Error occurred at Ksampler while running comfy with --force-fp16 command on Mac M1

Hi,
For me, Comfy is generating images when it is ran using the command python main.py . But when I ran it with python main.py --force-fp16 , it generates error at Ksampler saying "upsample_nearest2d_channels_last" not implemented for 'Half' .

I had similar issue with a1111, where by editing --no-half in Commandline args, it got solved. But with comfy, I am unable to find the arg line command or user.sh file. I don't know coding and new to comfyui. I have installed anaconda and PyTorch nightly using pip3 command and followed every damn step. Everything is up to date as well.

Help me!
Sushrutha D

> Here's what happens in the terminal-

When I run python main.py

Last login: Thu Dec 7 21:17:27 on ttys000
(base) sushruthad@Sushruthas-MacBook-Air ~ % cd /Users/sushruthad/Stable-Diffusion/ComfyUI
(base) sushruthad@Sushruthas-MacBook-Air ComfyUI % python main.py
** ComfyUI start up time: 2023-12-07 21:20:43.482025

Prestartup times for custom nodes:
0.0 seconds: /Users/sushruthad/Stable-Diffusion/ComfyUI/custom_nodes/ComfyUI-Manager

Total VRAM 16384 MB, total RAM 16384 MB
/Users/sushruthad/anaconda3/lib/python3.11/site-packages/transformers/utils/generic.py:260: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead.
torch.utils._pytree._register_pytree_node(
Set vram state to: SHARED
Device: mps
VAE dtype: torch.float32
Using sub quadratic optimization for cross attention, if you have memory or speed issues try using: --use-split-cross-attention
Adding extra search path checkpoints /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/Stable-diffusion
Adding extra search path configs /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/Stable-diffusion
Adding extra search path vae /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/VAE
Adding extra search path loras /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/Lora
Adding extra search path loras /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/LyCORIS
Adding extra search path upscale_models /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/ESRGAN
Adding extra search path upscale_models /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/RealESRGAN
Adding extra search path upscale_models /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/SwinIR
Adding extra search path embeddings /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/embeddings
Adding extra search path hypernetworks /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/hypernetworks
Adding extra search path controlnet /Users/sushruthad/Stable-Diffusion/stable-diffusion-webui/models/ControlNet
### Loading: ComfyUI-Manager (V1.6.4)
### ComfyUI Revision: 1789 [fbdb14d4] | Released on '2023-12-06'

Import times for custom nodes:
0.0 seconds: /Users/sushruthad/Stable-Diffusion/ComfyUI/custom_nodes/ComfyUI_NestedNodeBuilder
0.0 seconds: /Users/sushruthad/Stable-Diffusion/ComfyUI/custom_nodes/efficiency-nodes-comfyui
0.1 seconds: /Users/sushruthad/Stable-Diffusion/ComfyUI/custom_nodes/ComfyUI-Manager

Starting server

To see the GUI go to: http://127.0.0.1:8188
got prompt
model_type EPS
adm 2816
Using split attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using split attention in VAE
missing {'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_l.logit_scale'}
left over keys: dict_keys(['cond_stage_model.clip_l.transformer.text_model.embeddings.position_ids'])
Requested to load SDXLClipModel
Loading 1 new model
Requested to load SDXL
Loading 1 new model
0%| | 0/20 [00:00