diffusers
diffusers copied to clipboard
huggingface
Issue: Error handling of invalid orders
Describe the bug
When I attempt to create an XYZ grid using all available Samplers (other than Default) with an SDXL model in the X axis, and [Sampler] solver order: 1-5 in the Y axis, and nothing for Z axis, I get the below error (see relevant log export).
Final error is "UnboundLocalError: local variable 'orders' referenced before assignment".
I reported this to SD.Next here: https://github.com/vladmandic/automatic/issues/2454 The author responded: "Sampler DPM++ 1S has valid orders of 1,2,3. Issue occurs when using order out of range. If you want to have better error handling of invalid orders, you may want to report to https://github.com/huggingface/diffusers/"
Reproduction
Not sure how to do this in code, but I can share my process. In SD.Next, Stable Diffusion XL pipeline, I created an XYZ grid with all samplers in X axis, and Solver Order: 1-5 in Y axis. The issue took a long time to show up so it did complete the earlier solver orders. When it hit a solver order that was not valid, instead of noting that and moving on, it errored out and stopped executing the grid.
Logs
10:34:36-443850 INFO XYZ grid: images=70 grid=1 14x5 cells=1 steps=3500
10:34:36-443850 INFO Applying hypertile: unet=256
Progress 2.20it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
10:35:08-980475 INFO Processed: images=1 time=32.52 its=1.54 memory={'ram': {'used': 6.66, 'total': 31.85}, 'gpu': {'used':
15.13, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:35:08-996101 INFO Applying hypertile: unet=256
Progress 2.51it/s █████████████████████████████████ 100% 50/50 00:19 00:00 Base
10:35:35-698241 INFO Processed: images=1 time=26.69 its=1.87 memory={'ram': {'used': 6.67, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:35:35-698241 INFO Applying hypertile: unet=256
Progress 2.06it/s █████████████████████████████████ 100% 50/50 00:24 00:00 Base
10:36:06-557953 INFO Processed: images=1 time=30.86 its=1.62 memory={'ram': {'used': 7.22, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:36:06-560555 INFO Applying hypertile: unet=256
Progress 2.41it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:36:33-669330 INFO Processed: images=1 time=27.11 its=1.84 memory={'ram': {'used': 7.26, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:36:33-684957 INFO Applying hypertile: unet=256
Progress 2.41it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:37:01-641612 INFO Processed: images=1 time=27.96 its=1.79 memory={'ram': {'used': 7.35, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:37:01-641612 INFO Applying hypertile: unet=256
Progress 2.26it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
10:37:30-738992 INFO Processed: images=1 time=29.08 its=1.72 memory={'ram': {'used': 7.39, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:37:30-738992 INFO Applying hypertile: unet=256
Progress 1.22it/s █████████████████████████████████ 100% 50/50 00:40 00:00 Base
10:38:18-354242 INFO Processed: images=1 time=47.62 its=1.05 memory={'ram': {'used': 7.29, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:38:18-354242 INFO Applying hypertile: unet=256
Progress 1.20it/s █████████████████████████████████ 100% 50/50 00:41 00:00 Base
10:39:06-645770 INFO Processed: images=1 time=48.28 its=1.04 memory={'ram': {'used': 7.32, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:39:06-645770 INFO Applying hypertile: unet=256
Progress 2.48it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:39:34-048889 INFO Processed: images=1 time=27.39 its=1.83 memory={'ram': {'used': 7.37, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:39:34-064572 INFO Applying hypertile: unet=256
Progress 2.48it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:40:02-384658 INFO Processed: images=1 time=28.29 its=1.77 memory={'ram': {'used': 5.47, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:40:02-384658 INFO Applying hypertile: unet=256
Progress 1.20it/s █████████████████████████████████ 100% 50/50 00:41 00:00 Base
10:40:51-620433 INFO Processed: images=1 time=49.24 its=1.02 memory={'ram': {'used': 5.98, 'total': 31.85}, 'gpu': {'used':
17.52, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:40:51-636044 INFO Applying hypertile: unet=256
Progress 2.40it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:41:19-625417 INFO Processed: images=1 time=27.97 its=1.79 memory={'ram': {'used': 6.12, 'total': 31.85}, 'gpu': {'used':
17.53, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:41:19-641039 INFO Applying hypertile: unet=256
Progress 2.49it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:41:46-825088 INFO Processed: images=1 time=27.18 its=1.84 memory={'ram': {'used': 6.18, 'total': 31.85}, 'gpu': {'used':
17.53, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:41:46-840713 INFO Applying hypertile: unet=256
Progress 1.22it/s █████████████████████████████████ 100% 50/50 00:40 00:00 Base
10:42:33-958474 INFO Processed: images=1 time=47.12 its=1.06 memory={'ram': {'used': 6.12, 'total': 31.85}, 'gpu': {'used':
17.53, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:42:33-958474 INFO Applying hypertile: unet=256
Progress 2.41it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:43:00-998803 INFO Processed: images=1 time=27.02 its=1.85 memory={'ram': {'used': 6.18, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:43:01-014406 INFO Applying hypertile: unet=256
Progress 2.50it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:43:27-917324 INFO Processed: images=1 time=26.90 its=1.86 memory={'ram': {'used': 6.22, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:43:27-917324 INFO Applying hypertile: unet=256
Progress 2.06it/s █████████████████████████████████ 100% 50/50 00:24 00:00 Base
10:43:58-357471 INFO Processed: images=1 time=30.43 its=1.64 memory={'ram': {'used': 6.27, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:43:58-373037 INFO Applying hypertile: unet=256
Progress 2.42it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:44:25-368944 INFO Processed: images=1 time=27.00 its=1.85 memory={'ram': {'used': 6.32, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:44:25-384656 INFO Applying hypertile: unet=256
Progress 2.42it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:44:52-140387 INFO Processed: images=1 time=26.76 its=1.87 memory={'ram': {'used': 6.36, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:44:52-156098 INFO Applying hypertile: unet=256
Progress 2.29it/s █████████████████████████████████ 100% 50/50 00:21 00:00 Base
10:45:20-400224 INFO Processed: images=1 time=28.21 its=1.77 memory={'ram': {'used': 6.22, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:45:20-400224 INFO Applying hypertile: unet=256
Progress 1.23it/s █████████████████████████████████ 100% 50/50 00:40 00:00 Base
10:46:07-812175 INFO Processed: images=1 time=47.40 its=1.05 memory={'ram': {'used': 6.25, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:46:07-827750 INFO Applying hypertile: unet=256
Progress 1.23it/s █████████████████████████████████ 100% 50/50 00:40 00:00 Base
10:46:56-497078 INFO Processed: images=1 time=48.67 its=1.03 memory={'ram': {'used': 5.66, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:46:56-512694 INFO Applying hypertile: unet=256
Progress 2.50it/s █████████████████████████████████ 100% 50/50 00:19 00:00 Base
10:47:23-462955 INFO Processed: images=1 time=26.94 its=1.86 memory={'ram': {'used': 6.08, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:47:23-462955 INFO Applying hypertile: unet=256
Progress 2.50it/s █████████████████████████████████ 100% 50/50 00:19 00:00 Base
10:47:50-106419 INFO Processed: images=1 time=26.63 its=1.88 memory={'ram': {'used': 6.26, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:47:50-106419 INFO Applying hypertile: unet=256
Progress 1.20it/s █████████████████████████████████ 100% 50/50 00:41 00:00 Base
10:48:39-542567 INFO Processed: images=1 time=49.42 its=1.01 memory={'ram': {'used': 5.45, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:48:39-558186 INFO Applying hypertile: unet=256
Progress 2.42it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:49:06-583628 INFO Processed: images=1 time=27.03 its=1.85 memory={'ram': {'used': 6.14, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:49:06-583628 INFO Applying hypertile: unet=256
Progress 2.50it/s █████████████████████████████████ 100% 50/50 00:19 00:00 Base
10:49:33-497938 INFO Processed: images=1 time=26.90 its=1.86 memory={'ram': {'used': 6.34, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:49:33-520643 INFO Applying hypertile: unet=256
Progress 1.19it/s █████████████████████████████████ 100% 50/50 00:41 00:00 Base
10:50:21-962462 INFO Processed: images=1 time=48.43 its=1.03 memory={'ram': {'used': 6.42, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:50:21-978088 INFO Applying hypertile: unet=256
Progress 2.39it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:50:49-571212 INFO Processed: images=1 time=27.59 its=1.81 memory={'ram': {'used': 6.48, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:50:49-577037 INFO Applying hypertile: unet=256
Progress 2.45it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:51:16-990653 INFO Processed: images=1 time=27.40 its=1.82 memory={'ram': {'used': 6.52, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:51:16-990653 INFO Applying hypertile: unet=256
Progress 1.89it/s █████████████████████████████████ 100% 50/50 00:26 00:00 Base
10:51:50-989978 INFO Processed: images=1 time=33.99 its=1.47 memory={'ram': {'used': 5.5, 'total': 31.85}, 'gpu': {'used': 17.56,
'total': 23.99}, 'retries': 0, 'oom': 0}
10:51:50-996977 INFO Applying hypertile: unet=256
Progress 2.25it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
10:52:19-806701 INFO Processed: images=1 time=28.80 its=1.74 memory={'ram': {'used': 6.22, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:52:19-821209 INFO Applying hypertile: unet=256
Progress 2.30it/s █████████████████████████████████ 100% 50/50 00:21 00:00 Base
10:52:47-668490 INFO Processed: images=1 time=27.85 its=1.80 memory={'ram': {'used': 6.4, 'total': 31.85}, 'gpu': {'used': 17.56,
'total': 23.99}, 'retries': 0, 'oom': 0}
10:52:47-675555 INFO Applying hypertile: unet=256
Progress 2.09it/s █████████████████████████████████ 100% 50/50 00:23 00:00 Base
10:53:17-664819 INFO Processed: images=1 time=29.98 its=1.67 memory={'ram': {'used': 6.48, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:53:17-671882 INFO Applying hypertile: unet=256
Progress 1.16it/s █████████████████████████████████ 100% 50/50 00:42 00:00 Base
10:54:08-869732 INFO Processed: images=1 time=51.19 its=0.98 memory={'ram': {'used': 5.56, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:54:08-885460 INFO Applying hypertile: unet=256
Progress 1.14it/s █████████████████████████████████ 100% 50/50 00:43 00:00 Base
10:55:00-403189 INFO Processed: images=1 time=51.52 its=0.97 memory={'ram': {'used': 6.27, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:55:00-418917 INFO Applying hypertile: unet=256
Progress 2.46it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:55:27-935021 INFO Processed: images=1 time=27.50 its=1.82 memory={'ram': {'used': 6.47, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:55:27-950652 INFO Applying hypertile: unet=256
Progress 2.44it/s █████████████████████████████████ 100% 50/50 00:20 00:00 Base
10:55:55-304991 INFO Processed: images=1 time=27.35 its=1.83 memory={'ram': {'used': 6.53, 'total': 31.85}, 'gpu': {'used':
17.56, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:55:55-321113 INFO Applying hypertile: unet=256
Progress 1.10it/s █████████████████████████████████ 100% 50/50 00:45 00:00 Base
10:56:48-057360 INFO Processed: images=1 time=52.72 its=0.95 memory={'ram': {'used': 5.61, 'total': 31.85}, 'gpu': {'used':
17.57, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:56:48-063973 INFO Applying hypertile: unet=256
Progress 2.37it/s █████████████████████████████████ 100% 50/50 00:21 00:00 Base
10:57:13-170718 INFO Processed: images=1 time=25.10 its=1.99 memory={'ram': {'used': 10.85, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:57:13-170718 INFO Applying hypertile: unet=256
Progress 2.35it/s █████████████████████████████████ 100% 50/50 00:21 00:00 Base
10:57:40-697628 INFO Processed: images=1 time=27.53 its=1.82 memory={'ram': {'used': 6.5, 'total': 31.85}, 'gpu': {'used': 17.58,
'total': 23.99}, 'retries': 0, 'oom': 0}
10:57:40-709633 INFO Applying hypertile: unet=256
Progress 1.12it/s █████████████████████████████████ 100% 50/50 00:44 00:00 Base
10:58:31-496251 INFO Processed: images=1 time=50.78 its=0.98 memory={'ram': {'used': 6.59, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:58:31-499323 INFO Applying hypertile: unet=256
Progress 2.12it/s █████████████████████████████████ 100% 50/50 00:23 00:00 Base
10:59:01-295198 INFO Processed: images=1 time=29.78 its=1.68 memory={'ram': {'used': 6.52, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:59:01-303224 INFO Applying hypertile: unet=256
Progress 2.19it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
10:59:31-340037 INFO Processed: images=1 time=30.03 its=1.67 memory={'ram': {'used': 6.55, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
10:59:31-348038 INFO Applying hypertile: unet=256
Progress 1.91it/s █████████████████████████████████ 100% 50/50 00:26 00:00 Base
11:00:03-660307 INFO Processed: images=1 time=32.30 its=1.55 memory={'ram': {'used': 6.58, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:00:03-668276 INFO Applying hypertile: unet=256
Progress 2.24it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
11:00:31-869583 INFO Processed: images=1 time=28.19 its=1.77 memory={'ram': {'used': 6.64, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:00:31-869583 INFO Applying hypertile: unet=256
Progress 2.29it/s █████████████████████████████████ 100% 50/50 00:21 00:00 Base
11:01:00-117427 INFO Processed: images=1 time=28.23 its=1.77 memory={'ram': {'used': 6.68, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:01:00-130365 INFO Applying hypertile: unet=256
Progress 2.21it/s █████████████████████████████████ 100% 50/50 00:22 00:00 Base
11:01:28-741124 INFO Processed: images=1 time=28.61 its=1.75 memory={'ram': {'used': 6.72, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:01:28-747124 INFO Applying hypertile: unet=256
Progress 1.18it/s █████████████████████████████████ 100% 50/50 00:42 00:00 Base
11:02:17-792284 INFO Processed: images=1 time=49.03 its=1.02 memory={'ram': {'used': 6.59, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:02:17-802392 INFO Applying hypertile: unet=256
Progress 1.19it/s █████████████████████████████████ 100% 50/50 00:41 00:00 Base
11:03:06-410550 INFO Processed: images=1 time=48.60 its=1.03 memory={'ram': {'used': 6.62, 'total': 31.85}, 'gpu': {'used':
17.58, 'total': 23.99}, 'retries': 0, 'oom': 0}
11:03:06-426449 INFO Applying hypertile: unet=256
11:03:06-442449 ERROR Exception: local variable 'orders' referenced before assignment
11:03:06-442449 ERROR Arguments: args=('task(5vd9wyaymefv5xt)', 'realistic photo of a 45 year-old Mexican woman on a
farm.\n\ndetailed face, highres, realistic, photorealistic, pores, wrinkles, High Detail, Sharp focus,
dramatic, (looking at viewer:1.2), (natural light)\n\nExtremely detailed high quality photo, photorealistic,
realistic skin, sharp focus, realistic texture, focused, masterpiece. ', 'shiny skin, blur, bokeh, noise,
film grain, bad anatomy, airbrushed, misaligned eyes, bad hands, extra fingers, extra limbs, missing limbs,
illustration, drawing, painting, cartoon, anime, text, watermark, signature, ', [], 50, 11, 11, True, False,
True, 1, 1, 4, 6, 0.7, 1, -1.0, -1.0, 0, 0, 0, 1280, 1024, False, 0.5, 1.25, 'chaiNNer 4x ESRGAN
Helaman-LSDIRplus', True, 20, 0, 0, 5, 1, '', '', [], 3, False, False, 'positive', 'comma', 0, False, False,
'', 5, '', ['UniPC', 'DEIS', 'PNDM', 'DDPM', 'DDIM', 'LMSD', 'KDPM2', 'KDPM2 a', 'DPM++ 1S', 'DPM++ 2M',
'DPM SDE', 'Euler', 'Euler a', 'Heun'], 24, '1-5', [], 0, '', [], False, True, False, False, False, False,
0, 5, 'all', 'all', 'photograph', '', '', '', '1', 'none', False, '', '', 'comma', '', True, '', '20',
'female', 'all', 'all', 'all', 0, '', False, False, {'ad_model': 'face_yolov8s.pt', 'ad_prompt': '',
'ad_negative_prompt': '', 'ad_confidence': 0.3, 'ad_mask_k_largest': 0, 'ad_mask_min_ratio': 0,
'ad_mask_max_ratio': 1, 'ad_x_offset': 0, 'ad_y_offset': 0, 'ad_dilate_erode': 4, 'ad_mask_merge_invert':
'None', 'ad_mask_blur': 4, 'ad_denoising_strength': 0.4, 'ad_inpaint_only_masked': True,
'ad_inpaint_only_masked_padding': 32, 'ad_use_inpaint_width_height': False, 'ad_inpaint_width': 512,
'ad_inpaint_height': 512, 'ad_use_steps': False, 'ad_steps': 28, 'ad_use_cfg_scale': False, 'ad_cfg_scale':
7, 'ad_use_checkpoint': False, 'ad_checkpoint': 'Use same checkpoint', 'ad_use_vae': False, 'ad_vae': 'Use
same VAE', 'ad_use_sampler': False, 'ad_sampler': 'Default', 'ad_use_noise_multiplier': False,
'ad_noise_multiplier': 1, 'ad_use_clip_skip': False, 'ad_clip_skip': 1, 'ad_restore_face': False,
'ad_controlnet_model': 'None', 'ad_controlnet_module': 'None', 'ad_controlnet_weight': 1,
'ad_controlnet_guidance_start': 0, 'ad_controlnet_guidance_end': 1, 'is_api': ()}, {'ad_model': 'None',
'ad_prompt': '', 'ad_negative_prompt': '', 'ad_confidence': 0.3, 'ad_mask_k_largest': 0,
'ad_mask_min_ratio': 0, 'ad_mask_max_ratio': 1, 'ad_x_offset': 0, 'ad_y_offset': 0, 'ad_dilate_erode': 4,
'ad_mask_merge_invert': 'None', 'ad_mask_blur': 4, 'ad_denoising_strength': 0.4, 'ad_inpaint_only_masked':
True, 'ad_inpaint_only_masked_padding': 32, 'ad_use_inpaint_width_height': False, 'ad_inpaint_width': 512,
'ad_inpaint_height': 512, 'ad_use_steps': False, 'ad_steps': 28, 'ad_use_cfg_scale': False, 'ad_cfg_scale':
7, 'ad_use_checkpoint': False, 'ad_checkpoint': 'Use same checkpoint', 'ad_use_vae': False, 'ad_vae': 'Use
same VAE', 'ad_use_sampler': False, 'ad_sampler': 'Default', 'ad_use_noise_multiplier': False,
'ad_noise_multiplier': 1, 'ad_use_clip_skip': False, 'ad_clip_skip': 1, 'ad_restore_face': False,
'ad_controlnet_model': 'None', 'ad_controlnet_module': 'None', 'ad_controlnet_weight': 1,
'ad_controlnet_guidance_start': 0, 'ad_controlnet_guidance_end': 1, 'is_api': ()}, True, False, 1, False,
False, False, 1.1, 1.5, 100, 0.7, False, False, True, False, False, 0,
'Gustavosta/MagicPrompt-Stable-Diffusion', '', <scripts.animatediff_ui.AnimateDiffProcess object at
0x0000028217B34370>, None, False, '0', '0', 'inswapper_128.onnx', 'GFPGAN', 1, True, 'ESRGAN 8x NMKD Faces',
1, 1, False, True, 1, 0, 0, False, 0.1, True, False, 'CUDA', False, False, False, False, 'Hyperrealism')
kwargs={}
11:03:06-458483 ERROR gradio call: UnboundLocalError
╭──────────────────────────────────────────────── Traceback (most recent call last) ────────────────────────────────────────────────╮
│ C:\sdnext\modules\call_queue.py:34 in f │
│ │
│ 33 │ │ │ try: │
│ ❱ 34 │ │ │ │ res = func(*args, **kwargs) │
│ 35 │ │ │ │ progress.record_results(id_task, res) │
│ │
│ C:\sdnext\modules\txt2img.py:64 in txt2img │
│ │
│ 63 │ p.script_args = args │
│ ❱ 64 │ processed = modules.scripts.scripts_txt2img.run(p, *args) │
│ 65 │ if processed is None: │
│ │
│ C:\sdnext\modules\scripts.py:498 in run │
│ │
│ 497 │ │ parsed = p.per_script_args.get(script.title(), args[script.args_from:script.args │
│ ❱ 498 │ │ processed = script.run(p, *parsed) │
│ 499 │ │ s.record(script.title()) │
│ │
│ C:\sdnext\scripts\xyz_grid.py:667 in run │
│ │
│ 666 │ │ with SharedSettingsStackHelper(): │
│ ❱ 667 │ │ │ processed = draw_xyz_grid( │
│ 668 │ │ │ │ p, │
│ │
│ C:\sdnext\scripts\xyz_grid.py:323 in draw_xyz_grid │
│ │
│ 322 │ │ │ │ │ for ix, x in enumerate(xs): │
│ ❱ 323 │ │ │ │ │ │ process_cell(x, y, z, ix, iy, iz) │
│ 324 │
│ │
│ ... 7 frames hidden ... │
│ │
│ C:\sdnext\modules\sd_samplers_diffusers.py:55 in <lambda> │
│ │
│ 54 │ sd_samplers_common.SamplerData('KDPM2 a', lambda model: DiffusionSampler('KDPM2 a', │
│ ❱ 55 │ sd_samplers_common.SamplerData('DPM++ 1S', lambda model: DiffusionSampler('DPM++ 1S' │
│ 56 │ sd_samplers_common.SamplerData('DPM++ 2M', lambda model: DiffusionSampler('DPM++ 2M' │
│ │
│ C:\sdnext\modules\sd_samplers_diffusers.py:108 in __init__ │
│ │
│ 107 │ │ │ self.config['algorithm_type'] = 'deis' │
│ ❱ 108 │ │ self.sampler = constructor(**self.config) │
│ 109 │ │ self.sampler.name = name │
│ │
│ C:\sdnext\venv\lib\site-packages\diffusers\configuration_utils.py:644 in inner_init │
│ │
│ 643 │ │ getattr(self, "register_to_config")(**new_kwargs) │
│ ❱ 644 │ │ init(self, *args, **init_kwargs) │
│ 645 │
│ │
│ C:\sdnext\venv\lib\site-packages\diffusers\schedulers\scheduling_dpmsolver_singlestep.py:199 in __init__ │
│ │
│ 198 │ │ self.sample = None │
│ ❱ 199 │ │ self.order_list = self.get_order_list(num_train_timesteps) │
│ 200 │ │ self._step_index = None │
│ │
│ C:\sdnext\venv\lib\site-packages\diffusers\schedulers\scheduling_dpmsolver_singlestep.py:234 in get_order_list │
│ │
│ 233 │ │ │ │ orders = [1] * steps │
│ ❱ 234 │ │ return orders │
│ 235 │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
UnboundLocalError: local variable 'orders' referenced before assignment
System Info
Version Platform Description
Windows 11, Vivaldi
Using VENV: C:\sdnext\venv 10:00:36-084245 INFO Starting SD.Next 10:00:36-099876 INFO Python 3.10.6 on Windows 10:00:36-179570 INFO Version: app=sd.next updated=2023-11-07 hash=ca7cd437 url=https://github.com/vladmandic/automatic.git/tree/master 10:00:36-399245 INFO Platform: arch=AMD64 cpu=Intel64 Family 6 Model 183 Stepping 1, GenuineIntel system=Windows release=Windows-10-10.0.22621-SP0 python=3.10.6 10:00:36-414260 INFO nVidia CUDA toolkit detected: nvidia-smi present 10:00:36-468667 INFO Startup: standard 10:00:36-468667 INFO Verifying requirements 10:00:36-477170 INFO Verifying packages 10:00:36-477170 INFO Verifying submodules 10:00:51-957444 INFO Extensions enabled: ['Lora', 'sd-extension-chainner', 'sd-extension-system-info', 'sd-webui-agent-scheduler', 'stable-diffusion-webui-images-browser', 'stable-diffusion-webui-rembg', 'adetailer', 'OneButtonPrompt', 'sd-dynamic-prompts', 'sd-webui-animatediff', 'sd-webui-aspect-ratio-helper', 'sd-webui-reactor', 'Stable-Diffusion-Webui-Civitai-Helper', 'StyleSelectorXL'] 10:00:51-971533 INFO Verifying requirements 10:00:51-987240 INFO Updating Wiki 10:00:52-473754 INFO Extension preload: {'extensions-builtin': 0.0, 'extensions': 0.0} 10:00:52-473754 INFO Command line args: ['--upgrade'] upgrade=True 10:00:57-903639 INFO Load packages: torch=2.1.0+cu121 diffusers=0.22.0 gradio=3.43.2 10:00:58-374614 INFO Engine: backend=Backend.DIFFUSERS compute=cuda mode=no_grad device=cuda cross-optimization="Scaled-Dot-Product" 10:00:58-421892 INFO Device: device=NVIDIA GeForce RTX 4090 n=1 arch=sm_90 cap=(8, 9) cuda=12.1 cudnn=8801 driver=546.01 10:00:59-144140 INFO Available VAEs: path="models\VAE" items=4 10:00:59-144140 INFO Disabling uncompatible extensions: backend=Backend.DIFFUSERS ['multidiffusion-upscaler-for-automatic1111', 'a1111-sd-webui-lycoris'] 10:00:59-316801 INFO Available models: path="models\Stable-diffusion" items=74 time=0.17 10:01:00-572565 INFO Extension: script='extensions-builtin\sd-webui-agent-scheduler\scripts\task_scheduler.py' Using sqlite file: extensions-builtin\sd-webui-agent-scheduler\task_scheduler.sqlite3 10:01:01-688158 INFO Extension: script='extensions\adetailer\scripts!adetailer.py' [-] ADetailer initialized. version: 23.11.0, num models: 9 10:01:02-316187 INFO Extensions time: 2.80 { Lora=0.71 sd-webui-agent-scheduler=0.25 stable-diffusion-webui-images-browser=0.06 stable-diffusion-webui-rembg=0.74 adetailer=0.31 OneButtonPrompt=0.16 sd-webui-animatediff=0.08 sd-webui-aspect-ratio-helper=0.06 sd-webui-reactor=0.28 } 10:01:02-394326 INFO Load UI theme: name="invoked" style=Dark base=style.css 10:01:04-503736 INFO Local URL: http://127.0.0.1:7860/ 10:01:04-503736 INFO Initializing middleware 10:01:04-629910 INFO [AgentScheduler] Task queue is empty 10:01:04-629910 INFO [AgentScheduler] Registering APIs 10:01:04-913231 INFO Select: model="AA-FaceRealism\juggernautXL_version6Rundiffusion [1fe6c7ec54]" Loading weights: C:\sdnext\models\Stable-diffusion\AA-FaceRealism\juggernautXL_version6Rundiffusion.safetensors ━━━━━━ 0.0/7… -:--:-- GB 10:01:05-039853 INFO Setting Torch parameters: device=cuda dtype=torch.float32 vae=torch.float32 unet=torch.float32 context=inference_mode fp16=False bf16=False 10:01:05-039853 INFO Loading VAE: model=models\VAE\sdxl_vae.safetensors source=settings 10:01:05-055356 INFO Diffusers: vae="Stable Diffusion XL" class=StableDiffusionXLPipeline file="C:\sdnext\models\Stable-diffusion\AA-FaceRealism\juggernautXL_version6Rundiffusion.safetensors" size=6776MB 10:01:05-086621 INFO Diffusers: model="Stable Diffusion XL" class=StableDiffusionXLPipeline file="C:\sdnext\models\Stable-diffusion\AA-FaceRealism\juggernautXL_version6Rundiffusion.safetensors" size=6776MB
Who can help?
No response
@LankyPoet
we don't have any plan to add order 4 and 5 in DPM schedulers Is this what you are asking? I don't think we should file this as a bug report no?
let me know if I misunderstood something
Thanks. Sorry if I didn't word my post correctly. I am not asking to add higher orders to the existing samplers. I was using the XYZ grid and when I told it to step through higher sampler solver order 1-5 with all samplers, the entire process errored out. My request is that diffusers should be able to handle this error without stopping the entire process. Maybe note "highest order for sampler X is 3, using that instead of 4/5". Anything rather than stop the entire grid from finishing.
ohh I see what you mean we can probably raise a error message here if it helps
I think what you requested might be outside of our scope and should be handled by applications that using us
cc @patrickvonplaten
Thank you again for considering this. I did initially report to the app author, here is what he said:
https://github.com/vladmandic/automatic/issues/2454 The author responded: "Sampler DPM++ 1S has valid orders of 1,2,3. Issue occurs when using order out of range. If you want to have better error handling of invalid orders, you may want to report to https://github.com/huggingface/diffusers/"
@LankyPoet, could you try to reproduce the issue using just diffusers code? Otherwise it'd be hard for us to move forward here
Thanks. Trying to figure out how to accomplish that, I only know the GUI of the app I use. Will update if I can figure it out.
@patrickvonplaten this is a pretty trivial item, root cause is
https://github.com/huggingface/diffusers/blob/8092017d3f67b8da09a85c2368ea5d88ae8b4a6e/src/diffusers/schedulers/scheduling_dpmsolver_singlestep.py#L212-L234
there is if/elif/elif but no else so orders param never gets set. simple solution is change last if order == 3 to if order >= 3
i understand that order > 3 is not supported by this scheduler, but it is by some others and goal is to enable usage in our xyz grid so user can test multiple settings to produce a grid.
I don't think the solver's order is supposed to be larger than 3 - see: https://github.com/huggingface/diffusers/blob/8092017d3f67b8da09a85c2368ea5d88ae8b4a6e/src/diffusers/schedulers/scheduling_dpmsolver_singlestep.py#L96
I agree, ask is to either force 3 as max or add assert - both/either to avoid runtime errors. Diffusers perform validation for most of other user inputs.
Happy to throw a ValueError when order is > 3 - @LankyPoet would you maybe like to open an issue?
Yes but isn’t that what this issue already addresses? I can create another but I feel like it is duplicating.
On Tue, Nov 21, 2023 at 9:21 AM Patrick von Platen @.***> wrote:
Happy to throw a ValueError when order is > 3 - @LankyPoet https://github.com/LankyPoet would you maybe like to open an issue?
— Reply to this email directly, view it on GitHub https://github.com/huggingface/diffusers/issues/5705#issuecomment-1821016791, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDALA3MOJMVT7Y7PJGGKXQTYFS2FVAVCNFSM6AAAAAA7C66BA6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRRGAYTMNZZGE . You are receiving this because you were mentioned.Message ID: @.***>
@LankyPoet he meant "to open a PR to fix this":)
I tried my best but don't think I did it right. The pull request shows up in my own account, not in huggingface. I am sorry I am new to this. Here is the updated file if you can pull it in? https://github.com/huggingface/diffusers/compare/main...LankyPoet:diffusers:LankyPoet-dpmsolverpatch-1
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
@yiyixuxu Hi, I see the stale note, not sure if you saw my last reply and if it was usable? Thank you.
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
Hi, again just sending a reminder in case this has not been addressed. I think it's a good thing to incorporate to improve error-handling.
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
@yiyixuxu
Since this is a really easy fix it could be tagged with contributions-welcome and good first issue. If no one takes it I can do it.
