-
Hello,
I encountered an issue while using your code for training. When the number of frames in a sample is greater than 1 but less than 17, num_frames becomes 0, causing ratio_time to be 0, which r…
jike5 updated
3 months ago
-
Hello,
I'm interested in running command line inference using the .ckpt's of the model I trained, but after reading the instructions under `Inference` in `docs/GettingStarted.md` and the outputs o…
-
I use my own data and the stage1 training for v1.2. The loss doesn't go below 1. Do you have any suggestions why this happened?
-
Hi tzapu,
can you help me with configuring local server for websockets. I have installed your new nodes and dont know how to set up the server / Pi and then what do I put in the connection dialog in …
-
It might be an upstream issue.
I'm using Forge with default settings, except for the resolution. However, I’ve tried most of the samplers and schedulers to fix the problem, but without success. Wha…
-
As you know from SD3's paper they used Rectified Flow to make the training and the sampling process faster. However, in the past month a new modification of Rectified Flow named PiecewiseRectifiedFlow…
-
Thank you for this work! I have a simple question about installation. I see that there are empty InstaFlow and piecewise-rectified-flow subfolders, but I don't understand how to properly install those…
-
Do these both refer to the same thing? You cite Liu but there's no RectifiedFlowMatcher class, for example.
-
We are introducing Rectified Flow, a new ODE-based generative model, to this repository (in `RectifiedFlow` branch). The differences between Rectified Flow and the currently used DDPM will result in s…
-
##### SUMMARY
When using the os10_interface role, the flowcontrol.mode does not adhere to YAML syntax rules if you need to configure differing flow control modes.
##### ISSUE TYPE
- Bug …