Comments (10)
@moonnnpie in the ptp_util.py
, omit extra_set_kwargs = {"offset": 1}
, and change model.scheduler.set_timesteps(num_inference_steps, **extra_set_kwargs)
to model.scheduler.set_timesteps(num_inference_steps)
. This error is because in the previous diffusers, scheduler's set_timesteps
function has a param offset
.
from prompt-to-prompt.
ok fixed my case by reversing the advice - renamed the stuff in register_attention_control
function.
context
=> encoder_hidden_states
, mask
=> attention_mask
, CrossAttention
=> AttentionBlock
from prompt-to-prompt.
On latest 0.13.1 diffusers version I had to:
rename context
=> encoder_hidden_states
, mask
=> attention_mask
, self.reshape_heads_to_batch_dim
-> self.head_to_batch_dim
, self.reshape_batch_dim_to_heads
-> self.batch_to_head_dim
in register_attention_control
and it started working correctly
from prompt-to-prompt.
i also use 0.13.1 diffusers version but got a problem
TypeError: PNDMScheduler.set_timesteps() got an unexpected keyword argument 'offset'
how should i fix it
from prompt-to-prompt.
@eps696 I followed your advice, but found if change CrossAttention
to AttentionBlock
, the attention_store would be none. So in my case, I renamed context
and mask
, and stay with CrossAttention
.
from prompt-to-prompt.
hello, could you explain this issue a bit more?
i believe i have run into it in some "updated" form, getting the error forward() got an unexpected keyword argument 'encoder_hidden_states'
when running register_attention_control
on null-inversion - yet i don't quite catch what and where should be changed to avoid that. do you mean changing the original diffusers code?
from prompt-to-prompt.
@liangbingzhao @bonlime you are definitely right, i didn't notice that changing the block effectively disabled main workflow. yet the CrossAttention
refactoring after 0.11.1 was quite gross, so i'm afraid more issues might appear
from prompt-to-prompt.
@liangbingzhao thanks a lot
from prompt-to-prompt.
Hi, all, I have tried to port to diffusers==0.14.0, there is no runing error, but the result is actually bad...
https://github.com/JiauZhang/prompt-to-prompt for now, I just simple replace attention map.
Could anyone give me some advice?
from prompt-to-prompt.
Hi, all, I have tried to port to diffusers==0.14.0, there is no runing error, but the result is actually bad... https://github.com/JiauZhang/prompt-to-prompt for now, I just simple replace attention map. Could anyone give me some advice?
Is it solved?
from prompt-to-prompt.
Related Issues (20)
- "TypeError: 'NoneType' object is not callable" upon model.unet() call
- How does this method edit the actions of an object? HOT 1
- PNDMScheduler has no attribute time steps?
- The code doesn't run HOT 1
- support for resolutions lower than 512? HOT 2
- Inconsistency between timestep and noise level in Null-Text Inversion? HOT 1
- Where is the code that visualizes by time step? HOT 1
- Can't load a fine-tuned model due to using an old version of diffusers
- On the non-convergence of null text
- Is training code available? HOT 1
- Does it support SDXL? HOT 2
- When I use Diffusers==0.21.0, p2p seems to generate the same as w/o p2p. Code needs to be upgraded.
- bad results when changing the clothes's color HOT 1
- code for user-defined mask HOT 3
- Can Null text inversion apply to Stable diffusion 2.1? HOT 1
- If I want to add null text inversion to the training process to maintain the feature of the edited image, how can I achieve this?
- code about Delta Denoising Score
- Can we give image as input? HOT 1
- wrong DDIM inversion step HOT 1
- The setting of DDIMScheduler.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from prompt-to-prompt.