[bug]: Outpainting gives Error on Mac Studio Max M1, no image and no log info

[DrKucho]

trafficstars

Is there an existing issue for this?

• [X] I have searched the existing issues

OS

macOS

GPU

cuda

VRAM

32 GB SHARED memory

What happened?

I add a image to the unified canvas I select an square area on one side, half in half out, I click on invoke, I can see the image getting created under the coloured noise pictures more and more... but then when its about to finish it all goes away, and I see the Red Error Sign , I go to the log and nothing is there about the error

Screenshots

Additional context

I created tree images successfully on text to image, but it never worked on outputting, I tried with normal 1.5 model and 1.5 inpainting, it also fails but with this one it happens right after clicking Invoke, no noise generation process that I can see

Contact Details

[email protected]

[psychedelicious]

Hi, can you please look at the terminal window (from which you ran the invoke script to start the app) and copy-paste the output from ">> Image generation requested"?

[DrKucho]

See attached

K

On 11 Dec 2022, at 02:12, psychedelicious @.***> wrote:

Hi, can you please look at the terminal window (from which you ran the invoke script to start the app) and copy-paste the output from ">> Image generation requested"?

— Reply to this email directly, view it on GitHub https://github.com/invoke-ai/InvokeAI/issues/1905#issuecomment-1345425661, or unsubscribe https://github.com/notifications/unsubscribe-auth/AZN646NG7SJHXWK3JSDUGCLWMUTAVANCNFSM6AAAAAAS2PFGS4. You are receiving this because you authored the thread.

Traceback (most recent call last): File "/Users/kucho/Downloads/invokeAI/backend/invoke_ai_web_server.py", line 1124, in generate_images self.generate.prompt2image( File "/Users/kucho/Downloads/invokeAI/ldm/generate.py", line 488, in prompt2image results = generator.generate( File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/base.py", line 98, in generate image = make_image(x_T) File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/omnibus.py", line 123, in make_image samples, _ = sampler.sample( File "/Users/kucho/Downloads/invokeAI/installer_files/env/envs/invokeai/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/Users/kucho/Downloads/invokeAI/ldm/models/diffusion/ksampler.py", line 211, in sample attention_map_token_ids = range(1, extra_conditioning_info.tokens_count_including_eos_bos - 1) AttributeError: 'NoneType' object has no attribute 'tokens_count_including_eos_bos'

Image generation requested: {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 15, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_lms', 'seed': 973416146, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 384, 'y': 0, 'width': 256, 'height': 512}, 'init_img': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} ESRGAN parameters: False Facetool parameters: False {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 15, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_lms', 'seed': 973416146, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': <PIL.Image.Image image mode=RGBA size=256x512 at 0x165E41330>, 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 0, 'y': 0, 'width': 256, 'height': 512}, 'init_img': <PIL.Image.Image image mode=RGBA size=256x512 at 0x17717D4B0>, 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} using provided input image of size 256x512 using provided input image of size 256x512 Ksampler using karras noise schedule (steps < 30) Generating: 0%| | 0/1 [00:01<?, ?it/s] 'NoneType' object has no attribute 'tokens_count_including_eos_bos'

Traceback (most recent call last): File "/Users/kucho/Downloads/invokeAI/backend/invoke_ai_web_server.py", line 1124, in generate_images self.generate.prompt2image( File "/Users/kucho/Downloads/invokeAI/ldm/generate.py", line 488, in prompt2image results = generator.generate( File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/base.py", line 98, in generate image = make_image(x_T) File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/omnibus.py", line 123, in make_image samples, _ = sampler.sample( File "/Users/kucho/Downloads/invokeAI/installer_files/env/envs/invokeai/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context return func(*args, **kwargs) File "/Users/kucho/Downloads/invokeAI/ldm/models/diffusion/ksampler.py", line 211, in sample attention_map_token_ids = range(1, extra_conditioning_info.tokens_count_including_eos_bos - 1) AttributeError: 'NoneType' object has no attribute 'tokens_count_including_eos_bos'

Model change requested: stable-diffusion-1.5 Current VRAM usage: 0.00G Offloading inpainting-1.5 to CPU Retrieving model stable-diffusion-1.5 from system RAM cache Current embedding manager terms: * Setting Sampler to k_lms Image generation requested: {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 15, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_lms', 'seed': 384808453, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 384, 'y': 0, 'width': 256, 'height': 512}, 'init_img': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} ESRGAN parameters: False Facetool parameters: False {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 15, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_lms', 'seed': 384808453, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': <PIL.Image.Image image mode=RGBA size=256x512 at 0x1771613C0>, 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 0, 'y': 0, 'width': 256, 'height': 512}, 'init_img': <PIL.Image.Image image mode=RGBA size=256x512 at 0x177161750>, 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} using provided input image of size 256x512 using provided input image of size 256x512 Using recommended DDIM sampler for in>> Using recommended DDIM sampler for in>> Using recommended DDIM sampler for inpainting. target t_enc is 10 steps Generating: 0%| | 0/1 [00:00<?, ?it/s]>> Running DDIMSampler sampling starting at step 5 of 15 (10 new sampling steps) Decoding image: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [00:37<00:00, 3.71s/it] target t_enc is 7 steps████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [00:37<00:00, 3.68s/it] Running DDIMSampler sampling starting at step 3 of 10 (7 new sampling steps) Running DDIMSampler sampling starting at step 3 of 10 (7 new sampling steps) Decoding image: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00, 1>> target t_enc is 7 steps████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [00:37<00:00, 3.68s/it] Running DDIMSampler sampling starting at step 3 of 10 (7 new sampling steps) Decoding image: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00, 1.49s/it] target t_enc is 10 steps██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████>> target t_enc is 10 steps█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00, 1.50s/it] Generating: 0%| | 0/1 [00:57<?, ?it/s] list index out of range

Traceback (most recent call last): File "/Users/kucho/Downloads/invokeAI/backend/invoke_ai_web_server.py", line 1124, in generate_images self.generate.prompt2image( File "/Users/kucho/Downloads/invokeAI/ldm/generate.py", line 488, in prompt2image results = generator.generate( File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/base.py", line 106, in generate image_callback(image, seed, first_seed=first_seed, attention_maps_image=attention_maps_images[-1]) IndexError: list index out of range

Image generation requested: {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 20, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_dpm_2', 'seed': 1199575572, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 384, 'y': 0, 'width': 256, 'height': 512}, 'init_img': 'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAQAAAAIACAYAAABtmr...', 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} ESRGAN parameters: False Facetool parameters: False {'prompt': 'sci fi futuristic town with robot machines 4k art by h r giger', 'iterations': 1, 'steps': 20, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 512, 'sampler_name': 'k_dpm_2', 'seed': 1199575572, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'unifiedCanvas', 'init_mask': <PIL.Image.Image image mode=RGBA size=256x512 at 0x1771612D0>, 'fit': False, 'strength': 0.72, 'invert_mask': False, 'bounding_box': {'x': 0, 'y': 0, 'width': 256, 'height': 512}, 'init_img': <PIL.Image.Image image mode=RGBA size=256x512 at 0x177161EA0>, 'inpaint_width': 384, 'inpaint_height': 704, 'seam_size': 96, 'seam_blur': 16, 'seam_strength': 0.7, 'seam_steps': 10, 'tile_size': 32, 'infill_method': 'tile', 'force_outpaint': False, 'variation_amount': 0} Setting Sampler to k_dpm_2 using provided input image of size 256x512 using provided input image of size 256x512 Using recommended DDIM sampler for inpainting. target t_enc is 14 steps Generating: 0%| | 0/1 [00:00<?, ?it/s]>> Running DDIMSampler sampling starting at step 6 of 20 (14 new sampling steps) Decoding image: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 14/14 [00:51<00:00, 3.69s/it] target t_enc is 7 steps████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 14/14 [00:51<00:00, 3.69s/it] Running DDIMSampler sampling starting at step 3 of 10 (7 new sampling steps) Decoding image: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00, 1.49s/it] target t_enc is 14 steps█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 7/7 [00:10<00:00, 1.50s/it] Generating: 0%| | 0/1 [01:11<?, ?it/s] list index out of range

Traceback (most recent call last): File "/Users/kucho/Downloads/invokeAI/backend/invoke_ai_web_server.py", line 1124, in generate_images self.generate.prompt2image( File "/Users/kucho/Downloads/invokeAI/ldm/generate.py", line 488, in prompt2image results = generator.generate( File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/base.py", line 106, in generate image_callback(image, seed, first_seed=first_seed, attention_maps_image=attention_maps_images[-1]) IndexError: list index out of range

[DrKucho]

I have found that it fails as well doing image o image

{'prompt': 'robot machine futuristic city art by h r giger', 'iterations': 1, 'steps': 8, 'cfg_scale': 7.5, 'threshold': 0, 'perlin': 0, 'height': 512, 'width': 1024, 'sampler_name': 'k_dpm_2', 'seed': 2536543295, 'progress_images': False, 'progress_latents': True, 'save_intermediates': 5, 'generation_mode': 'img2img', 'init_mask': '', 'seamless': False, 'hires_fix': False, 'init_img': <PIL.Image.Image image mode=RGB size=1024x512 at 0x17710A380>, 'strength': 0.72, 'fit': True, 'variation_amount': 0}

using provided input image of size 1024x512 This input is larger than your defaults. If you run out of memory, please use a smaller image. image will be resized to fit inside a box 1024x512 in size. after adjusting image dimensions to be multiples of 64, init image is 1024x512 Ksampler using karras noise schedule (steps < 30) Generating: 0%| | 0/1 [00:00<?, ?it/s]>> Sampling with k_dpm_2 starting at step 3 of 8 (5 new sampling steps) 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 5/5 [01:53<00:00, 22.76s/it] Generating: 0%| | 0/1 [02:05<?, ?it/s] list index out of range

Traceback (most recent call last): File "/Users/kucho/Downloads/invokeAI/backend/invoke_ai_web_server.py", line 1124, in generate_images self.generate.prompt2image( File "/Users/kucho/Downloads/invokeAI/ldm/generate.py", line 488, in prompt2image results = generator.generate( File "/Users/kucho/Downloads/invokeAI/ldm/invoke/generator/base.py", line 106, in generate image_callback(image, seed, first_seed=first_seed, attention_maps_image=attention_maps_images[-1]) IndexError: list index out of range

[psychedelicious]

@DrKucho Thanks, this bug is fixed by 8cbb50c204b7ae24f805574aa59ede1ed2156318, which will be included in the next update.

[github-actions[bot]]

There has been no activity in this issue for 14 days. If this issue is still being experienced, please reply with an updated confirmation that the issue is still being experienced with the latest release.