Google Colab RAM usage being 100% #600
Unanswered
Mainakcris7
asked this question in
Q&A
Replies: 2 comments 1 reply
-
What about the batch size? Is it similar to what on the tutorial which is 32? Mr. Dbourke also taught about computation optimization using |
Beta Was this translation helpful? Give feedback.
1 reply
-
Try using 2 things
1) use smaller batch of 16 or 8
2) use garbage collection- gc.collect()
…On Mon, Oct 30, 2023, 6:56 AM Ammar Azman ***@***.***> wrote:
What about the batch size? Is it similar to what on the tutorial which is
32? Mr. Dbourke also taught about computation optimization using
.batch(batch_size=32).prefetch(buffer_size=tf.data.AUTOTUNE) too in order
to avoid the spike usage on the RAM. Did you follow?
—
Reply to this email directly, view it on GitHub
<#600 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AWDNWCFPKDSZGAXXVAH7TZ3YB364DAVCNFSM6AAAAAA6GRUIECVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TIMJYHAYDM>
.
You are receiving this because you are subscribed to this thread.Message
ID:
<mrdbourke/tensorflow-deep-learning/repo-discussions/600/comments/7418806@
github.com>
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I was trying mixed precision and prefetching techniques as per as Section 9 (Food Vision) on CIFAR10 dataset. I followed all the steps what were showed in the videos. But still the RAM usage being 12.26 GB out of 12.68 GB, after just processing the data for half (or, maybe less) epoch. Why it is so ??
Any solution please????
Beta Was this translation helpful? Give feedback.
All reactions