-
Notifications
You must be signed in to change notification settings - Fork 66
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat(tuner): add gpu for paddle and tf #121
Conversation
paddlepaddle-gpu | ||
torch | ||
torchvision | ||
scipy |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
scipy??
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this is used in creating easy data for overfitting test if i remember correctly cc @tadejsv
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes exactly, so I have a one-liner to calculate distances
.github/requirements-cicd-gpu.txt
Outdated
@@ -0,0 +1,6 @@ | |||
numpy |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
not necessary, as base dep is jina and jina includes numpy already
@@ -128,24 +128,35 @@ def fit( | |||
|
|||
optimizer = tf.keras.optimizers.RMSprop(learning_rate=0.01) | |||
|
|||
if device == 'cuda': |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
do we have the same device name across all frameworks, what @tadejsv using in pytorch?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yes we agreed to use cuda
and cpu
.
for tensorflow, since it's session based, i can not verify it's using gpu for training in an explicit way, however, printing los on gpu machine shows that it's using gpu: