Benjamin Gregg
Benjamin Gregg
This is not an option still in current controlnet train script
turned out to be problem when using torch dynamo to fix pip install vision_aided_loss dont use torch dynamo
Mind if I bump this feature? looking for a alternative to vdd and parsecs implementation to be able use rtx HDR from browser streaming to tv. This project looks promising...
@nv-ajaus [NvRemixBridge.exe_20231019_202915.dmp](https://github.com/NVIDIAGameWorks/rtx-remix/files/13173077/NvRemixBridge.exe_20231019_202915.dmp) EDIT: Above link is BattleFront 2 .dmp Below is .dmp refrenced for BattleFront 1 [NvRemixBridge.exe_20231019_204342.dmp](https://github.com/NVIDIAGameWorks/rtx-remix/files/13180409/NvRemixBridge.exe_20231019_204342.dmp)
> If you change your mind and need some files or have some files up for testing im gladly doing it in case it could help to support HDR Same...
Adding onto Mesh / Material Properties or if another route is thought of Once we have the ability to talk to meshes not created by API, or if we can...
Thanks to help from Pete This was fixed by capturing scene as usda and editing `string lightspeed_game_name = "The Hobbit\x99"` to `string lightspeed_game_name = "The Hobbit"`
Having this too
@sayakpaul i tried newest commit and now flux training will crash before training ``` 09/28/2024 20:53:43 - INFO - __main__ - Initializing controlnet weights from transformer 09/28/2024 20:53:54 - INFO...
@PromeAIpro Training no longer crashes on start but OOM is still happening even with one 512x512 val image I follow flux readme to the T here is readout and launch...