Failed assertion `[MPSNDArray initWithDevice:descriptor:] Error: total bytes of NDArray > 2**32

[timojaask]

I wanted to try this out, so I downloaded the woman with a pearl earring image from the README.md and ran the example command:

imagine --model SD-2.0-depth --init-image girl_with_a_pearl_earring.jpg --init-image-strength 0.05  "professional headshot photo of a woman with a pearl earring" -r 4 -w 1024 -h 1024 --steps 50

Which results in the following output:

 % imagine --model SD-2.0-depth --init-image girl_with_a_pearl_earring.jpg --init-image-strength 0.05  "professional headshot photo of a woman with a pearl earring" -r 4 -w 1024 -h 1024 --steps 50

🤖🧠 imaginAIry received 1 prompt(s) and will repeat them 4 times to create 4 images.
Generating 🖼  1/4: "professional headshot photo of a woman with a pearl earring" 1024x1024px negative-prompt:"Ugly, duplication, duplicates, mutilation, deformed, mutilated, mutation, twisted body, disfigured, bad anatomy, out of frame, extra fingers, mutated hands, poorly drawn hands, extra limbs, malformed limbs, missing arms, extra arms, missing legs, extra legs, mutated hands, extra hands, fused fingers, missing fingers, extra fingers, long neck, small head, closed eyes, rolling eyes, weird eyes, smudged face, blurred face, poorly drawn face, mutation, mutilation, cloned face, strange mouth, grainy, blurred, blurry, writing, calligraphy, signature, text, watermark, bad art," seed:104916828 prompt-strength:7.5 steps:50 sampler-type:k_dpmpp_2m
Loading model /Users/timojaaskelainen/.cache/huggingface/hub/models--stabilityai--stable-diffusion-2-depth/snapshots/d41a0687231847e8bd55f43fb1f576afaeefef19/512-depth-ema.ckpt onto mps:0 backend...
  0%|                                                                                                                                                                                     | 0/46 [00:00<?, ?it/s]/AppleInternal/Library/BuildRoots/5b8a32f9-5db2-11ed-8aeb-7ef33c48bc85/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:724: failed assertion `[MPSNDArray initWithDevice:descriptor:] Error: total bytes of NDArray > 2**32'
zsh: abort      imagine --model SD-2.0-depth --init-image girl_with_a_pearl_earring.jpg  0.05
timojaaskelainen@Timos-M1-MBP ~ % /Users/timojaaskelainen/.pyenv/versions/3.10.9/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 what this is about, but looks like some array allocation goes over the framework limits?
Some other examples that I've tried are working without errors.

I'm running macOS Ventura 13.1, Python 3.10.9, on MacBook Pro, M1 Max, 64 GB RAM.

[brycedrennan]

hmm that is strange. Thanks for all the information. will investigate

[brycedrennan]

I'm at 35% complete on my Macbook Pro M1 Max 32 GB, Ventura 13.0.1. Says it's gonna take 20 minutes to complete. I certainly ran this originally on the Desktop with a 2080TI graphics card and got it done in a few seconds.

[brycedrennan]

yeah worked for me. took 26 minutes. can you provide the output of pip freeze?

[timojaask]

Sure:

% pip freeze
aiohttp==3.8.3
aiosignal==1.3.1
antlr4-python3-runtime==4.9.3
async-timeout==4.0.2
attrs==22.2.0
certifi==2022.12.7
charset-normalizer==2.1.1
click==8.1.3
click-shell==2.1
contourpy==1.0.7
cycler==0.11.0
diffusers==0.11.1
einops==0.6.0
facexlib==0.2.5
fairscale==0.4.13
filelock==3.9.0
filterpy==1.4.5
fonttools==4.38.0
frozenlist==1.3.3
fsspec==2023.1.0
ftfy==6.1.1
huggingface-hub==0.11.1
idna==3.4
imageio==2.24.0
imaginAIry==8.0.0
importlib-metadata==6.0.0
kiwisolver==1.4.4
kornia==0.6.9
lightning-utilities==0.5.0
llvmlite==0.39.1
matplotlib==3.6.3
multidict==6.0.4
numba==0.56.4
numpy==1.23.5
omegaconf==2.3.0
open-clip-torch==2.9.3
opencv-python==4.7.0.68
packaging==23.0
Pillow==9.4.0
protobuf==3.20.3
psutil==5.9.4
pyparsing==3.0.9
python-dateutil==2.8.2
pytorch-lightning==1.9.0
PyYAML==6.0
regex==2022.10.31
requests==2.28.2
safetensors==0.2.8
scipy==1.10.0
semantic-version==2.10.0
sentencepiece==0.1.97
setuptools-rust==1.5.2
six==1.16.0
timm==0.6.12
tokenizers==0.13.2
torch==1.13.1
torchdiffeq==0.2.3
torchmetrics==0.11.0
torchvision==0.14.1
tqdm==4.64.1
transformers==4.25.1
typing_extensions==4.4.0
urllib3==1.26.14
wcwidth==0.2.6
yarl==1.8.2
zipp==3.11.0

[brycedrennan]

Please confirm this is fixed in just release 8.0.4