|
|
在Colab 上用MimicTalk训练数字人
脚本地址：
https://colab.research.google.co ... rollTo=OjU021Om659c
地址2
https://colab.research.google.co ... 46S2g3V?usp=sharing
能顺利安装mmcv和pytorch3d
Building wheels for collected packages: mmcv
Building wheel for mmcv (setup.py) ... done
Created wheel for mmcv: filename=mmcv-2.1.0-cp39-cp39-linux_x86_64.whl size=1322078 sha256=55fe258d3108ce4a73093be7e6accbd37780e35774b6912a61fd85cd0687fe06
Stored in directory: /root/.cache/pip/wheels/eb/be/7a/1cbc064b54f14e1f1e5b9133bc0b7e416904b784e86b478e10
Successfully built mmcv
Installing collected packages: addict, tomli, termcolor, pyparsing, platformdirs, opencv-python, kiwisolver, importlib-resources, fonttools, cycler, contourpy, yapf, matplotlib, mmengine, mmcv
Successfully installed addict-2.4.0 contourpy-1.3.0 cycler-0.12.1 fonttools-4.55.3 importlib-resources-6.4.5 kiwisolver-1.4.7 matplotlib-3.9.3 mmcv-2.1.0 mmengine-0.10.5 opencv-python-4.10.0.84 platformdirs-4.3.6 pyparsing-3.2.0 termcolor-2.5.0 tomli-2.2.1 yapf-0.43.0
Collecting git+https://github.com/facebookresearch/pytorch3d.git@stable
Cloning https://github.com/facebookresearch/pytorch3d.git (to revision stable) to /tmp/pip-req-build-r7ge5w39
Running command git clone --filter=blob:none --quiet https://github.com/facebookresearch/pytorch3d.git /tmp/pip-req-build-r7ge5w39
Running command git checkout -q 75ebeeaea0908c5527e7b1e305fbc7681382db47
Resolved https://github.com/facebookresearch/pytorch3d.git to commit 75ebeeaea0908c5527e7b1e305fbc7681382db47
Preparing metadata (setup.py) ... done
Collecting iopath (from pytorch3d==0.7.8)
Downloading iopath-0.1.10.tar.gz (42 kB)
Preparing metadata (setup.py) ... done
Requirement already satisfied: tqdm in /content/local/miniconda/envs/mimictalk/lib/python3.9/site-packages (from iopath->pytorch3d==0.7.8) (4.65.2)
Requirement already satisfied: typing_extensions in /content/local/miniconda/envs/mimictalk/lib/python3.9/site-packages (from iopath->pytorch3d==0.7.8) (4.9.0)
Collecting portalocker (from iopath->pytorch3d==0.7.8)
Downloading portalocker-3.0.0-py3-none-any.whl.metadata (8.5 kB)
Downloading portalocker-3.0.0-py3-none-any.whl (19 kB)
Building wheels for collected packages: pytorch3d, iopath
Building wheel for pytorch3d (setup.py) ... done
Created wheel for pytorch3d: filename=pytorch3d-0.7.8-cp39-cp39-linux_x86_64.whl size=980119 sha256=55d70f0222733cde2c0e3e69ab5fb9530a994eaf4739e5746c353b3e6e2d249e
Stored in directory: /tmp/pip-ephem-wheel-cache-thub4ums/wheels/d7/e2/25/51fb3f170ac9f43bf7d1bd4bdbf67e4a34104adc2ecfe5f62e
Building wheel for iopath (setup.py) ... done
Created wheel for iopath: filename=iopath-0.1.10-py3-none-any.whl size=31545 sha256=8e1071c6aac511e5e1d4c5483b8bf98653e35aa883e11767f47454c811576d83
Stored in directory: /root/.cache/pip/wheels/c1/13/6d/441d8f2af76ee6d2a3e67eebb1d0c556fefcee0a8b32266a8e
Successfully built pytorch3d iopath
Installing collected packages: portalocker, iopath, pytorch3d
Successfully installed iopath-0.1.10 portalocker-3.0.0 pytorch3d-0.7.8
运行视频训练和推理前还是要修改源代码的
下面是修改方法：
Traceback (most recent call last):
File "/content/local/miniconda/envs/mimictalk/MimicTalk/inference/mimictalk_infer.py", line 18, in <module>
from utils.commons.hparams import hparams, set_hparams
ModuleNotFoundError: No module named 'utils'
在 inference/mimictalk_infer.py 和 inference/train_mimictalk_on_a_video.py文件中添加：
import sys
sys.path.append('./') #将当前目录 (./) 添加到 Python 模块搜索路径中，使得可以导入当前目录下的模块。
===========================================
安装组件时遇到的问题：
Building wheels for collected packages: pyaudio
Running command Building wheel for pyaudio (pyproject.toml)
running bdist_wheel
running build
running build_py
creating build/lib.linux-x86_64-cpython-39/pyaudio
copying src/pyaudio/__init__.py -> build/lib.linux-x86_64-cpython-39/pyaudio
running build_ext
building 'pyaudio._portaudio' extension
creating build/temp.linux-x86_64-cpython-39/src/pyaudio
gcc -pthread -B /content/local/miniconda/envs/mimictalk/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -O2 -Wall -fPIC -O2 -isystem /content/local/miniconda/envs/mimictalk/include -I/content/local/miniconda/envs/mimictalk/include -fPIC -O2 -isystem /content/local/miniconda/envs/mimictalk/include -fPIC -I/usr/local/include -I/usr/include -I/content/local/miniconda/envs/mimictalk/include/python3.9 -c src/pyaudio/device_api.c -o build/temp.linux-x86_64-cpython-39/src/pyaudio/device_api.o
src/pyaudio/device_api.c:9:10: fatal error: portaudio.h: No such file or directory
9 | #include "portaudio.h"
| ^~~~~~~~~~~~~
compilation terminated.
error: command '/usr/bin/gcc' failed with exit code 1
error: subprocess-exited-with-error
× Building wheel for pyaudio (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
full command: /content/local/miniconda/envs/mimictalk/bin/python /content/local/miniconda/envs/mimictalk/lib/python3.9/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py build_wheel /tmp/tmpy328lh56
cwd: /tmp/pip-install-1703o6g1/pyaudio_66d22200ef194f3986428a21635963fd
Building wheel for pyaudio (pyproject.toml) ... error
ERROR: Failed building wheel for pyaudio
Failed to build pyaudio
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (pyaudio)
问题出在安装 pyaudio 时无法找到 portaudio.h 文件，这个文件是 PyAudio 库所依赖的 PortAudio 库的头文件。以下是问题的原因和解决方法：
安装 portaudio 系统依赖
在安装 PyAudio 之前，您需要确保 PortAudio 库已安装。运行以下命令安装：
# 在 Colab 或 Linux 系统中安装 PortAudio
!apt-get install -y portaudio19-dev
===========================
WARNING: The following packages were previously imported in this runtime:
[_distutils_hack,setuptools]
You must restart the runtime in order to use newly installed versions.
此警告表明，在当前的 Colab 运行时中已经加载了一些与新安装的包相关的模块（例如 setuptools 和 _distutils_hack），但这些模块是旧版本的。新安装的版本无法覆盖已经加载的旧模块，因此需要重启运行时以便使用新安装的包 |
|
