Not sure it's a bug, maybe due to some mac pytorch support issues.
Every time I start to train, it could last about 1 min, than python craches.
Below finds the the log.
I was using C0nsumption's workflow
load motion module from /Users/gabrie/ComfyUI/custom_nodes/ComfyUI-AnimateDiff-Evolved/models/v3_sd15_mm.ckpt
load domain lora from /Volumes/Thunder_1T/AI_Models/LoRAs/AnimateDiff/v3_sd15_adapter.ckpt
Using Lion optimizer for training
Using LoRA Version: cloneofsimo
Lora successfully injected into UNet3DConditionModel.
Using LoRA Version: cloneofsimo
Lora successfully injected into UNet3DConditionModel.
initialize scaler
global_step: 0
Steps: 0%| | 0/50 [00:00<?, ?it/s]/AppleInternal/Library/BuildRoots/4e1473ee-9f66-11ee-8daf-cedaeb4cabe2/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:120: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: subRange.start (16) is not less than length of dimension[2] (16)'
/Users/gabrie/Desktop/Comfy.command: line 12: 1674 Abort trap: 6 python main.py --force-fp16
/Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
warnings.warn('resource_tracker: There appear to be %d '
Not sure it's a bug, maybe due to some mac pytorch support issues.
Every time I start to train, it could last about 1 min, than python craches. Below finds the the log. I was using C0nsumption's workflow
load motion module from /Users/gabrie/ComfyUI/custom_nodes/ComfyUI-AnimateDiff-Evolved/models/v3_sd15_mm.ckpt load domain lora from /Volumes/Thunder_1T/AI_Models/LoRAs/AnimateDiff/v3_sd15_adapter.ckpt Using Lion optimizer for training Using LoRA Version: cloneofsimo Lora successfully injected into UNet3DConditionModel. Using LoRA Version: cloneofsimo Lora successfully injected into UNet3DConditionModel. initialize scaler global_step: 0 Steps: 0%| | 0/50 [00:00<?, ?it/s]/AppleInternal/Library/BuildRoots/4e1473ee-9f66-11ee-8daf-cedaeb4cabe2/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:120: failed assertion `[MPSNDArrayDescriptor sliceDimension:withSubrange:] error: subRange.start (16) is not less than length of dimension[2] (16)' /Users/gabrie/Desktop/Comfy.command: line 12: 1674 Abort trap: 6 python main.py --force-fp16 /Library/Frameworks/Python.framework/Versions/3.10/lib/python3.10/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown warnings.warn('resource_tracker: There appear to be %d '