf
cifar = tf.keras.datasets.cifar100
(x_train, y_train), (x_test, y_test) = cifar.load_data()
model = tf.keras.applications.ResNet50(
include_top=True,
weights=None,
input_shape=(32, 32, 3),
classes=100,)
loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
model.compile(optimizer="adam", loss=loss_fn, metrics=["accuracy"])
model.fit(x_train, y_train, epochs=5, batch_size=64)
This way you can check GPU usage by running TF_MLC_LOGGING=1 python tf_arch_test.py and watching Activity Monitor. While feeling which side of the laptop is burning your legs can be a helpful indicator of GPU usage, it's not always reliable.
Fix #1: Add conda-forge as a channel, use the legacy optimizer in your code.
conda config --add channels conda-forge
conda config --set channel_priority strict
conda create -y --name cv
conda activate cv
conda install -y -c apple tensorflow-deps
python -m pip install tensorflow-macos
python -m pip install tensorflow-metal
python --version
conda list|grep -E '(tensorflow|numpy)'
TF_MLC_LOGGING=1 python tf_arch_test.py
You will get:
Python 3.10.8
numpy 1.23.2 py310h127c7cf_0 conda-forge
tensorflow-deps 2.10.0 0 apple
tensorflow-estimator 2.11.0 pypi_0 pypi
tensorflow-macos 2.11.0 pypi_0 pypi
tensorflow-metal 0.7.0 pypi_0 pypi
You will need to modify your code to use one of the legacy optimizers. e.g.:
model.compile(
optimizer=tf.keras.optimizers.legacy.Adam(learning_rate=1e-3),
loss=loss_fn,
metrics=["accuracy"],
)
Fix #2: Add conda-forge as a channel, pin the version numbers to avoid the pluggable architecture issue.
conda config --add channels conda-forge
conda config --set channel_priority strict
conda create -y --name cv
conda activate cv
conda install -y -c apple tensorflow-deps==2.10.0
python -m pip install tensorflow-macos==2.10.0
python -m pip install tensorflow-metal==0.6.0
python --version
conda list|grep -E '(tensorflow|numpy)'
You will get:
Python 3.10.8
numpy 1.23.2 py310h127c7cf_0 conda-forge
tensorflow-deps 2.10.0 0 apple
tensorflow-estimator 2.10.0 pypi_0 pypi
tensorflow-macos 2.10.0 pypi_0 pypi
tensorflow-metal 0.6.0 pypi_0 pypi
Fix #3: Don't add conda-forge, pin the version numbers way back to the last ones that actually worked:
conda create -y --name cv python
conda activate cv
conda install -y -c apple tensorflow-deps==2.9.0
python -m pip install tensorflow-macos==2.9.2
python -m pip install tensorflow-metal==0.5.1
python --version
conda list|grep -E '(tensorflow|numpy)'
You will get:
Python 3.10.9
numpy 1.22.3 py310hdb36b11_0
numpy-base 1.22.3 py310h5e3e9f0_0
tensorflow-deps 2.9.0 0 apple
tensorflow-estimator 2.9.0 pypi_0 pypi
tensorflow-macos 2.9.2 pypi_0 pypi
tensorflow-metal 0.5.1 pypi_0 pypi
Download and install Conda env:
https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-arm64.sh
chmod +x ~/Downloads/Miniforge3-MacOSX-arm64.sh
sh ~/Downloads/Miniforge3-MacOSX-arm64.sh
source ~/miniforge3/bin/activate
Install the TensorFlow dependencies:
conda install -c apple tensorflow-deps
Install base TensorFlow:
python -m pip install tensorflow-macos
Install base TensorFlow-metal:
python -m pip install tensorflow-metal
Create Conda Environment:
conda create -n tensorflow-env tensorflow
conda activate tensorflow-env
First of all, TensorFlow does not support officially the Mac M1. They don't distribute packages precompiled for the Mac M1 (and its specific arm64 arch), hence the tensorflow-macos package, which is maintained by Apple. TensorFlow distributes, as far as I know, official wheels only for x86 (Linux, Windows, Mac), and the Raspberry PI (arm64).
Apple is using a specific plugin in Tensorflow to make the framework compatible with Metal, the graphic stack of MacOS. To put it in a other way, they are leveraging the PluggableDevice API of Tensorflow to write code that translates the TensorFlow operations to code that the GPU of the M1 understands.
Those two packages contain respectively:
tensorflow-deps the dependencies to run Tensorflow on arm64, i.e python, numpy, grpcio and h5py. This is more of a convenience package, I believe.
tensorflow-metal: a plugin to make tensorflow able to run on metal, the shader API of MacOS (comparable to the low level APIs of Vulkan or DirectX12 on other platforms). You can think of it as a replacement of CUDA, if you are used to run TensorFlow on Nvidia GPUs.
Without the tensorflow-metal package, TensorFlow won't be able to leverage the GPU of the M1, but will still be able to run code on the CPU.
The two answers I got have helped better understand how to install TensorFlow on m1. But I would like share my experience too.
About tensorflow-deps. I do need it, without it pip failed to installed grpcio, and thus actually failed to install tensorflow-macos. When I first asked the question I didn't pay enough attention to output of pip install tensorflow-macos.
About tensorflow-macos package, actually https://blog.tensorflow.org/2020/11/accelerating-tensorflow-performance-on-mac.html has the full information. BTW that article, published 2020-11-18, said "In the near future, we’ll be making updates like this even easier for users to get these performance numbers by integrating the forked version into the TensorFlow master branch." But according to Lescurel's answer it seems they have not.
I didn't know the concept of PluggableDevice (as in Lescurel's), so even when I visited https://github.com/apple/tensorflow_macos I was still confused. Take a look at that article if you do not know that either, basically it will let TensorFlow support new devices.
For 4 articles I listed, "works for M1 too" is the most helpful. It actually explained why I need tensorflow-deps & tensorflow-metal. But part of reasons I did not pay enough attention beforehand were: a) I want to use conda, not miniforge, all these package manager tools scare me a little bit (come from nodejs background, npm, yarn, yarn2, pnmp). The answer from merv also suggested another one mamba, but I think I will pass. b) I don't use homebrew, basically all the articles talking about installing ts on m1 mentioned installing homebrew first. But I use macport, for the reason I mentioned here (again I am bit scared of these these package manager tools)
Using environment.yaml like the one in merv's answer is a reliable way to install tensorflow!
BTW, once I have figured out the whole process of installing tensorflow, install pytorch is a lot easier as pytorch also supports M1 now, check here https://pytorch.org/blog/introducing-accelerated-pytorch-training-on-mac/
Official instructions from Apple are available here.
At the time of writing:
conda create python=3.10.6 --name <NAME>
conda activate <NAME>
conda install -c apple tensorflow-deps
python -m pip install tensorflow-macos
python -m pip install tensorflow-metal
I installed a working version of tensorflow 2.9.2 on my M1 Mac with pip. First, install pyenv and python 3.10.6. Next, pip install tensorflow-metal and finally pip install tensorflow-macos. That's it, no need for tensorflow-deps.
If your model complains about the unavailability of cuDNN and runs slowly, try adjusting your script to enable cuDNN as per tensorflow docs
Worked for me using Python 3.10.8 via Homebrew and following the instructions from Apple but using the instructions for "x86: AMD" instead.
Check Python version:
% which python3.10
/opt/homebrew/bin/python3.10
Create venv, activate it (prompt will change), and update pip:
% python3.10 -m venv ~/py310-tf-metal
% source ~/py310-tf-metal/bin/activate
(py310-tf-metal) % which python
~/py310-tf-metal/bin/python
(py310-tf-metal) % python -m pip install -U pip
...
Successfully installed pip-22.3.1
Install tensorflow-macos:
(py310-tf-metal) % python -m pip install tensorflow-macos
...
Successfully installed MarkupSafe-2.1.1 absl-py-1.3.0 astunparse-1.6.3 cachetools-5.2.0 certifi-2022.9.24 charset-normalizer-2.1.1 flatbuffers-22.11.23 gast-0.4.0 google-auth-2.14.1 google-auth-oauthlib-0.4.6 google-pasta-0.2.0 grpcio-1.50.0 h5py-3.7.0 idna-3.4 keras-2.10.0 keras-preprocessing-1.1.2 libclang-14.0.6 markdown-3.4.1 numpy-1.23.5 oauthlib-3.2.2 opt-einsum-3.3.0 packaging-21.3 protobuf-3.19.6 pyasn1-0.4.8 pyasn1-modules-0.2.8 pyparsing-3.0.9 requests-2.28.1 requests-oauthlib-1.3.1 rsa-4.9 six-1.16.0 tensorboard-2.10.1 tensorboard-data-server-0.6.1 tensorboard-plugin-wit-1.8.1 tensorflow-estimator-2.10.0 tensorflow-macos-2.10.0 termcolor-2.1.1 typing-extensions-4.4.0 urllib3-1.26.13 werkzeug-2.2.2 wheel-0.38.4 wrapt-1.14.1
Install tensorflow-metal:
(py310-tf-metal) % python -m pip install tensorflow-metal
Collecting tensorflow-metal
Downloading tensorflow_metal-0.6.0-cp310-cp310-macosx_12_0_arm64.whl (1.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 MB 25.5 MB/s eta 0:00:00
Requirement already satisfied: six>=1.15.0 in ./Venvs/py310-tf-metal/lib/python3.10/site-packages (from tensorflow-metal) (1.16.0)
Requirement already satisfied: wheel~=0.35 in ./Venvs/py310-tf-metal/lib/python3.10/site-packages (from tensorflow-metal) (0.38.4)
Installing collected packages: tensorflow-metal
Successfully installed tensorflow-metal-0.6.0
Test using the CIFAR training script at the Apple page:
import tensorflow as tf
cifar = tf.keras.datasets.cifar100
(x_train, y_train), (x_test, y_test) = cifar.load_data()
model = tf.keras.applications.ResNet50(
include_top=True,
weights=None,
input_shape=(32, 32, 3),
classes=100,)
loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
model.compile(optimizer="adam", loss=loss_fn, metrics=["accuracy"])
model.fit(x_train, y_train, epochs=5, batch_size=64)
Save above as testcifar.py and run it:
(py310-tf-metal) % python testcifar.py
Downloading data from https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz
169001437/169001437 [==============================] - 3s 0us/step
Metal device set to: Apple M1
systemMemory: 16.00 GB
maxCacheSize: 5.33 GB
2022-11-28 07:58:10.715660: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2022-11-28 07:58:10.715837: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2022-11-28 07:58:14.736843: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz
Epoch 1/5
...
2022-11-28 07:58:21.975675: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.
...
Epoch 5/5
782/782 [==============================] - 206s 264ms/step - loss: 4.0877 - accuracy: 0.1292
I had to downgrade tensorflow to get it to work on Macbook Pro M2:
pip install tensorflow-macos==2.9
pip install tensorflow-metal==0.5.0
I am trying to install Tensorflow Object Detection API, following those instructions. Everything goes well till the moment I have to run python -m pip install within the models/research directory. It starts collecting and installing the various packages, but when it reaches the installation of tf-models-official, it produces the following error :
ERROR: Cannot install object-detection because these package versions have conflicting dependencies.
The conflict is caused by:
tf-models-official 2.7.0 depends on tensorflow-addons
tf-models-official 2.6.1 depends on tensorflow-addons
tf-models-official 2.6.0 depends on tensorflow-addons
tf-models-official 2.5.1 depends on tensorflow-addons
To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict
I did up a bit of digging online, but didn't manage to find anything in the Github repo of tensorflow or other websites. I am using Tensorflow 2.7.0 and Python 3.9.2, on a Linux 64-bit OS. I also tried with Python 3.7.12 but the same error persisted, while manual installation of tensorflow-addons using pip install indicates that no matching distribution was found.
python-3.10.2-amd64: error
python-3.9.10-amd64: ok
python-3.9.2-amd64: ok
python-3.8.10-amd64: ok
python -m pip install --use-feature=2020-resolver tensorflow
'--use-feature=2020-resolver' could be a good solution.
Did you use "tensorflow==2.5.0"?
Don't use specific version like "==2.5.0".
UPDATE
See https://pypi.org/project/tensorflow/#history
tensorflow 2.8.0 (Feb 3, 2022)
I installed tf 2.7.0 and then when I tried to install object detection, tf was reinstalled into tf 2.8.0. So I thought it is weird but I didn't know that tf 2.8.0 doesn't exist yet when you tried to install object detection.
If you try again, tf will be reinstalled automatically into 2.8.0.
I suddenly encounter this situation today, all below output is under the same environment.
pip show tensorflow, the output gives:
Name: tensorflow
Version: 1.14.0rc1
But, if I enter python
python
>>> import tensorflow as tf
>>> print(tf.__version__) . It gives:
'1.13.1'
I am sure I am using the same python under the environment I need with pure terminal or in IDE. This is weird, cuz I used to not having this issue. I just installed few other packages these two days, but I believe they have nothing to do with tensorflow. And you can also verify this through pip show tensorflow or pip list output to see the version is 1.14.0rc1. So why when I actually use python, the tensorflow is not loaded properly?
Solved it by the following:
(under the same environment)
pip uninstall tensorflow (in order to reinstall tensorflow the right version)
pip install tensorflow==1.14.0rc1
Note during installing, there is a piece of info from the terminal:
Installing collected packages: tensorflow
Attempting uninstall: tensorflow
Found existing installation: tensorflow 1.13.1
Uninstalling tensorflow-1.13.1:
Successfully uninstalled tensorflow-1.13.1
So it seems like the tensorflow 1.13.1 has been accidentally installed previously, and whenever in a python program import tensorflow, during searching stage, it hits the tensorflow==1.13.1 before finding the 1.14.0rc1 version. Altho pip list only displays tensorflow==1.14.0rc1 but not tensorflow==1.13.1 might also be due to its internal search or duplicated package resolve mechanism.
I am new to tensorflow. In fact using it because the server code I am writing calls that.
I am using conda to setup the various packages. I did conda install -c anaconda tensorflow-mkl. (Note: I dont have a GPU - using a CPU)
I always get this error:
Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX
The specific line of code where this happens:
tensorflow.contrib.predictor.from_saved_model(path)
On further research, I figured out that this is because the tensorflow package I have does not support this above instruction and needs to have support for the same.
Some questions:
1. How do we ensure the tensorflow package I have does support the above function? Any source from which I can download?
If it is not important, is there a way to suppress this instruction or any errors from it?
Thanks in advance!
You can use conda or pip installations to download the tensorflow that supports cpu. You can use the following commands from your terminal
conda install tensorflow -c anaconda
or
pip install tensorflow==1.13.1
You can use this link if you havent installed pip yet
How to install pip3 on Windows?
Hope this helps..
Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX
This is just a warning.Not an error.
To suppress this warning please add the following lines before your actual code:
import os
os.environ['TF_CPP_MIN_LOG_LEVEL']='2'
import tensorflow as tf
As per the tensorflow official documentation,starting with TensorFlow 1.6, binaries use AVX instructions which may not run on older CPUs.
You can refer the below url for more details on Intel Optimized tensorflow installation:
https://software.intel.com/en-us/articles/intel-optimization-for-tensorflow-installation-guide
Hope this answer your query.Thank you.
I am using ubuntu 16, with python 3, tf-GPU with keras.
I downgraded to tf 1.4 due to cuda errors as explained here
But now I am getting this error
TypeError: softmax() got an unexpected keyword argument 'axis'
Seems that this is an API change in tensorflow and new keras is not suitable for the old tf.
I can't find what is the correct keras version to use with tf 1.4 gpu. What is the correct one?
Keras - Tensorflow versions compatibility is a frequent problem that i have faced many times myself. I am keeping in my bookmarks this compatibility table, with matches of tensorflow and keras versions. It would seem that keras 2.0.8 is compatible with tensorflow 1.4.
If you are using keras exclusively with the tensorflow backend, I would recommend to use the keras implementation found in tf.keras rather than the keras module. That way, you won't scratch your head about possible incompatibilities or bugs (see also that question).
There does not seem to be proper documentation on which Keras version targets which TensorFlow version. The quickest way to solve your problem may be just downgrading Keras one version at a time until you find one that works (or, conversely, upgrading one version at a time from one that you know to work until it breaks). If you find that tedious you can do it as a binary search.
Looking at the releases page, it seems that version 2.0.8 should be compatible with TensorFlow 1.4; it's about a year old already but at least you have an starting point there.
I was able to use the conda package manager to install keras and keras-gpu, with a compatible tensorflow and cuda versions to get past your TypeError: softmax()... error message when I was trying to load the original BERT tensorflow checkpoint using the keras-bert package:
$ conda create -n bert python=3.6
$ conda activate bert
$ conda install keras==2.0.8
$ conda install keras-gpu==2.0.8
$ pip install keras-pos-embd==0.10.0
$ pip install keras-transformer==0.22.0
I am using tensorflow version 1.15.4 and installing Keras version 2.3.1 fixed my problem.
BTW, this page is useful. Find your intended tensorflow version and then click on corresponding Packages and Nvidia Settings.
setuptools==41.0.0
numpy
matplotlib
pandas
pydub
scipy
tensorflow==1.15.4
keras==2.3.1
python_speech_features
praat-parselmouth
pyquaternion