[Open-Sora-Plan v1.3.0]: inference and training #717
Conversation
This file contains 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
Collaborator
wtomin
commented
Oct 29, 2024
•
wtomin
commented
- vae inference & training under pynative mode;
- dit inference & training under pynative mode;
- dynamic resolution dataloader;
- prompt refiner inference.
wtomin
requested review from
CaitinZhao,
SamitHuang and
zhanghuiyao
as code owners
wtomin
force-pushed
the
op-v1.3-update
branch
4 times, most recently
from
d2bfac6 to
4f29b01
Compare
wtomin
force-pushed
the
op-v1.3-update
branch
from
4f29b01 to
609abc9
Compare
wtomin
force-pushed
the
op-v1.3-update
branch
2 times, most recently
from
c068a2f to
1b79586
Compare
wtomin
force-pushed
the
op-v1.3-update
branch
from
a3ea865 to
a196345
Compare
wtomin
force-pushed
the
op-v1.3-update
branch
13 times, most recently
from
2e7783b to
de77902
Compare
This reverts commit 410e689.
wtomin
force-pushed
the
op-v1.3-update
branch
from
83aa679 to
31bdcfd
Compare
wtomin
changed the title
wtomin
force-pushed
the
op-v1.3-update
branch
from
f6d4d2f to
5a87ff4
Compare
wtomin
force-pushed
the
op-v1.3-update
branch
from
5a87ff4 to
3452269
Compare
SamitHuang
reviewed
Dec 24, 2024
examples/opensora_pku/README.md
Outdated
|
|
||
| #### Performance | ||
|
|
||
| The training performance are tested on ascend 910* with mindspore 2.3.1 graph mode. The results are as follows. | ||
| We evaluated the training performance on Ascend NPUs. All experiments are running in PYNATIVE mode. The results are as follows. |
| | OpenSoraT2V-ROPE-L-122 | 8 | 3 | 1 | 29 | 640x480 | 6mins | zero2 | ON | ON | O0 | 3.68 | 63.04 | | ||
| | OpenSoraT2V-ROPE-L-122 | 8 | 4 | 1 | 29 |1280x720 | 10mins | zero2 + SP(sp_size=8) | OFF | ON | O0 | 4.32 | 6.71 | | ||
| | OpenSoraT2V-ROPE-L-122 | 8 | 5 | 1 | 93 | 1280x720 | 15mins | zero2 + SP(sp_size=8) | ON | ON | O0 | 24.40 | 3.81 | | ||
| | model name | cards | stage | batch size (global) | video size | Paramllelism |recompute |data sink | jit level| step time | train imgs/s | |
There was a problem hiding this comment.
Explain what train imgs/s is and how it's computed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.