moaeddy
commented
3 months ago pip install pytorch
Closed inobyte closed 3 months ago
moaeddy
commented
3 months ago pip install pytorch
mars7233
commented
3 months ago Traceback (most recent call last): File "D:\Download\Deep-Live-Cam\Deep-Live-Cam-DML\run.py", line 3, in from modules import core File "D:\Download\Deep-Live-Cam\Deep-Live-Cam-DML\modules\core.py", line 14, in import torch ModuleNotFoundError: No module named 'torch'
i installed/reinstalled everything, but it slill does not work And yes, a have toutch installed
After I reused conda to download all the dependent packages, I solved a series of problems such as tourch and other dependencies.
ochristianlove
commented
3 weeks ago HERE IS A SOLUTION FROM FIX I HAD SAME WITH WINWON 11
Steps to Fix: 1) Activate your virtual environment (if not already activated):
C:\Users\Christian\Deep-Live-Cam>venv\Scripts\activate
venv\Scripts\activate
2) Install numpy: Run this command to install numpy in your virtual environment:
pip install numpy
3) pip install onnxruntime:
pip install onnxruntime
4)Run the script again: Once the required modules are installed, try running your script again:
python run.py --execution-provider cuda
YOU MAY GET THIS ERROR
File "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THE ABOVE ERROR:
nstall tensorflow: Use the following command to install the TensorFlow package:
pip install tensorflow
Note: If you're using a GPU and want to leverage it for TensorFlow, you should install tensorflow-gpu instead:
pip install tensorflow-gpu
Run the script again: After installing TensorFlow, run your script again:
python run.py --execution-provider cuda
AFATER THIS YOU MAY STILL GET ERROR:
ile "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THIS:
Install customtkinter: You can install the customtkinter package by running:
pip install customtkinter
Run the script again: After installing the package, try running the script again:
python run.py --execution-provider cuda
YOUR MAY GET THIS ERROR:
(venv) C:\Users\Christian\Deep-Live-Cam>python run.py --execution-provider cuda
Traceback (most recent call last):
File "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THIS:
Install OpenCV (cv2): You can install OpenCV using the following command:
pip install opencv-python
If you also need additional OpenCV functionalities (like video processing support), you might want to install the following package as well:
pip install opencv-python-headless
Run the script again: After installing OpenCV, try running the script again:
python run.py --execution-provider cuda
YOU MAY GET THIS ERROR:
venv) C:\Users\Christian\Deep-Live-Cam>python run.py --execution-provider cuda
Traceback (most recent call last):
File "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THIS:
Install cv2_enumerate_cameras: You can try installing it using pip:
pip install cv2-enumerate-cameras
If cv2_enumerate_cameras is not available on PyPI, it may be part of a custom module or repository. In that case, you may need to:
Check the project's documentation or code repository for installation instructions.
Alternatively, search for the GitHub repository (or similar) for the module and install it from there.
For example, if it’s hosted on GitHub:
pip install git+https://github.com/username/cv2_enumerate_cameras.git
Run the script again: Once the module is installed, try running the script again:
python run.py --execution-provider cuda
IT MAY SHOW YOU THIS ERROR
venv) C:\Users\Christian\Deep-Live-Cam>python run.py --execution-provider cuda
Traceback (most recent call last):
File "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THIS:
Install Pillow (which includes PIL): Run the following command to install Pillow:
pip install Pillow
Run the script again: After installing Pillow, try running your script again:
python run.py --execution-provider cuda
IT MAY SHOW YOU THIS ERROR
aceback (most recent call last):
File "C:\Users\Christian\Deep-Live-Cam\run.py", line 3, in
TO FIX THIS:
Install insightface: Install the package using pip:
pip install insightface
Run the script again: After the installation completes, try running your script again:
python run.py --execution-provider cuda
IT MAY SHOW YOU THIS ERROR:
uilding 'insightface.thirdparty.face3d.mesh.cython.mesh_core_cython' extension error: Microsoft Visual C++ 14.0 or greater is required. Get it with "Microsoft C++ Build Tools": https://visualstudio.microsoft.com/visual-cpp-build-tools/ [end of output]
note: This error originates from a subprocess, and is likely not a problem with pip. ERROR: Failed building wheel for insightface Failed to build insightface ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (insightface)
TO FIX THIS: The error you're encountering indicates that a C++ compiler is required to build the insightface package from source. Specifically, you need Microsoft Visual C++ 14.0 or greater. It happen because didn't check the box for c++ (at the right hand side) during the installer so make sure to check the box for "Desktop development with C++" during installation, download check the box (all c++ build tools and only 1 win 11 sdk) then lunch and continue the fix on Deep Live terminal
3) Install insightface again: After the build tools are installed, you can attempt to install insightface again:
pip install insightface
Run the script again: After successful installation, try running your script:
python run.py --execution-provider cuda
YOU MAY GET THIS ERROR: RROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. tensorboard 2.10.1 requires protobuf<3.20,>=3.9.2, but you have protobuf 5.28.2 which is incompatible. tensorflow-gpu 2.10.1 requires protobuf<3.20,>=3.9.2, but you have protobuf 5.28.2 which is incompatible. tensorflow-intel 2.17.0 requires keras>=3.2.0, but you have keras 2.10.0 which is incompatible. tensorflow-intel 2.17.0 requires protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3, but you have protobuf 5.28.2 which is incompatible. tensorflow-intel 2.17.0 requires tensorboard<2.18,>=2.17, but you have tensorboard 2.10.1 which is incompatible. Successfully installed PyYAML-6.0.2 albucore-0.0.17 albumentations-1.4.18 annotated-types-0.7.0 colorama-0.4.6 contourpy-1.3.0 cycler-0.12.1 cython-3.0.11 easydict-1.13 eval-type-backport-0.2.0 fonttools-4.54.1 imageio-2.36.0 insightface-0.7.3 joblib-1.4.2 kiwisolver-1.4.7 lazy-loader-0.4 matplotlib-3.9.2 onnx-1.17.0 prettytable-3.11.0 protobuf-5.28.2 pydantic-2.9.2 pydantic-core-2.23.4 pyparsing-3.2.0 python-dateutil-2.9.0.post0 scikit-image-0.24.0 scikit-learn-1.5.2 scipy-1.14.1 threadpoolctl-3.5.0 tifffile-2024.9.20 tqdm-4.66.5 wcwidth-0.2.13
TO FIX THIS:
Uninstall conflicting packages: First, let's uninstall the problematic packages:
pip uninstall protobuf tensorboard keras tensorflow-gpu tensorflow-intel
(KERAS and tensorflow only uninstall, it may bring other options to uninstall along)
Install compatible versions: Now, install versions that are compatible with each other:
Install the required version of protobuf:
pip install protobuf==3.20.3
Install the required version of tensorboard (for TensorFlow 2.17.0):
pip install tensorboard==2.17.0
Install a compatible version of tensorflow-gpu or tensorflow-intel (depending on your hardware setup). If you're using a GPU:
pip install tensorflow-gpu==2.17.0
(if it show small error ignore, "Could not find a version that satisfies the requirement' but it works)
If you prefer the Intel-optimized version:
pip install tensorflow-intel==2.17.0
Install the compatible version of keras:
pip install keras==3.2.0
Run your script again: After fixing these dependencies, try running your script again:
python run.py --execution-provider cuda
YOU MAY GET THIS ERROR (venv) C:\Users\Christian\Deep-Live-Cam>python run.py --execution-provider cuda usage: run.py [-h] [-s SOURCE_PATH] [-t TARGET_PATH] [-o OUTPUT_PATH] [--frame-processor {face_swapper,face_enhancer} [{face_swapper,face_enhancer} ...]] [--keep-fps] [--keep-audio] [--keep-frames] [--many-faces] [--nsfw-filter] [--map-faces] [--video-encoder {libx264,libx265,libvpx-vp9}] [--video-quality [0-51]] [--live-mirror] [--live-resizable] [--max-memory MAX_MEMORY] [--execution-provider {azure,cpu} [{azure,cpu} ...]] [--execution-threads EXECUTION_THREADS] [-v] run.py: error: argument --execution-provider: invalid choice: 'cuda' (choose from 'azure', 'cpu')
TO FIX THIS: Note that the cuda run code "python run.py --execution-provider cuda " it trying to use the gpu in my pc, this above error show i haven't intalled the requirement for this command and that only cup command can works, for those who dont have gpu you can lunch the Deep Live from here with this command
python run.py --execution-provider cpu
For those with GPU lets proceed to install dependencies with for gpu for this command to work
1) ensure nvidia cuda gpu is installed or repeat to ensure it works : Install CUDA Toolkit 11.8 from
lunch cuda toolkit and let it run
2) Install dependencies:
uninstall Existing onnxruntime and onnxruntime-gpu: This ensures that you have a clean environment before installing the specific version:
pip uninstall onnxruntime onnxruntime-gpu
Install the Correct Version of onnxruntime-gpu: Install the GPU-enabled version:
pip install onnxruntime-gpu==1.16.3
3) Run the Script: Now try running your script with the cuda execution provider:
python run.py --execution-provider cuda
it should work now, enjoy cheers
hacksider
commented
3 weeks ago @ochristianlove this is awesome! Can you post this on the Wiki as well? Thanks a lot bro!
ochristianlove
commented
3 weeks ago Also i discovered that even after this it conflicting with older versions of installed requirements files, so it didn't work properly
TO FIX:
1) Delete all directory of deeo live cam
2)cd path\to\your\project
3) clone the url from git
get clone htts://URL
5) cd to the downloaded clone directory (deep live cam)
😂 Am not with my pc so bear with me typing from my phone
And do all this in any prompt Terminal I used powershell as administrator
3) create virtual environment
python -m venv venv
4) activate venviromen
venv\Scripts\activate
5) it came with it own default requirement.txt so we just make the command
pip install -r requirements.txt
6) download the model and put in the model directory in deep live cam folde, also note: inswapper_128_fp16.onnx will be rename again in the model path to old in inswapper_128.onnx (am not sure i renamed were well but the 4kb file is there just use the same file name)
7) uninstall and reinstall
pip uninstall onnxruntime onnxruntime-gpu
pip install onnxruntime-gpu==1.16.3
8)- run the program
python run.py --execution-provider cuda
Enjoy (am very impressed 😁 with this tech) tip Enable many faces to keep constant track with your face, bye
hacksider
commented
3 weeks ago Thanks @ochristianlove I will put this on the wiki (will credit you with it as well). Thanks a lot for the help!
Traceback (most recent call last): File "D:\Download\Deep-Live-Cam\Deep-Live-Cam-DML\run.py", line 3, in
from modules import core
File "D:\Download\Deep-Live-Cam\Deep-Live-Cam-DML\modules\core.py", line 14, in
import torch
ModuleNotFoundError: No module named 'torch'
i installed/reinstalled everything, but it slill does not work And yes, a have toutch installed