LCM img2img error | TypeError: slice indices must be integers or None or have an __index__ method

[PladsElsker]

I re-installed diffusers twice, I keep running into this error (LCM txt2img works fine though, only img2img is broken):

Traceback (most recent call last):
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\gradio\routes.py", line 488, in run_predict
    output = await app.get_blocks().process_api(
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\gradio\blocks.py", line 1431, in process_api
    result = await self.call_function(
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\gradio\blocks.py", line 1103, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\extensions\sd-webui-lcm\scripts\main.py", line 172, in generate_i2i
    result = pipe(
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\extensions\sd-webui-lcm\lcm\lcm_i2i_pipeline.py", line 347, in __call__
    latents, denoised = self.scheduler.step(model_pred, i, t, latents, return_dict=False)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\diffusers\schedulers\scheduling_lcm.py", line 425, in step
    self._init_step_index(timestep)
  File "C:\Users\Plads\Documents\GitHub\stable diffusion\stable-diffusion-avidhjeren\venv\lib\site-packages\diffusers\schedulers\scheduling_lcm.py", line 264, in _init_step_index
    step_index = index_candidates[0]
IndexError: index 0 is out of bounds for dimension 0 with size 0

This is where the code breaks:

# in scheduling_lcm.py
[...]

        index_candidates = (self.timesteps == timestep).nonzero()  # <- this returns a tensor with shape (1, 0) because timestep not in self.timesteps apparently? 

        # The sigma index that is taken for the **very** first `step`
        # is always the second index (or the last index if there is only 1)
        # This way we can ensure we don't accidentally skip a sigma in
        # case we start in the middle of the denoising schedule (e.g. for image-to-image)
        if len(index_candidates) > 1:
            step_index = index_candidates[1]
        else:
            step_index = index_candidates[0]  # <- breaks here

[...]

Any idea why this happens? Thanks.

[codercv]

Approve TXT2IMG - work, IMG2IMG - Not

Traceback (most recent call last):
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\gradio\routes.py", line 488, in run_predict
    output = await app.get_blocks().process_api(
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1431, in process_api
    result = await self.call_function(
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1103, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "D:\AISD\stable-diffusion-webui\extensions\sd-webui-lcm\scripts\main.py", line 172, in generate_i2i
    result = pipe(
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "D:\AISD\stable-diffusion-webui\extensions\sd-webui-lcm\lcm\lcm_i2i_pipeline.py", line 305, in __call__
    self.scheduler.set_timesteps(strength, num_inference_steps, original_inference_steps)
  File "d:\AISD\stable-diffusion-webui\venv\lib\site-packages\diffusers\schedulers\scheduling_lcm.py", line 377, in set_timesteps
    timesteps = lcm_origin_timesteps[::-skipping_step][:num_inference_steps]
TypeError: slice indices must be integers or None or have an __index__ method

[PladsElsker]

Oh I forgot that I actually had your error first. Your error is the right one.

I fixed the stack trace that you had by manually re-ordering the parameters of a function call in the library files (don't do this). Then, mine came next.

When I tried to re-install diffusers, it did not bring your error back, so I forgot about it haha. I think it should have brought the error back, but it didn't for some reason.

Not sure if all that was super clear, so let me re-iterate: @codercv's error is the one I should have issued.

[ricodp54]

I'm having this problem too, but I was able to use it fine until a week ago.

[wandrzej]

Same issue here TypeError: slice indices must be integers or None or have an __index__ method

[lukecageit]

Same issue here TypeError: slice indices must be integers or None or have an __index__ method

Same here, i don't know if it can be a new dependency version or something like that?

[zenphyl]

Same issue here TypeError: slice indices must be integers or None or have an __index__ method

same here

[Nilok7]

Having the same issue here for img2img, while txt2img works fine. Seems to be related to gradio and anyio.

S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\diffusers\pipelines\pipeline_utils.py:749: FutureWarning: `torch_dtype` is deprecated and will be removed in version 0.25.0.
  deprecate("torch_dtype", "0.25.0", "")
S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\diffusers\pipelines\pipeline_utils.py:752: FutureWarning: `torch_device` is deprecated and will be removed in version 0.25.0.
  deprecate("torch_device", "0.25.0", "")
Traceback (most recent call last):
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\gradio\routes.py", line 488, in run_predict
    output = await app.get_blocks().process_api(
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1431, in process_api
    result = await self.call_function(
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1103, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\gradio\utils.py", line 707, in wrapper
    response = f(*args, **kwargs)
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\extensions\sd-webui-lcm\scripts\main.py", line 172, in generate_i2i
    result = pipe(
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\extensions\sd-webui-lcm\lcm\lcm_i2i_pipeline.py", line 305, in __call__
    self.scheduler.set_timesteps(strength, num_inference_steps, original_inference_steps)
  File "S:\StableDiffusion-Automatic1111\stable-diffusion-webui\venv\lib\site-packages\diffusers\schedulers\scheduling_lcm.py", line 382, in set_timesteps
    timesteps = lcm_origin_timesteps[::-skipping_step][:num_inference_steps]
TypeError: slice indices must be integers or None or have an __index__ method

[CalladsEssence]

Can also confirm this issue. Updated the diffusers conform the LCM README. (which fixed the error in LCM txt2img)

This is my version of AUTOMATIC1111 version: [v1.6.0-2-g4afaaf8a]  •  python: 3.10.6  •  torch: 2.0.1+cu118  •  xformers: N/A  •  gradio: 3.41.2

The line in 'scheduling_lcm.py' that seems to cause trouble is this one (382): timesteps = lcm_origin_timesteps[::-skipping_step][:num_inference_steps]

Since the error is TypeError: slice indices must be integers or None or have an index method

So 'timesteps' must be filled with an integer?

A few lines back (379) an 'int' is mentioned and it is used for 'lcm_origin_timesteps'. lcm_origin_timesteps = np.asarray(list(range(1, int(original_steps strength) + 1))) c - 1

So, is there a problem with either line 379 or line 382?

(no clue if this was useful mentioning, if not: plz forgive me -I'm totally blonde-)

[lilyzlt]

the same error, img2img and video2video all has this problems