You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: site/en/install/pip.md
+24-17Lines changed: 24 additions & 17 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -231,12 +231,29 @@ The following NVIDIA® software are only required for GPU support.
231
231
232
232
**Note:** Do not install TensorFlow with `conda`. It may not have the latest stable version. `pip` is recommended since TensorFlow is only officially released to PyPI.
233
233
234
-
### 5. Virtual environment configuration
234
+
### 6. Verify the installation
235
235
236
-
You can skip this section if you only run TensorFlow on the CPU.
237
-
238
-
Note: Symbolic links are only necessary in case the intended way doesn't work, i.e. the components aren't being
239
-
detected, and/or conflict with the existing system CUDA installation.
236
+
Verify the CPU setup:
237
+
238
+
```bash
239
+
python3 -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"
240
+
```
241
+
242
+
If a tensor is returned, you've installed TensorFlow successfully.
243
+
244
+
Verify the GPU setup:
245
+
246
+
```bash
247
+
python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
248
+
```
249
+
250
+
If a list of GPU devices is returned, you've installed TensorFlow
251
+
successfully. **If not continue to the next step**.
0 commit comments