Disable autocast for cpu to fix error. Remove unused precision arg. #518
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
|
This is tangential to this PR, but does this and other work in this repo mean that generation works on Intel CPUs (obviously very slowly), or is more work needed? |
Contributor
Author
|
This PR makes generation work for (intel) CPUs, yes. ~5mins for -s30 -Addim
for an i7 few year old laptop.
…On Mon, Sep 12, 2022 at 7:21 PM tildebyte ***@***.***> wrote:
This is tangential to this PR, but does this and other work in this repo
mean that generation works on Intel CPUs (obviously very slowly), or is
more work needed?
—
Reply to this email directly, view it on GitHub
<#518 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACJAB4AU6OSBCYCVSKHSSDV55KCNANCNFSM6AAAAAAQKRK42Y>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float Fix buggy device handling in model.py. Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
Collaborator
lstein
approved these changes
Sep 12, 2022
Collaborator
lstein
left a comment
lstein
left a comment
There was a problem hiding this comment.
Looks good to me. Tested and didn't find any problems on CUDA.
Contributor
Author
|
Only orig_scripts/ may have questionable uses of autocast left. I do have an idea for #526 that I'll comment there. |
afiaka87
pushed a commit
to afiaka87/lstein-stable-diffusion
that referenced
this pull request
Sep 19, 2022
…nvoke-ai#518) When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float Fix buggy device handling in model.py. Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
afiaka87
pushed a commit
to afiaka87/lstein-stable-diffusion
that referenced
this pull request
Sep 19, 2022
…nvoke-ai#518) When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float Fix buggy device handling in model.py. Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
afiaka87
pushed a commit
to afiaka87/lstein-stable-diffusion
that referenced
this pull request
Sep 21, 2022
…nvoke-ai#518) When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float Fix buggy device handling in model.py. Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
brycedrennan
added a commit
to brycedrennan/imaginAIry
that referenced
this pull request
Nov 13, 2022
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
brycedrennan
added a commit
to brycedrennan/imaginAIry
that referenced
this pull request
Nov 13, 2022
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
brycedrennan
added a commit
to brycedrennan/imaginAIry
that referenced
this pull request
Nov 13, 2022
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
austinbrown34
pushed a commit
to cognidesign/InvokeAI
that referenced
this pull request
Dec 30, 2022
…nvoke-ai#518) When running on just cpu (intel), a call to torch.layer_norm would error with RuntimeError: expected scalar type BFloat16 but found Float Fix buggy device handling in model.py. Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.
mattstern31
added a commit
to mattstern31/imagin-AIry-Python
that referenced
this pull request
Nov 11, 2023
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
superlucky19971023
added a commit
to superlucky19971023/imaginAIry
that referenced
this pull request
Aug 18, 2024
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
PyPilot-ai
pushed a commit
to PyPilot-ai/PyPilot-ai
that referenced
this pull request
Apr 9, 2025
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
PyPilot-ai
added a commit
to PyPilot-ai/PyPilot-ai
that referenced
this pull request
Apr 13, 2025
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
tune-project1
added a commit
to tune-project1/Viziax
that referenced
this pull request
Apr 13, 2025
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
PyPilot-ai
added a commit
to PyPilot-ai/surveylogic
that referenced
this pull request
Apr 14, 2025
autocast not supported on CPU pytorch/pytorch#55374 invoke-ai/InvokeAI#518
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When running on just cpu (intel), a call to torch.layer_norm would error with "RuntimeError: expected scalar type BFloat16 but found Float"
Fix buggy device handling in model.py.
Tested with scripts/dream.py --full_precision on just cpu on intel laptop. Works but slow at ~10s/it.