Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
docs.it4i.cz
Manage
Activity
Members
Labels
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Container registry
Model registry
Operate
Environments
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
SCS
docs.it4i.cz
Commits
6140f612
Commit
6140f612
authored
2 years ago
by
Jan Siwiec
Browse files
Options
Downloads
Patches
Plain Diff
Update tensorflow.md
parent
62568cb5
No related branches found
No related tags found
1 merge request
!411
Add section about using TensorFlow with GPUs
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
docs.it4i/software/machine-learning/tensorflow.md
+5
-2
5 additions, 2 deletions
docs.it4i/software/machine-learning/tensorflow.md
with
5 additions
and
2 deletions
docs.it4i/software/machine-learning/tensorflow.md
+
5
−
2
View file @
6140f612
...
@@ -4,14 +4,16 @@ TensorFlow (TF) is an open-source software library which can compile tensor oper
...
@@ -4,14 +4,16 @@ TensorFlow (TF) is an open-source software library which can compile tensor oper
very quickly on both CPUs and GPUs. It is often used as a backend for machine learning libraries
very quickly on both CPUs and GPUs. It is often used as a backend for machine learning libraries
and models.
and models.
We heavily recommend the usage of
`Tensor
f
low 2.x`
. TensorFlow 1 has been long deprecated and it
We heavily recommend the usage of
`Tensor
F
low 2.x`
. TensorFlow 1 has been long deprecated and it
will probably be difficult to make it run on GPUs on our clusters.
will probably be difficult to make it run on GPUs on our clusters.
## Installation
## Installation
For TensorFlow to work with GPUs, you have to use several libraries (CUDA, cuDNN, NCCL etc.)
For TensorFlow to work with GPUs, you have to use several libraries (CUDA, cuDNN, NCCL etc.)
with versions that are compatible together.
with versions that are compatible together.
You can load the correct modules with the following command:
You can load the correct modules with the following command:
```
console
```
console
$
ml TensorFlow
$
ml TensorFlow
```
```
...
@@ -19,6 +21,7 @@ $ ml TensorFlow
...
@@ -19,6 +21,7 @@ $ ml TensorFlow
If you want to upgrade the TensorFlow version used in this package or install additional Python
If you want to upgrade the TensorFlow version used in this package or install additional Python
modules, you can simply create a virtual environment and install a different TensorFlow version
modules, you can simply create a virtual environment and install a different TensorFlow version
inside it:
inside it:
```
console
```
console
$
python3
-m
venv venv
$
python3
-m
venv venv
$
source
venv/bin/activate
$
source
venv/bin/activate
...
@@ -53,7 +56,7 @@ Using the available `TensorFlow` module should make sure that these modules will
...
@@ -53,7 +56,7 @@ Using the available `TensorFlow` module should make sure that these modules will
### Selecting GPUs
### Selecting GPUs
You can select how many and which (N
vidia
) GPUs will be used by TensorFlow with the
You can select how many and which (N
VIDIA
) GPUs will be used by TensorFlow with the
`CUDA_VISIBLE_DEVICES`
environment variable.
`CUDA_VISIBLE_DEVICES`
environment variable.
```
console
```
console
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment