Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

XYZ Prompt: NotImplementedError: Cannot copy out of meta tensor; no data! #173

Open
windpiligrim opened this issue Oct 23, 2024 · 0 comments
Open

XYZ Prompt: NotImplementedError: Cannot copy out of meta tensor; no data! #173

windpiligrim opened this issue Oct 23, 2024 · 0 comments
Assignees
@wkpark
Labels
bug Something isn't working

Comments

@windpiligrim
Copy link

windpiligrim commented Oct 23, 2024 •
edited
Loading

Hello. This is my first bug report ever :D
Please let me know if this is not relevant or any other issues.

I have recently updated Model Mixer and my routine of playing with Dare merges broke. Here is what I used to do:

  1. Set up Model A to a pony model,
  2. Use GPU, Fast Rebasin
  3. Set Model B to a pony model
  4. Set Dare merge (Lambda = 1.0)
  5. Use MBW - ALL
  6. Enable XYZ:
  7. X: Model Mixer Model B (2-3 models)
  8. Y: Model Mixer MBW Alpha B (For example = "1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,0,1,0.5,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,0,1,0.3,1,1,1,1,1,1,1,1","0,1,1,1,1,1,1,1,1,0,1,0,1,1,1,1,1,1,1,1")
  9. Z: 2-3 Styles with prompts to test the merges

Here is the log of where it breaks:

Console Log

debugs = ['elemental merge']
use_extra_elements = True, ?it/s]

  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1
    config hash = 552bbe35376b043403c508fac3e8107c7c3173709c786d4ae6e605d31dd7481a
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['humana_v1.safetensors [62f6b5f0a5]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['OUT00']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|███████████████████████████████████████████████████████████████████| 218/218 [00:00<00:00, 851.34it/s]
    Check uninitialized support mbw presets #2/3: 100%|██████████████████████████████████████████████████| 218/218 [00:00<00:00, 220062.16it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 36100.98it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 0.6s (prepare: 0.1s, merging: 0.5s).
  • loading scripts.patches...
  • lora patch
  • update UNet block output_blocks.0.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:33<00:00, 2.66s/it]
    debugs = ['elemental merge']█████████████████████████ | 35/70 [01:40<01:34, 2.70s/it]
    use_extra_elements = True
  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1
    config hash = c944033c76af9bc58de90a62672e9b3e703614557f04b788a679caa202e67798
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['humana_v1.safetensors [62f6b5f0a5]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['OUT00']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 0.7, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|████████████████████████████████████████████████████████████████████| 218/218 [00:05<00:00, 42.56it/s]
    Check uninitialized support mbw presets #2/3: 100%|██████████████████████████████████████████████████| 218/218 [00:00<00:00, 218536.87it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 31510.74it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 5.6s (prepare: 0.2s, merging: 5.4s).
  • loading scripts.patches...
  • lora patch
  • update UNet block output_blocks.0.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:33<00:00, 2.66s/it]
    Total progress: 100%|██████████████████████████████████████████████████████████████████| 70/70 [03:38<00:00, 3.13s/it]
    X/Y/Z plot will create 16 images on 1 2x2 grid; 4 images per cell. (Total steps to process: 140)03:38<00:00, 2.70s/it]
    debugs = ['elemental merge']
    use_extra_elements = True, ?it/s]
  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1
    config hash = ada4d1f16841761a82b7b8ede66aa354f9008f3c40c0510558ca546910a65bd3
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['opticon_v10.safetensors [cd7397d079]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['IN00', 'IN01', 'IN02', 'IN03', 'IN04', 'IN05', 'IN06', 'IN07', 'M00', 'OUT00', 'OUT01', 'OUT02', 'OUT03', 'OUT04', 'OUT05', 'OUT06', 'OUT07', 'OUT08']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|█████████████████████████████████████████████████████████████████| 1460/1460 [00:01<00:00, 875.32it/s]
    Check uninitialized support mbw presets #2/3: 100%|████████████████████████████████████████████████| 1460/1460 [00:00<00:00, 291451.33it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 36012.42it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 2.1s (prepare: 0.2s, merging: 1.9s).
  • loading scripts.patches...
  • lora patch
  • update UNet block input_blocks.0.
  • update UNet block input_blocks.1.
  • update UNet block input_blocks.2.
  • update UNet block input_blocks.3.
  • update UNet block input_blocks.4.
  • update UNet block input_blocks.5.
  • update UNet block input_blocks.6.
  • update UNet block input_blocks.7.
  • update UNet block middle_block.
  • update UNet block output_blocks.0.
  • update UNet block output_blocks.1.
  • update UNet block output_blocks.2.
  • update UNet block output_blocks.3.
  • update UNet block output_blocks.4.
  • update UNet block output_blocks.5.
  • update UNet block output_blocks.6.
  • update UNet block output_blocks.7.
  • update UNet block output_blocks.8.
  • update UNet block time_embed.
  • update UNet block out.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:34<00:00, 2.71s/it]
    debugs = ['elemental merge']████████▎ | 35/140 [01:55<04:48, 2.75s/it]
    use_extra_elements = True
  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1
    config hash = 23f1aeafb1d704786feebdb4c6102b7ff900d8c00f41fd88385ffd87a1539fb7
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['opticon_v10.safetensors [cd7397d079]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['OUT00']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 0.7, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|████████████████████████████████████████████████████████████████████| 218/218 [00:05<00:00, 43.15it/s]
    Check uninitialized support mbw presets #2/3: 100%|██████████████████████████████████████████████████████████████| 218/218 [00:00<?, ?it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 28157.19it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 5.4s (merging: 5.3s).
  • loading scripts.patches...
  • lora patch
  • update UNet block output_blocks.0.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:35<00:00, 2.72s/it]
    debugs = ['elemental merge']████████████████████████▌ | 70/140 [03:45<03:12, 2.75s/it]
    use_extra_elements = True
  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1
    config hash = ada4d1f16841761a82b7b8ede66aa354f9008f3c40c0510558ca546910a65bd3
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['opticon_v10.safetensors [cd7397d079]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,1,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['OUT00']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|██████████████████████████████████████████████████████████████████| 218/218 [00:00<00:00, 1028.62it/s]
    Check uninitialized support mbw presets #2/3: 100%|██████████████████████████████████████████████████| 218/218 [00:00<00:00, 218641.38it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 36099.75it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 0.7s (prepare: 0.2s, merging: 0.5s).
  • loading scripts.patches...
  • lora patch
  • update UNet block output_blocks.0.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:35<00:00, 2.72s/it]
    debugs = ['elemental merge']████████████████████████████████████████ | 105/140 [05:31<01:36, 2.75s/it]
    use_extra_elements = True
  • mm_max_models = 3
    XYZ: mbw alpha b -> 0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1
    config hash = 23f1aeafb1d704786feebdb4c6102b7ff900d8c00f41fd88385ffd87a1539fb7
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['opticon_v10.safetensors [cd7397d079]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['0,1,1,1,1,1,1,1,1,0,1,0.7,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['OUT00']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 0.0, 1.0, 0.7, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|████████████████████████████████████████████████████████████████████| 218/218 [00:05<00:00, 42.97it/s]
    Check uninitialized support mbw presets #2/3: 100%|██████████████████████████████████████████████████| 218/218 [00:00<00:00, 218589.12it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 252/252 [00:00<00:00, 28019.10it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 5.5s (prepare: 0.2s, merging: 5.3s).
  • loading scripts.patches...
  • lora patch
  • update UNet block output_blocks.0.
  • UNet partial blocks have been successfully updated
    Applying attention optimization: Doggettx... done.
  • remove old checkpointinfo
    100%|██████████████████████████████████████████████████████████████████████████████████| 35/35 [01:34<00:00, 2.71s/it]
    Total progress: 100%|████████████████████████████████████████████████████████████████| 140/140 [07:35<00:00, 3.25s/it]
    X/Y/Z plot will create 84 images on 1 7x3 grid; 4 images per cell. (Total steps to process: 735)07:35<00:00, 2.74s/it]
    debugs = ['elemental merge']
    use_extra_elements = True, ?it/s]
  • mm_max_models = 3
    XYZ: mbw alpha b -> 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1
    config hash = 70cad8728437980840661c1ed79d7204dbb5d940362e26893a79e862f28009d8
  • mm_use [True, False, False]
  • model_a butapony_v20.safetensors [4fda762c92]
  • base_model None
  • max_models 3
  • models ['oneTrickPonyxl_v20.safetensors [932797f91f]']
  • modes ['Dare-Fixed']
  • calcmodes ['Normal']
  • usembws [['ALL']]
  • weights ['1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1']
  • alpha [0.5]
  • adjust
  • use elemental [False]
  • elementals ['']
    model_a = butapony_v20
    sdversion = XL
  • Parse elemental merge...
    compact_mode = True
  • check possible UNet partial update...
  • partial changed blocks = ['BASE', 'IN00', 'IN01', 'IN02', 'IN03', 'IN04', 'IN05', 'IN06', 'IN07', 'IN08', 'M00', 'OUT00', 'OUT01', 'OUT02', 'OUT03', 'OUT04', 'OUT05', 'OUT06', 'OUT07', 'OUT08']
  • UNet partial update mode
    mode = Dare-Fixed, mbw mode, alpha = [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]
    Stage Issue where model in merged but not save leads to "Cannot copy out of meta tensor; no data!" - persists through restart - had to disable app to continue #1/3: 100%|█████████████████████████████████████████████████████████████████| 2260/2260 [00:02<00:00, 873.02it/s]
    Check uninitialized support mbw presets #2/3: 100%|████████████████████████████████████████████████| 2260/2260 [00:00<00:00, 282411.05it/s]
    Save unchanged weights Add to A1111 Extension List #3/3: 100%|████████████████████████████████████████████████| 255/255 [00:00<00:00, 31988.86it/s]
    Broken clip!
    [3, 6, 12, 23, 24, 29, 41, 46, 48, 51, 53, 58, 63]
  • merge processing in 3.0s (prepare: 0.1s, merging: 2.9s).
  • loading scripts.patches...
  • lora patch
  • Textencoder(BASE) has been successfully updated
  • update UNet block input_blocks.0.
  • update UNet block input_blocks.1.
  • update UNet block input_blocks.2.
  • update UNet block input_blocks.3.
  • update UNet block input_blocks.4.
  • update UNet block input_blocks.5.
  • update UNet block input_blocks.6.
  • update UNet block input_blocks.7.
  • update UNet block input_blocks.8.
  • update UNet block middle_block.
  • update UNet block output_blocks.0.
  • update UNet block output_blocks.1.
  • update UNet block output_blocks.2.
  • update UNet block output_blocks.3.
  • update UNet block output_blocks.4.
  • update UNet block output_blocks.5.
  • update UNet block output_blocks.6.
  • update UNet block output_blocks.7.
  • update UNet block output_blocks.8.
  • update UNet block time_embed.
  • update UNet block out.
  • UNet partial blocks have been successfully updated
  • Reload full state_dict...
  • remove old checkpointinfo
    WARN: workaround load_model() to fix 'copy out of meta tensor' error (launched without --medvram cmd option case.)
    *** Error running before_process: D:\Programs\SD\webui\extensions\sd-webui-model-mixer\scripts\model_mixer.py
    Traceback (most recent call last):
    File "D:\Programs\SD\webui\modules\scripts.py", line 824, in before_process
    script.before_process(p, *script_args)
    File "D:\Programs\SD\webui\extensions\sd-webui-model-mixer\scripts\model_mixer.py", line 4816, in before_process
    sd_models.unload_model_weights()
    File "D:\Programs\SD\webui\modules\sd_models.py", line 1006, in unload_model_weights
    send_model_to_cpu(sd_model or shared.sd_model)
    File "D:\Programs\SD\webui\modules\sd_models.py", line 743, in send_model_to_cpu
    m.to(devices.cpu)
    File "D:\Programs\SD\system\python\lib\site-packages\lightning_fabric\utilities\device_dtype_mixin.py", line 54, in to
    return super().to(*args, **kwargs)
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1160, in to
    return self._apply(convert)
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
    module._apply(fn)
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
    module._apply(fn)
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 810, in _apply
    module._apply(fn)
    [Previous line repeated 1 more time]
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 833, in _apply
    param_applied = fn(param)
    File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1158, in convert
    return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
    NotImplementedError: Cannot copy out of meta tensor; no data!

generating image for xyz plot: NotImplementedError
Traceback (most recent call last):
File "D:\Programs\SD\webui\scripts\xyz_grid.py", line 728, in cell
res = process_images(pc)
File "D:\Programs\SD\webui\modules\processing.py", line 847, in process_images
res = process_images_inner(p)
File "D:\Programs\SD\webui\modules\processing.py", line 966, in process_images_inner
p.setup_conds()
File "D:\Programs\SD\webui\modules\processing.py", line 1520, in setup_conds
super().setup_conds()
File "D:\Programs\SD\webui\modules\processing.py", line 503, in setup_conds
self.c = self.get_conds_with_caching(prompt_parser.get_multicond_learned_conditioning, prompts, total_steps, [self.cached_c], self.extra_network_data)
File "D:\Programs\SD\webui\modules\processing.py", line 488, in get_conds_with_caching
cache[1] = function(shared.sd_model, required_prompts, steps, hires_steps, shared.opts.use_old_scheduling)
File "D:\Programs\SD\webui\modules\prompt_parser.py", line 261, in get_multicond_learned_conditioning
learned_conditioning = get_learned_conditioning(model, prompt_flat_list, steps, hires_steps, use_old_scheduling)
File "D:\Programs\SD\webui\modules\prompt_parser.py", line 188, in get_learned_conditioning
conds = model.get_learned_conditioning(texts)
File "D:\Programs\SD\webui\modules\sd_models_xl.py", line 32, in get_learned_conditioning
c = self.conditioner(sdxl_conds, force_zero_embeddings=['txt'] if force_zero_negative_prompt else [])
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "D:\Programs\SD\webui\repositories\generative-models\sgm\modules\encoders\modules.py", line 141, in forward
emb_out = embedder(batch[embedder.input_key])
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "D:\Programs\SD\webui\repositories\generative-models\sgm\util.py", line 59, in do_autocast
return f(*args, **kwargs)
File "D:\Programs\SD\webui\repositories\generative-models\sgm\modules\encoders\modules.py", line 391, in forward
outputs = self.transformer(
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\transformers\models\clip\modeling_clip.py", line 822, in forward
return self.text_model(
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\torch\nn\modules\module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "D:\Programs\SD\system\python\lib\site-packages\transformers\models\clip\modeling_clip.py", line 734, in forward
causal_attention_mask = _make_causal_mask(input_shape, hidden_states.dtype, device=hidden_states.device)
File "D:\Programs\SD\system\python\lib\site-packages\transformers\models\clip\modeling_clip.py", line 684, in _make_causal_mask
mask = torch.full((tgt_len, tgt_len), torch.tensor(torch.finfo(dtype).min, device=device), device=device)

I reinstalled Automatic1111 and Model Mixer, and still got the error.

What helped me to resolve the issue was to roll back to a September version of Model Mixer. Now it all works again.

Could this be related to this commit? : 211e13c#diff-81675046c08690da75a12e021fb5086b96e41641277e8feff350ff94541cb218R4765
@wkpark wkpark self-assigned this Oct 23, 2024
@wkpark wkpark added the bug Something isn't working label Oct 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@wkpark wkpark

Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@wkpark @windpiligrim