CODEWITHSUNDEEP
pythontensorflow
roudan
roudan

Reputation: 4200

tensorflow GPU is not available

I followed through Jeff Heaton's youtube video to install tensorflow with GPU. I have Eva nivida GTX 1660 Ti. After that, it show GPU is not available. any suggestion if something goes wrong? Thanks

https://www.youtube.com/watch?v=PnK1jO2kXOQ&list=PLrx-l85_irc5IQhZ-yOLq6NjV7GajbT0h&index=8&t=86s https://github.com/jeffheaton/t81_558_deep_learning/blob/master/install/tensorflow-install-jul-2020.ipynb

# What version of Python do you have?
# What version of Python do you have?
import sys

import tensorflow.keras
import pandas as pd
import sklearn as sk
import tensorflow as tf

print(f"Tensor Flow Version: {tf.__version__}")
print(f"Keras Version: {tensorflow.keras.__version__}")
print()
print(f"Python {sys.version}")
print(f"Pandas {pd.__version__}")
print(f"Scikit-Learn {sk.__version__}")
gpu = len(tf.config.list_physical_devices('GPU'))>0
print("GPU is", "available" if gpu else "NOT AVAILABLE")

Tensor Flow Version: 2.3.0
Keras Version: 2.4.0

Python 3.8.5 (default, Sep  3 2020, 21:29:08) [MSC v.1916 64 bit (AMD64)]
Pandas 1.2.0
Scikit-Learn 0.23.2
GPU is NOT AVAILABLE

Upvotes: 1

Views: 1407

Answers (2)

DapperDuck
DapperDuck

Reputation: 2876

Tensorflow requires CUDA to be installed to utilize a GPU. Unfortunately, the 1660 Ti is not a
CUDA-enabled GPU. This means that you can't use tensorflow with GPU acceleration.

If you want to use a gpu, you can use a free one with Google Colab or Kaggle.

Upvotes: 2

Armadillan
Armadillan

Reputation: 570

Tensorflow uses CUDA, which the 1660 Ti does not support. Here is a list of all supported gpus: https://developer.nvidia.com/cuda-gpus

Unfortunately, you can not run tensorflow with gpu-acceleration on that machine. :(

Upvotes: 1

Related Questions