Open Huey-Lewy opened 1 year ago
Error loading script: api_t2v.py
Traceback (most recent call last):
File "G:\stable-diffusion-webui\modules\scripts.py", line 263, in load_scripts
script_module = script_loading.load_module(scriptfile.path)
File "G:\stable-diffusion-webui\modules\script_loading.py", line 10, in load_module
module_spec.loader.exec_module(module)
File "
Error loading script: text2vid.py
Traceback (most recent call last):
File "G:\stable-diffusion-webui\modules\scripts.py", line 263, in load_scripts
script_module = script_loading.load_module(scriptfile.path)
File "G:\stable-diffusion-webui\modules\script_loading.py", line 10, in load_module
module_spec.loader.exec_module(module)
File "
Error loading script: api_t2v.py Traceback (most recent call last): File "G:\stable-diffusion-webui\modules\scripts.py", line 263, in load_scripts script_module = script_loading.load_module(scriptfile.path) File "G:\stable-diffusion-webui\modules\script_loading.py", line 10, in load_module module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "G:\stable-diffusion-webui\extensions\sd-webui-text2video\scripts\api_t2v.py", line 35, in from t2v_helpers.args import T2VArgs_sanity_check, T2VArgs, T2VOutputArgs File "G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\t2v_helpers\args.py", line 7, in from modelscope.t2v_model import has_torch2 ImportError: cannot import name 'has_torch2' from 'modelscope.t2v_model' (G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\modelscope\t2v_model.py)
Error loading script: text2vid.py Traceback (most recent call last): File "G:\stable-diffusion-webui\modules\scripts.py", line 263, in load_scripts script_module = script_loading.load_module(scriptfile.path) File "G:\stable-diffusion-webui\modules\script_loading.py", line 10, in load_module module_spec.loader.exec_module(module) File "", line 883, in exec_module File "", line 241, in _call_with_frames_removed File "G:\stable-diffusion-webui\extensions\sd-webui-text2video\scripts\text2vid.py", line 20, in from t2v_helpers.render import run File "G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\t2v_helpers\render.py", line 5, in from modelscope.process_modelscope import process_modelscope File "G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\modelscope\process_modelscope.py", line 21, in from t2v_helpers.args import get_outdir, process_args File "G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\t2v_helpers\args.py", line 7, in from modelscope.t2v_model import has_torch2 ImportError: cannot import name 'has_torch2' from 'modelscope.t2v_model' (G:\stable-diffusion-webui/extensions/sd-webui-text2video/scripts\modelscope\t2v_model.py)
Why are you sending an error in your console in my bug report?
It seems the issue is that unpickler is catching this as potentially malicious:
Error verifying pickled file from C:\Users\Herculean/.cache\huggingface\hub\models--openai--clip-vit-large-patch14\snapshots\8d052a0f05efbaefbc9e8786ba291cfdf93e5bff\pytorch_model.bin:
I've seen other people have similar problems with the safety check. First thing I would try is finding that file and deleting it. It could have only downloaded partially or somehow been corrupted. The file should download automatically again. See if that fixes your issue.
That being said, I'm not certain both tracebacks are related, but it's a good place to start.
Is there an existing issue for this?
Are you using the latest version of the extension?
What happened?
Running into not enough memory problems, can't figure out why. It was working before but not anymore for some reason.
Model Type: ModelScope
RuntimeError: [enforce fail at ..\c10\core\impl\alloc_cpu.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 52428800 bytes.
Model Type: VideoCrafter (WIP) Note: Refer to Settings section Image 1
RuntimeError: [enforce fail at ..\c10\core\impl\alloc_cpu.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 4194304 bytes.
Model Type: VideoCrafter (WIP) Note: Refer to Settings section Image 2
RuntimeError: [enforce fail at ..\c10\core\impl\alloc_cpu.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 19660800 bytes.
Steps to reproduce the problem
What should have happened?
It should've generated a video, but instead, it ran into a memory error . For this bug report purposes, I did "anime girl dancing" as the prompt. Instead of it generating the video based on the prompt, it shows a video showing "If you see this video it means an error check your logs".
WebUI and Deforum extension Commit IDs
WebUI commit id - v1.3.0-72-gb957dcfe txt2vid commit id - a8937bafb63867d15bd5e62afe31f431fcde1c49
Torch version
torch: 2.0.1+cu118
What GPU were you using for launching?
ASUS TUF Gaming GeForce RTX 3080 Ti with 12GB of VRAM
On which platform are you launching the webui backend with the extension?
Local PC setup (Windows)
Settings
(Here are the settings I've tried using for figuring this stuff out)
![image](https://github.com/deforum-art/sd-webui-text2video/assets/88630941/3f1c57b1-baa6-4eb0-b673-68ca7535226e)
Console logs
Additional information
I'm very new trying out things on my own with Stable Diffusion. I've followed the exact instructions given by TroubleChute's video tutorial and for the first couple of times with VideoCrafter it was working. After I did a restart, it suddenly refused to do anything due to memory issues.