TensorFlow CPU

2.21.0 · active · verified Sat Apr 11

TensorFlow-cpu is the CPU-optimized variant of TensorFlow, an open-source machine learning framework developed by Google. It is designed for high-performance numerical computation, suitable for training and deploying machine learning models without GPU acceleration. The current version is 2.21.0, and it follows a frequent release cadence, typically aligning with the main TensorFlow project's minor and patch updates every few months.

Warnings

breaking Support for Python 3.9 has been removed starting with TensorFlow 2.21.
Fix: Upgrade to Python 3.10 or newer. The `tensorflow-cpu` package generally requires Python >=3.10.
breaking Keras 3 became the default Keras version starting with TensorFlow 2.16. This may introduce breaking changes if your code relies on Keras 2 APIs.
Fix: Update your code to be compatible with Keras 3. To continue using Keras 2 (tf-keras), install `tf-keras` via `pip install tf-keras~=2.16` and set the environment variable `TF_USE_LEGACY_KERAS=1` before importing TensorFlow.
breaking The `tf.estimator` API has been removed.
Fix: Migrate your code away from `tf.estimator`. If you must use this API, you will need to use TensorFlow 2.15 or an earlier version.
gotcha If a GPU is present on your system, TensorFlow might prioritize it even if you've installed `tensorflow-cpu`. This can lead to unexpected GPU utilization.
Fix: To explicitly restrict TensorFlow to CPU, add `tf.config.set_visible_devices([], 'GPU')` at the beginning of your script, before any other TensorFlow operations.
gotcha Optimizing TensorFlow performance on CPU often requires tuning specific runtime options and environment variables.
Fix: Consider adjusting `tf.config.threading.set_intra_op_parallelism_threads()` and `tf.config.threading.set_inter_op_parallelism_threads()`. For Intel CPUs, setting the environment variable `TF_ENABLE_ONEDNN_OPTS=1` can enable oneDNN optimizations (default on Windows x64 & x86 since TF 2.15).

Install

pip install tensorflow-cpu Install CPU-only TensorFlow

Imports

tensorflow
```
import tensorflow as tf
```

Quickstart

This quickstart code verifies the installed TensorFlow version, checks if CPU devices are recognized, and performs a basic tensor operation to confirm that TensorFlow is functioning correctly on the CPU.

import tensorflow as tf

# Verify TensorFlow version
print(f"TensorFlow Version: {tf.__version__}")

# Verify CPU device is recognized
cpu_devices = tf.config.list_physical_devices('CPU')
print(f"CPU Devices: {cpu_devices}")

if cpu_devices:
    print(f"TensorFlow is configured to use CPU: {cpu_devices[0].name}")
    # Perform a simple operation to confirm functionality
    a = tf.constant([[1.0, 2.0], [3.0, 4.0]])
    b = tf.constant([[1.0, 1.0], [1.0, 1.0]])
    print(f"Result of a + b:\n{a + b}")
else:
    print("No CPU devices found by TensorFlow. Check installation.")

view raw JSON →