本文分享自华为云社区《Open-Sora 文生视频原来在AI Gallery上也能体验了》，作者：码上开花_Lancer。
体验链接：Open-Sora 文生视频案例体验
不久前，OpenAI Sora 凭借其惊人的视频生成效果迅速走红，在一堆文本转视频模型中脱颖而出，成为全球关注的焦点。之后，Colossal-AI团队又推出了新的开源解决方案“Open-Sora 1.0”，涵盖了整个训练过程，包括数据处理、所有训练细节和模型检查点，与世界各地的AI爱好者携手推进视频创作的新时代。
详细内容请参考：https://hpc-ai.com/blog/open-sora-v1.0
2024年4月份又更新了Open-Sora 1.1，它可以生成2s~15s，144p到720p分辨率的视频，支持文本到图像、文本到视频以及图像到视频的生成，让我们来看看Open-Sora 1.1的实际视频生成效果：
案例体验
🔹 本案例需使用 Pytorch-2.0.1 GPU-V100 及以上规格运行
🔹 点击Run in ModelArts，将会进入到ModelArts CodeLab中，这时需要你登录华为云账号，如果没有账号，则需要注册一个，且要进行实名认证，参考《如何创建华为云账号并且实名认证》 即可完成账号注册和实名认证。 登录之后，等待片刻，即可进入到CodeLab的运行环境
🔹 出现 Out Of Memory ，请检查是否为您的参数配置过高导致，修改参数配置，重启kernel或更换更高规格资源进行规避❗❗❗
1. 下载代码和模型
此处运行大约需要1分钟，请耐心等待！
import os import moxing as mox if not os.path.exists('Open-Sora'):mox.file.copy_parallel('obs://modelbox-course/open-sora_1.1/Open-Sora', 'Open-Sora')if not os.path.exists('/home/ma-user/.cache/huggingface'):mox.file.copy_parallel('obs://modelbox-course/huggingface', '/home/ma-user/.cache/huggingface')if not os.path.exists('Open-Sora/opensora/models/sd-vae-ft-ema'):mox.file.copy_parallel('obs://modelbox-course/sd-vae-ft-ema', 'Open-Sora/opensora/models/sd-vae-ft-ema') if not os.path.exists('Open-Sora/opensora/models/text_encoder/t5-v1_1-xxl'):mox.file.copy_parallel('obs://modelbox-course/t5-v1_1-xxl', 'Open-Sora/opensora/models/text_encoder/t5-v1_1-xxl')if not os.path.exists('/home/ma-user/work/t5.py'):mox.file.copy_parallel('obs://modelbox-course/open-sora_1.1/t5.py', '/home/ma-user/work/t5.py')if not os.path.exists('Open-Sora/opus-mt-zh-en'):mox.file.copy_parallel('obs://modelarts-labs-bj4-v2/course/ModelBox/opus-mt-zh-en', 'Open-Sora/opus-mt-zh-en')if not os.path.exists('/home/ma-user/work/frpc_linux_amd64'):mox.file.copy_parallel('obs://modelarts-labs-bj4-v2/course/ModelBox/frpc_linux_amd64', '/home/ma-user/work/frpc_linux_amd64')
INFO:root:Using MoXing-v2.1.6.879ab2f4-879ab2f4 INFO:root:List OBS time cost: 0.02 seconds. INFO:root:Copy parallel total time cost: 41.71 seconds. INFO:root:List OBS time cost: 0.14 seconds. INFO:root:Copy parallel total time cost: 2.91 seconds.
2. 配置运行环境
本案例依赖Python3.10.10及以上环境，因此我们首先创建虚拟环境：
!/home/ma-user/anaconda3/bin/conda clean -i !/home/ma-user/anaconda3/bin/conda create -n python-3.10.10 python=3.10.10 -y --override-channels --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main !/home/ma-user/anaconda3/envs/python-3.10.10/bin/pip install ipykernel
/home/ma-user/anaconda3/lib/python3.7/site-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.15) or chardet (3.0.4) doesn't match a supported version! RequestsDependencyWarning) /home/ma-user/anaconda3/lib/python3.7/site-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.15) or chardet (3.0.4) doesn't match a supported version! RequestsDependencyWarning) Collecting package metadata (current_repodata.json): done Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source. Collecting package metadata (repodata.json): done Solving environment: done ## Package Plan ## environment location: /home/ma-user/anaconda3/envs/python-3.10.10 added / updated specs:- python=3.10.10
The following packages will be downloaded: package | build---------------------------|-----------------_libgcc_mutex-0.1 | main 3 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main_openmp_mutex-5.1 | 1_gnu 21 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainbzip2-1.0.8 | h5eee18b_6 262 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainca-certificates-2024.3.11 | h06a4308_0 127 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainld_impl_linux-64-2.38 | h1181459_1 654 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainlibffi-3.4.4 | h6a678d5_1 141 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainlibgcc-ng-11.2.0 | h1234567_1 5.3 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainlibgomp-11.2.0 | h1234567_1 474 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainlibstdcxx-ng-11.2.0 | h1234567_1 4.7 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainlibuuid-1.41.5 | h5eee18b_0 27 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainncurses-6.4 | h6a678d5_0 914 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainopenssl-1.1.1w | h7f8727e_0 3.7 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainpip-24.0 | py310h06a4308_0 2.7 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainpython-3.10.10 | h7a1cb2a_2 26.9 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainreadline-8.2 | h5eee18b_0 357 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainsetuptools-69.5.1 | py310h06a4308_0 1012 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainsqlite-3.45.3 | h5eee18b_0 1.2 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/maintk-8.6.14 | h39e8969_0 3.4 MB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/maintzdata-2024a | h04d1e81_0 116 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainwheel-0.43.0 | py310h06a4308_0 110 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainxz-5.4.6 | h5eee18b_1 643 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/mainzlib-1.2.13 | h5eee18b_1 111 KB https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main------------------------------------------------------------Total: 52.8 MB The following NEW packages will be INSTALLED: _libgcc_mutex anaconda/pkgs/main/linux-64::_libgcc_mutex-0.1-main_openmp_mutex anaconda/pkgs/main/linux-64::_openmp_mutex-5.1-1_gnubzip2 anaconda/pkgs/main/linux-64::bzip2-1.0.8-h5eee18b_6ca-certificates anaconda/pkgs/main/linux-64::ca-certificates-2024.3.11-h06a4308_0ld_impl_linux-64 anaconda/pkgs/main/linux-64::ld_impl_linux-64-2.38-h1181459_1libffi anaconda/pkgs/main/linux-64::libffi-3.4.4-h6a678d5_1libgcc-ng anaconda/pkgs/main/linux-64::libgcc-ng-11.2.0-h1234567_1libgomp anaconda/pkgs/main/linux-64::libgomp-11.2.0-h1234567_1libstdcxx-ng anaconda/pkgs/main/linux-64::libstdcxx-ng-11.2.0-h1234567_1libuuid anaconda/pkgs/main/linux-64::libuuid-1.41.5-h5eee18b_0ncurses anaconda/pkgs/main/linux-64::ncurses-6.4-h6a678d5_0openssl anaconda/pkgs/main/linux-64::openssl-1.1.1w-h7f8727e_0pip anaconda/pkgs/main/linux-64::pip-24.0-py310h06a4308_0python anaconda/pkgs/main/linux-64::python-3.10.10-h7a1cb2a_2readline anaconda/pkgs/main/linux-64::readline-8.2-h5eee18b_0setuptools anaconda/pkgs/main/linux-64::setuptools-69.5.1-py310h06a4308_0sqlite anaconda/pkgs/main/linux-64::sqlite-3.45.3-h5eee18b_0tk anaconda/pkgs/main/linux-64::tk-8.6.14-h39e8969_0tzdata anaconda/pkgs/main/noarch::tzdata-2024a-h04d1e81_0wheel anaconda/pkgs/main/linux-64::wheel-0.43.0-py310h06a4308_0xz anaconda/pkgs/main/linux-64::xz-5.4.6-h5eee18b_1zlib anaconda/pkgs/main/linux-64::zlib-1.2.13-h5eee18b_1
Downloading and Extracting Packages libffi-3.4.4 | 141 KB | ##################################### | 100% _openmp_mutex-5.1 | 21 KB | ##################################### | 100% xz-5.4.6 | 643 KB | ##################################### | 100% tzdata-2024a | 116 KB | ##################################### | 100% _libgcc_mutex-0.1 | 3 KB | ##################################### | 100% zlib-1.2.13 | 111 KB | ##################################### | 100% bzip2-1.0.8 | 262 KB | ##################################### | 100% libuuid-1.41.5 | 27 KB | ##################################### | 100% ca-certificates-2024 | 127 KB | ##################################### | 100% libstdcxx-ng-11.2.0 | 4.7 MB | ##################################### | 100% ncurses-6.4 | 914 KB | ##################################### | 100% openssl-1.1.1w | 3.7 MB | ##################################### | 100% wheel-0.43.0 | 110 KB | ##################################### | 100% python-3.10.10 | 26.9 MB | ##################################### | 100% pip-24.0 | 2.7 MB | ##################################### | 100% readline-8.2 | 357 KB | ##################################### | 100% tk-8.6.14 | 3.4 MB | ##################################### | 100% setuptools-69.5.1 | 1012 KB | ##################################### | 100% libgcc-ng-11.2.0 | 5.3 MB | ##################################### | 100% ld_impl_linux-64-2.3 | 654 KB | ##################################### | 100% libgomp-11.2.0 | 474 KB | ##################################### | 100% sqlite-3.45.3 | 1.2 MB | ##################################### | 100% Preparing transaction: done Verifying transaction: done Executing transaction: done # # To activate this environment, use # # $ conda activate python-3.10.10 # # To deactivate an active environment, use # # $ conda deactivate Looking in indexes: http://repo.myhuaweicloud.com/repository/pypi/simple Collecting ipykernelDownloading http://repo.myhuaweicloud.com/repository/pypi/packages/53/9d/40d5207db523363d9b5698f33778c18b0d591e3fdb6e0116b894b2a2491c/ipykernel-6.29.4-py3-none-any.whl (117 kB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m117.1/117.1 kB•[0m •[31m10.6 MB/s•[0m eta •[36m0:00:00•[0m ...... Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/80/03/6ea8b1b2a5ab40a7a60dc464d3daa7aa546e0a74d74a9f8ff551ea7905db/executing-2.0.1-py2.py3-none-any.whl (24 kB) Collecting asttokens>=2.1.0 (from stack-data->ipython>=7.23.1->ipykernel)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/45/86/4736ac618d82a20d87d2f92ae19441ebc7ac9e7a581d7e58bbe79233b24a/asttokens-2.4.1-py2.py3-none-any.whl (27 kB) Collecting pure-eval (from stack-data->ipython>=7.23.1->ipykernel)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/2b/27/77f9d5684e6bce929f5cfe18d6cfbe5133013c06cb2fbf5933670e60761d/pure_eval-0.2.2-py3-none-any.whl (11 kB) Installing collected packages: wcwidth, pure-eval, ptyprocess, typing-extensions, traitlets, tornado, six, pyzmq, pygments, psutil, prompt-toolkit, platformdirs, pexpect, parso, packaging, nest-asyncio, executing, exceptiongroup, decorator, debugpy, python-dateutil, matplotlib-inline, jupyter-core, jedi, comm, asttokens, stack-data, jupyter-client, ipython, ipykernel Successfully installed asttokens-2.4.1 comm-0.2.2 debugpy-1.8.1 decorator-5.1.1 exceptiongroup-1.2.1 executing-2.0.1 ipykernel-6.29.4 ipython-8.25.0 jedi-0.19.1 jupyter-client-8.6.2 jupyter-core-5.7.2 matplotlib-inline-0.1.7 nest-asyncio-1.6.0 packaging-24.0 parso-0.8.4 pexpect-4.9.0 platformdirs-4.2.2 prompt-toolkit-3.0.46 psutil-5.9.8 ptyprocess-0.7.0 pure-eval-0.2.2 pygments-2.18.0 python-dateutil-2.9.0.post0 pyzmq-26.0.3 six-1.16.0 stack-data-0.6.3 tornado-6.4 traitlets-5.14.3 typing-extensions-4.12.1 wcwidth-0.2.13
import json import os data = {"display_name": "python-3.10.10","env": {"PATH": "/home/ma-user/anaconda3/envs/python-3.10.10/bin:/home/ma-user/anaconda3/envs/python-3.7.10/bin:/modelarts/authoring/notebook-conda/bin:/opt/conda/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/home/ma-user/modelarts/ma-cli/bin:/home/ma-user/modelarts/ma-cli/bin:/home/ma-user/anaconda3/envs/PyTorch-1.8/bin"},"language": "python","argv": ["/home/ma-user/anaconda3/envs/python-3.10.10/bin/python","-m","ipykernel","-f","{connection_file}"] } if not os.path.exists("/home/ma-user/anaconda3/share/jupyter/kernels/python-3.10.10/"):os.mkdir("/home/ma-user/anaconda3/share/jupyter/kernels/python-3.10.10/") with open('/home/ma-user/anaconda3/share/jupyter/kernels/python-3.10.10/kernel.json', 'w') as f:json.dump(data, f, indent=4)
conda env list
/home/ma-user/anaconda3/lib/python3.7/site-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.15) or chardet (3.0.4) doesn't match a supported version! RequestsDependencyWarning) # conda environments: # base * /home/ma-user/anaconda3 python-3.10.10 /home/ma-user/anaconda3/envs/python-3.10.10 python-3.7.10 /home/ma-user/anaconda3/envs/python-3.7.10
Note: you may need to restart the kernel to use updated packages.
创建完成后，稍等片刻，或刷新页面，点击右上角kernel选择python-3.10.10
查看Python版本
!python -V
Python 3.10.10
检查可用GPU，至少需要32GB显存
!nvidia-smi
Wed Jun 5 16:22:37 2024 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 470.57.02 Driver Version: 470.57.02 CUDA Version: 11.4 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 Tesla V100-PCIE... On | 00000000:00:0D.0 Off | 0 | | N/A 28C P0 25W / 250W | 0MiB / 32510MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------++-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | No running processes found | +-----------------------------------------------------------------------------+
安装依赖包
!pip install --upgrade pip !pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 xformers==0.0.22
Looking in indexes: http://repo.myhuaweicloud.com/repository/pypi/simple Requirement already satisfied: pip in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (24.0) Looking in indexes: http://repo.myhuaweicloud.com/repository/pypi/simple Collecting torch==2.0.1Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/8c/4d/17e07377c9c3d1a0c4eb3fde1c7c16b5a0ce6133ddbabc08ceef6b7f2645/torch-2.0.1-cp310-cp310-manylinux1_x86_64.whl (619.9 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m619.9/619.9 MB•[0m •[31m8.2 MB/s•[0m eta •[36m0:00:00•[0m00:01•[0m00:01•[0m •[?25hCollecting torchvision==0.15.2Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/87/0f/88f023bf6176d9af0f85feedf4be129f9cf2748801c4d9c690739a10c100/torchvision-0.15.2-cp310-cp310-manylinux1_x86_64.whl (6.0 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m6.0/6.0 MB•[0m •[31m109.5 MB/s•[0m eta •[36m0:00:00•[0ma •[36m0:00:01•[0m •[?25hCollecting torchaudio==2.0.2Downloading •[?25hCollecting certifi>=2017.4.17 (from requests->torchvision==0.15.2)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/5b/11/1e78951465b4a225519b8c3ad29769c49e0d8d157a070f681d5b6d64737f/certifi-2024.6.2-py3-none-any.whl (164 kB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m164.4/164.4 kB•[0m •[31m23.1 MB/s•[0m eta •[36m0:00:00•[0m •[?25hCollecting mpmath<1.4.0,>=1.1.0 (from sympy->torch==2.0.1)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl (536 kB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m536.2/536.2 kB•[0m •[31m32.8 MB/s•[0m eta •[36m0:00:00•[0m •[?25hInstalling collected packages: mpmath, lit, urllib3, sympy, pillow, nvidia-nvtx-cu11, nvidia-nccl-cu11, nvidia-cusparse-cu11, nvidia-curand-cu11, nvidia-cufft-cu11, nvidia-cuda-runtime-cu11, nvidia-cuda-nvrtc-cu11, nvidia-cuda-cupti-cu11, nvidia-cublas-cu11, numpy, networkx, MarkupSafe, idna, filelock, cmake, charset-normalizer, certifi, requests, nvidia-cusolver-cu11, nvidia-cudnn-cu11, jinja2, triton, torch, xformers, torchvision, torchaudio Successfully installed MarkupSafe-2.1.5 certifi-2024.6.2 charset-normalizer-3.3.2 cmake-3.29.3 filelock-3.14.0 idna-3.7 jinja2-3.1.4 lit-18.1.6 mpmath-1.3.0 networkx-3.3 numpy-1.26.4 nvidia-cublas-cu11-11.10.3.66 nvidia-cuda-cupti-cu11-11.7.101 nvidia-cuda-nvrtc-cu11-11.7.99 nvidia-cuda-runtime-cu11-11.7.99 nvidia-cudnn-cu11-8.5.0.96 nvidia-cufft-cu11-10.9.0.58 nvidia-curand-cu11-10.2.10.91 nvidia-cusolver-cu11-11.4.0.1 nvidia-cusparse-cu11-11.7.4.91 nvidia-nccl-cu11-2.14.3 nvidia-nvtx-cu11-11.7.91 pillow-10.3.0 requests-2.32.3 sympy-1.12.1 torch-2.0.1 torchaudio-2.0.2 torchvision-0.15.2 triton-2.0.0 urllib3-2.2.1 xformers-0.0.22
%cd Open-Sora
/home/ma-user/work/ma_share/open-spra_1/Open-Sora
/home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/IPython/core/magics/osm.py:417: UserWarning: This is now an optional IPython functionality, setting dhist requires you to install the `pickleshare` library.self.shell.db['dhist'] = compress_dhist(dhist)[-100:]
'/home/ma-user/work/ma_share/open-spra_1/Open-Sora'
!pip install colossalai==0.3.6 accelerate==0.29.2 diffusers==0.27.2 ftfy==6.2.0 gdown==5.1.0 mmengine==0.10.3 pre-commit==3.7.0 pyav==12.0.5 tensorboard==2.16.2 timm==0.9.16 transformers==4.39.3 wandb==0.16.6
Looking in indexes: http://repo.myhuaweicloud.com/repository/pypi/simple Collecting colossalai==0.3.6Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/05/ed/57e80620ea8e35c3aa63a3207720b1890700fd12eea38b6592e9833e5c1b/colossalai-0.3.6.tar.gz (1.1 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m1.1/1.1 MB•[0m •[31m36.5 MB/s•[0m eta •[36m0:00:00•[0m •[?25h Preparing metadata (setup.py) ... •[?25ldone •[?25hCollecting accelerate==0.29.2Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/1b/e8/2fc7af3fa77ddac89a9c9b390d2d31d1db0612247ba2274009946959604e/accelerate-0.29.2-py3-none-any.whl (297 kB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m297.4/297.4 kB•[0m •[31m14.5 MB/s•[0m eta •[36m0:00:00•[0m •[?25hCollecting diffusers==0.27.2Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/75/c5/3b84fd731dd93c549a0c25657e4ce5a957aeccd32d60dba2958cd3cdac23/diffusers-0.27.2-py3-none-any.whl (2.0 MB)
!pip install .
Looking in indexes: http://repo.myhuaweicloud.com/repository/pypi/simple Processing /home/ma-user/work/ma_share/open-spra_1/Open-SoraPreparing metadata (setup.py) ... •[?25ldone •[?25hRequirement already satisfied: colossalai in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.3.6) Requirement already satisfied: accelerate in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.29.2) Requirement already satisfied: diffusers in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.27.2) Requirement already satisfied: ftfy in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (6.2.0) Requirement already satisfied: gdown in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (5.1.0) Requirement already satisfied: mmengine in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.10.3) Collecting pandas (from opensora==1.1.0)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/89/1b/12521efcbc6058e2673583bb096c2b5046a9df39bd73eca392c1efed24e5/pandas-2.2.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.0 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m13.0/13.0 MB•[0m •[31m60.4 MB/s•[0m eta •[36m0:00:00•[0m00:01•[0m00:01•[0m •[?25hRequirement already satisfied: pre-commit in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (3.7.0) Collecting pyarrow (from opensora==1.1.0)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/91/83/57572c088ec185582f04b607d545a4a6ef7599c0a3c1e60d397743b0d609/pyarrow-16.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (40.9 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m40.9/40.9 MB•[0m •[31m36.9 MB/s•[0m eta •[36m0:00:00•[0m00:01•[0m00:01•[0m •[?25hCollecting av (from opensora==1.1.0)Downloading http://repo.myhuaweicloud.com/repository/pypi/packages/0a/11/2b501d0a4de22826217a0b909e832f52fb5d503df50f424f3e31023a7bcc/av-12.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (34.3 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m34.3/34.3 MB•[0m •[31m96.1 MB/s•[0m eta •[36m0:00:00•[0m00:01•[0m00:01•[0m •[?25hRequirement already satisfied: tensorboard in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (2.16.2) Requirement already satisfied: timm in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.9.16) Requirement already satisfied: tqdm in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (4.66.4) Requirement already satisfied: transformers in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (4.39.3) Requirement already satisfied: wandb in /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages (from opensora==1.1.0) (0.16.6) Collecting rotary_embedding_torch (from opensora==1.1.0)Downloading Building wheels for collected packages: opensora, pandarallelBuilding wheel for opensora (setup.py) ... •[?25ldone •[?25h Created wheel for opensora: filename=opensora-1.1.0-py3-none-any.whl size=195249 sha256=86c66de7ded305b2e4fb07992d0147c0408086cc31cdc31d97bcea44d8f69596Stored in directory: /home/ma-user/.cache/pip/wheels/ae/34/85/7f84dd36f2e448d8d4455272d3358f557d0a570011d1701074Building wheel for pandarallel (setup.py) ... •[?25ldone •[?25h Created wheel for pandarallel: filename=pandarallel-1.6.5-py3-none-any.whl size=16673 sha256=b97386c92d34443f19cc88ea717c6cca143ef2b8f1f1ac79f4645c37d230bafcStored in directory: /home/ma-user/.cache/pip/wheels/f6/dd/25/a1c3775e721641ff67c71b3652e901e7e52611c6c3091784c9 Successfully built opensora pandarallel Installing collected packages: pytz, tzdata, pyarrow, dill, beartype, av, pandas, pandarallel, rotary_embedding_torch, opensora Successfully installed av-12.1.0 beartype-0.18.5 dill-0.3.8 opensora-1.1.0 pandarallel-1.6.5 pandas-2.2.2 pyarrow-16.1.0 pytz-2024.1 rotary_embedding_torch-0.6.2 tzdata-2024.1
!pip install spaces gradio MoviePy -i https://pypi.tuna.tsinghua.edu.cn/simple --trusted-host pypi.tuna.tsinghua.edu.cn !cp /home/ma-user/work/frpc_linux_amd64 /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/gradio/frpc_linux_amd64_v0.2 !chmod +x /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/gradio/frpc_linux_amd64_v0.2
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Collecting spacesDownloading https://pypi.tuna.tsinghua.edu.cn/packages/b2/3c/6205090507ea96e6e56d0deda8d0fc4c507026ef3772e55b637a5d0b7c61/spaces-0.28.3-py3-none-any.whl (18 kB) Collecting gradioDownloading https://pypi.tuna.tsinghua.edu.cn/packages/d1/37/f49320600cdf1fa856cc605a2e20e9debd34b5425b53f49abdb2ea463716/gradio-4.32.2-py3-none-any.whl (12.3 MB) •[2K •[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━•[0m •[32m12.3/12.3 MB•[0m •[31m5.2 MB/s•[0m eta •[36m0:00:00•[0m00:01•[0m00:01•[0m Successfully uninstalled decorator-5.1.1 •[31mERROR: 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. fabric 3.2.2 requires decorator>=5, but you have decorator 4.4.2 which is incompatible.•[0m•[31m •[0mSuccessfully installed MoviePy-1.0.3 aiofiles-23.2.1 altair-5.3.0 anyio-4.4.0 decorator-4.4.2 dnspython-2.6.1 email_validator-2.1.1 fastapi-0.111.0 fastapi-cli-0.0.4 ffmpy-0.3.2 gradio-4.32.2 gradio-client-0.17.0 h11-0.14.0 httpcore-1.0.5 httptools-0.6.1 httpx-0.27.0 imageio-2.34.1 imageio_ffmpeg-0.5.1 importlib-resources-6.4.0 orjson-3.10.3 proglog-0.1.10 pydub-0.25.1 python-dotenv-1.0.1 python-multipart-0.0.9 ruff-0.4.7 semantic-version-2.10.0 shellingham-1.5.4 sniffio-1.3.1 spaces-0.28.3 starlette-0.37.2 tomlkit-0.12.0 toolz-0.12.1 typer-0.12.3 ujson-5.10.0 uvicorn-0.30.1 uvloop-0.19.0 watchfiles-0.22.0 websockets-11.0.3
3. 生成视频
修改模型配置文件：
%%writefile configs/opensora-v1-1/inference/sample.py num_frames = 16 frame_interval = 3 fps = 24 image_size = (240, 426) multi_resolution = "STDiT2" # Define model model = dict(type="STDiT2-XL/2",from_pretrained="hpcai-tech/OpenSora-STDiT-v2-stage3",input_sq_size=512, # 使用huggingface上下载好的模型权重qk_norm=True,enable_flash_attn=True,enable_layernorm_kernel=True, ) vae = dict(type="VideoAutoencoderKL",from_pretrained="./opensora/models/sd-vae-ft-ema",cache_dir=None, # 修改为从当前目录加载micro_batch_size=4, ) text_encoder = dict(type="t5",from_pretrained="./opensora/models/text_encoder/t5-v1_1-xxl",cache_dir=None, # 修改为从当前目录加载model_max_length=200, ) scheduler = dict(type="iddpm",num_sampling_steps=100,cfg_scale=7.0,cfg_channel=3, # or None ) dtype = "fp16" # Condition prompt_path = "./assets/texts/t2v_samples.txt" prompt = None # prompt has higher priority than prompt_path # Others batch_size = 1 seed = 42 save_dir = "./samples/samples/"
Overwriting configs/opensora-v1-1/inference/sample.py
import os os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com'
!cp /home/ma-user/work/t5.py /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/opensora/models/text_encoder/t5.py
# text to video !python scripts/inference.py configs/opensora-v1-1/inference/sample.py --prompt "A fashion girl walking on the streets of Tokyo" --num-frames 32 --image-size 240 426
/home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/colossalai/shardformer/layer/normalization.py:45: UserWarning: Please install apex from source (https://github.com/NVIDIA/apex) to use the fused layernorm kernelwarnings.warn("Please install apex from source (https://github.com/NVIDIA/apex) to use the fused layernorm kernel") Config (path: configs/opensora-v1-1/inference/sample.py): {'num_frames': 32, 'frame_interval': 3, 'fps': 24, 'image_size': [240, 426], 'multi_resolution': 'STDiT2', 'model': {'type': 'STDiT2-XL/2', 'from_pretrained': 'hpcai-tech/OpenSora-STDiT-v2-stage3', 'input_sq_size': 512, 'qk_norm': True, 'enable_flash_attn': True, 'enable_layernorm_kernel': True}, 'vae': {'type': 'VideoAutoencoderKL', 'from_pretrained': './opensora/models/sd-vae-ft-ema', 'cache_dir': None, 'micro_batch_size': 4}, 'text_encoder': {'type': 't5', 'from_pretrained': './opensora/models/text_encoder/t5-v1_1-xxl', 'cache_dir': None, 'model_max_length': 200}, 'scheduler': {'type': 'iddpm', 'num_sampling_steps': 100, 'cfg_scale': 7.0, 'cfg_channel': 3}, 'dtype': 'fp16', 'prompt_path': './assets/texts/t2v_samples.txt', 'prompt': ['A fashion girl walking on the streets of Tokyo'], 'batch_size': 1, 'seed': 42, 'save_dir': './samples/samples/', 'config': 'configs/opensora-v1-1/inference/sample.py', 'prompt_as_path': False, 'reference_path': None, 'loop': 1, 'sample_name': None, 'num_sample': 1} Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]/home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/torch/_utils.py:776: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()return self.fget.__get__(instance, owner)() Loading checkpoint shards: 100%|██████████████████| 2/2 [00:35<00:00, 17.87s/it] /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.warnings.warn( 100%|█████████████████████████████████████████| 100/100 [02:11<00:00, 1.32s/it] Prompt: A fashion girl walking on the streets of Tokyo Saved to ./samples/samples/sample_0.mp4
生成的视频保存在Open-Sora/samples文件夹中，随机查看：
import os import random from moviepy.editor import * from IPython.display import Image # 视频存放目录 video_root = 'samples/samples' # 列出所有文件 videos = os.listdir(video_root) # 随机抽取视频 video = random.sample(videos, 1)[0] # 视频输入路径 video_path = os.path.join(video_root, video) # 加载原始视频 clip = VideoFileClip(video_path) # 保存为GIF文件 clip.write_gif("output_animation.gif", fps=10) # 显示生成结果 Image(open('output_animation.gif','rb').read())
MoviePy - Building file output_animation.gif with imageio.
4. Gradio 界面
修改配置文件：
%%writefile configs/opensora-v1-1/inference/sample-ref.py num_frames = 16 frame_interval = 3 fps = 24 image_size = (240, 426) multi_resolution = "STDiT2" # Condition prompt_path = None prompt = ["A car driving on the ocean.","In an ornate, historical hall, a massive tidal wave peaks and begins to crash. Two surfers, seizing the moment, skillfully navigate the face of the wave.", ] loop = 2 condition_frame_length = 4 # ( # loop id, [the loop index of the condition image or video] # reference id, [the index of the condition image or video in the reference_path] # reference start, [the start frame of the condition image or video] # target start, [the location to insert] # length, [the number of frames to insert] # edit_ratio [the edit rate of the condition image or video] # ) # See https://github.com/hpcaitech/Open-Sora/blob/main/docs/config.md#advanced-inference-config for more details # See https://github.com/hpcaitech/Open-Sora/blob/main/docs/commands.md#inference-with-open-sora-11 for more examples mask_strategy = ["0,0,0,0,8,0.3",None,"0", ] reference_path = ["https://cdn.openai.com/tmp/s/interp/d0.mp4",None,"assets/images/condition/wave.png", ] # Define model model = dict(type="STDiT2-XL/2",from_pretrained="hpcai-tech/OpenSora-STDiT-v2-stage3",input_sq_size=512, # 使用huggingface上下载好的模型权重qk_norm=True,enable_flash_attn=True,enable_layernorm_kernel=True, ) vae = dict(type="VideoAutoencoderKL",from_pretrained="./opensora/models/sd-vae-ft-ema",cache_dir=None, # 修改为从当前目录加载micro_batch_size=4, ) text_encoder = dict(type="t5",from_pretrained="./opensora/models/text_encoder/t5-v1_1-xxl",cache_dir=None, # 修改为从当前目录加载model_max_length=200, ) scheduler = dict(type="iddpm",num_sampling_steps=100,cfg_scale=7.0,cfg_channel=3, # or None ) dtype = "fp16" # Others batch_size = 1 seed = 42 save_dir = "./samples/samples/"
Overwriting configs/opensora-v1-1/inference/sample-ref.py
修改Gradio应用：
%%writefile gradio/app-ref.py import argparse import importlib import os import subprocess import sys import re import json import math import spaces import torch import gradio as gr from tempfile import NamedTemporaryFile import datetime from transformers import pipeline zh2en = pipeline("translation", model="./opus-mt-zh-en") MODEL_TYPES = ["v1.1-stage2", "v1.1-stage3"] CONFIG_MAP = {"v1.1-stage2": "configs/opensora-v1-1/inference/sample-ref.py","v1.1-stage3": "configs/opensora-v1-1/inference/sample-ref.py", } HF_STDIT_MAP = {"v1.1-stage2": "hpcai-tech/OpenSora-STDiT-v2-stage2","v1.1-stage3": "hpcai-tech/OpenSora-STDiT-v2-stage3", } RESOLUTION_MAP = {"144p": {"16:9": (256, 144), "9:16": (144, 256),"4:3": (221, 165),"3:4": (165, 221),"1:1": (192, 192),},"240p": {"16:9": (426, 240), "9:16": (240, 426),"4:3": (370, 278),"3:4": (278, 370),"1:1": (320, 320),},"360p": {"16:9": (640, 360), "9:16": (360, 640),"4:3": (554, 416),"3:4": (416, 554),"1:1": (480, 480),},"480p": {"16:9": (854, 480), "9:16": (480, 854),"4:3": (740, 555),"3:4": (555, 740),"1:1": (640, 640),},"720p": {"16:9": (1280, 720), "9:16": (720, 1280),"4:3": (1108, 832),"3:4": (832, 1110),"1:1": (960, 960),}, } # ============================ # Utils # ============================ def collect_references_batch(reference_paths, vae, image_size):from opensora.datasets.utils import read_from_path refs_x = []for reference_path in reference_paths:if reference_path is None:refs_x.append([])continueref_path = reference_path.split(";")ref = []for r_path in ref_path:r = read_from_path(r_path, image_size, transform_name="resize_crop")r_x = vae.encode(r.unsqueeze(0).to(vae.device, vae.dtype))r_x = r_x.squeeze(0)ref.append(r_x)refs_x.append(ref)# refs_x: [batch, ref_num, C, T, H, W]return refs_x def process_mask_strategy(mask_strategy):mask_batch = []mask_strategy = mask_strategy.split(";")for mask in mask_strategy:mask_group = mask.split(",")assert len(mask_group) >= 1 and len(mask_group) <= 6, f"Invalid mask strategy: {mask}"if len(mask_group) == 1:mask_group.extend(["0", "0", "0", "1", "0"])elif len(mask_group) == 2:mask_group.extend(["0", "0", "1", "0"])elif len(mask_group) == 3:mask_group.extend(["0", "1", "0"])elif len(mask_group) == 4:mask_group.extend(["1", "0"])elif len(mask_group) == 5:mask_group.append("0")mask_batch.append(mask_group)return mask_batch def apply_mask_strategy(z, refs_x, mask_strategys, loop_i):masks = []for i, mask_strategy in enumerate(mask_strategys):mask = torch.ones(z.shape[2], dtype=torch.float, device=z.device)if mask_strategy is None:masks.append(mask)continuemask_strategy = process_mask_strategy(mask_strategy)for mst in mask_strategy:loop_id, m_id, m_ref_start, m_target_start, m_length, edit_ratio = mstloop_id = int(loop_id)if loop_id != loop_i:continuem_id = int(m_id)m_ref_start = int(m_ref_start)m_length = int(m_length)m_target_start = int(m_target_start)edit_ratio = float(edit_ratio)ref = refs_x[i][m_id] # [C, T, H, W]if m_ref_start < 0:m_ref_start = ref.shape[1] + m_ref_startif m_target_start < 0:# z: [B, C, T, H, W]m_target_start = z.shape[2] + m_target_startz[i, :, m_target_start : m_target_start + m_length] = ref[:, m_ref_start : m_ref_start + m_length]mask[m_target_start : m_target_start + m_length] = edit_ratiomasks.append(mask)masks = torch.stack(masks)return masks def process_prompts(prompts, num_loop):from opensora.models.text_encoder.t5 import text_preprocessing ret_prompts = []for prompt in prompts:if prompt.startswith("|0|"):prompt_list = prompt.split("|")[1:]text_list = []for i in range(0, len(prompt_list), 2):start_loop = int(prompt_list[i])text = prompt_list[i + 1]text = text_preprocessing(text)end_loop = int(prompt_list[i + 2]) if i + 2 < len(prompt_list) else num_looptext_list.extend([text] * (end_loop - start_loop))assert len(text_list) == num_loop, f"Prompt loop mismatch: {len(text_list)} != {num_loop}"ret_prompts.append(text_list)else:prompt = text_preprocessing(prompt)ret_prompts.append([prompt] * num_loop)return ret_prompts def extract_json_from_prompts(prompts):additional_infos = []ret_prompts = []for prompt in prompts:parts = re.split(r"(?=[{\[])", prompt)assert len(parts) <= 2, f"Invalid prompt: {prompt}"ret_prompts.append(parts[0])if len(parts) == 1:additional_infos.append({})else:additional_infos.append(json.loads(parts[1]))return ret_prompts, additional_infos # ============================ # Model-related # ============================ def read_config(config_path):""" Read the configuration file.""" from mmengine.config import Config return Config.fromfile(config_path) def build_models(model_type, config, enable_optimization=False):""" Build the models for the given model type and configuration.""" # build vaefrom opensora.registry import MODELS, build_module vae = build_module(config.vae, MODELS).cuda() # build text encodertext_encoder = build_module(config.text_encoder, MODELS) # T5 must be fp32text_encoder.t5.model = text_encoder.t5.model.cuda() # build stdit# we load model from HuggingFace directly so that we don't need to# handle model download logic in HuggingFace Spacefrom opensora.models.stdit.stdit2 import STDiT2 stdit = STDiT2.from_pretrained(HF_STDIT_MAP[model_type],enable_flash_attn=enable_optimization,trust_remote_code=True,).cuda() # build schedulerfrom opensora.registry import SCHEDULERS scheduler = build_module(config.scheduler, SCHEDULERS) # hack for classifier-free guidancetext_encoder.y_embedder = stdit.y_embedder # move modelst to devicevae = vae.to(torch.float16).eval()text_encoder.t5.model = text_encoder.t5.model.eval() # t5 must be in fp32stdit = stdit.to(torch.float16).eval() # clear cudatorch.cuda.empty_cache()return vae, text_encoder, stdit, scheduler def parse_args():parser = argparse.ArgumentParser()parser.add_argument("--model-type",default="v1.1-stage3",choices=MODEL_TYPES,help=f"The type of model to run for the Gradio App, can only be {MODEL_TYPES}",)parser.add_argument("--output", default="./outputs", type=str, help="The path to the output folder")parser.add_argument("--port", default=None, type=int, help="The port to run the Gradio App on.")parser.add_argument("--host", default=None, type=str, help="The host to run the Gradio App on.")parser.add_argument("--share", action="store_true", help="Whether to share this gradio demo.")parser.add_argument("--enable-optimization",action="store_true",help="Whether to enable optimization such as flash attention and fused layernorm",)return parser.parse_args() # ============================ # Main Gradio Script # ============================ # as `run_inference` needs to be wrapped by `spaces.GPU` and the input can only be the prompt text # so we can't pass the models to `run_inference` as arguments. # instead, we need to define them globally so that we can access these models inside `run_inference` # read config args = parse_args() config = read_config(CONFIG_MAP[args.model_type]) # make outputs dir os.makedirs(args.output, exist_ok=True) # disable torch jit as it can cause failure in gradio SDK # gradio sdk uses torch with cuda 11.3 torch.jit._state.disable() # import after installation from opensora.datasets import IMG_FPS, save_sample from opensora.utils.misc import to_torch_dtype # some global variables dtype = to_torch_dtype(config.dtype) device = torch.device("cuda") # build model vae, text_encoder, stdit, scheduler = build_models(args.model_type, config, enable_optimization=args.enable_optimization) def run_inference(mode, prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale):torch.manual_seed(seed)with torch.inference_mode():# ======================# 1. Preparation# ======================# parse the inputsresolution = RESOLUTION_MAP[resolution][aspect_ratio] # gather args from confignum_frames = config.num_framesframe_interval = config.frame_intervalfps = config.fpscondition_frame_length = config.condition_frame_length # compute number of loopsif mode == "Text2Image":num_frames = 1num_loop = 1else:num_seconds = int(length.rstrip('s'))if num_seconds <= 16:num_frames = num_seconds * fps // frame_intervalnum_loop = 1else:config.num_frames = 16total_number_of_frames = num_seconds * fps / frame_intervalnum_loop = math.ceil((total_number_of_frames - condition_frame_length) / (num_frames - condition_frame_length)) # prepare model argsif config.num_frames == 1:fps = IMG_FPS model_args = dict()height_tensor = torch.tensor([resolution[0]], device=device, dtype=dtype)width_tensor = torch.tensor([resolution[1]], device=device, dtype=dtype)num_frames_tensor = torch.tensor([num_frames], device=device, dtype=dtype)ar_tensor = torch.tensor([resolution[0] / resolution[1]], device=device, dtype=dtype)fps_tensor = torch.tensor([fps], device=device, dtype=dtype)model_args["height"] = height_tensormodel_args["width"] = width_tensormodel_args["num_frames"] = num_frames_tensormodel_args["ar"] = ar_tensormodel_args["fps"] = fps_tensor # compute latent sizeinput_size = (num_frames, *resolution)latent_size = vae.get_latent_size(input_size) # process promptprompt = zh2en(prompt_text)[0].get("translation_text")prompt_raw = [prompt]print(prompt_raw)prompt_raw, _ = extract_json_from_prompts(prompt_raw)prompt_loops = process_prompts(prompt_raw, num_loop)video_clips = [] # prepare mask strategyif mode == "Text2Image":mask_strategy = [None]elif mode == "Text2Video":if reference_image is not None:mask_strategy = ['0']else:mask_strategy = [None]else:raise ValueError(f"Invalid mode: {mode}") # =========================# 2. Load reference images# =========================if mode == "Text2Image":refs_x = collect_references_batch([None], vae, resolution)elif mode == "Text2Video":if reference_image is not None:# save image to diskfrom PIL import Imageim = Image.fromarray(reference_image) with NamedTemporaryFile(suffix=".jpg") as temp_file:im.save(temp_file.name)refs_x = collect_references_batch([temp_file.name], vae, resolution)else:refs_x = collect_references_batch([None], vae, resolution)else:raise ValueError(f"Invalid mode: {mode}") # 4.3. long video generationfor loop_i in range(num_loop):# 4.4 sample in hidden spacebatch_prompts = [prompt[loop_i] for prompt in prompt_loops]z = torch.randn(len(batch_prompts), vae.out_channels, *latent_size, device=device, dtype=dtype) # 4.5. apply mask strategymasks = None # if cfg.reference_path is not None:if loop_i > 0:ref_x = vae.encode(video_clips[-1])for j, refs in enumerate(refs_x):if refs is None:refs_x[j] = [ref_x[j]]else:refs.append(ref_x[j])if mask_strategy[j] is None:mask_strategy[j] = ""else:mask_strategy[j] += ";"mask_strategy[j] += f"{loop_i},{len(refs)-1},-{condition_frame_length},0,{condition_frame_length}" masks = apply_mask_strategy(z, refs_x, mask_strategy, loop_i) # 4.6. diffusion sampling# hack to update num_sampling_steps and cfg_scalescheduler_kwargs = config.scheduler.copy()scheduler_kwargs.pop('type')scheduler_kwargs['num_sampling_steps'] = sampling_stepsscheduler_kwargs['cfg_scale'] = cfg_scale scheduler.__init__(**scheduler_kwargs)samples = scheduler.sample(stdit,text_encoder,z=z,prompts=batch_prompts,device=device,additional_args=model_args,mask=masks, # scheduler must support mask)samples = vae.decode(samples.to(dtype))video_clips.append(samples) # 4.7. save videoif loop_i == num_loop - 1:video_clips_list = [video_clips[0][0]] + [video_clips[i][0][:, config.condition_frame_length :] for i in range(1, num_loop)]video = torch.cat(video_clips_list, dim=1)current_datetime = datetime.datetime.now()timestamp = current_datetime.timestamp()save_path = os.path.join(args.output, f"output_{timestamp}")saved_path = save_sample(video, save_path=save_path, fps=config.fps // config.frame_interval)return saved_path @spaces.GPU(duration=200) def run_image_inference(prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale):return run_inference("Text2Image", prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale) @spaces.GPU(duration=200) def run_video_inference(prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale):return run_inference("Text2Video", prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale) def main():# create demowith gr.Blocks() as demo:with gr.Row():with gr.Column():gr.HTML("""<h1 align="center">Open-Sora 1.1</h1>""") with gr.Row():with gr.Column():prompt_text = gr.Textbox(label="Prompt",placeholder="请输入中文提示词",lines=4,)resolution = gr.Radio(choices=["144p", "240p", "360p", "480p", "720p"],value="240p",label="Resolution", )aspect_ratio = gr.Radio(choices=["9:16", "16:9", "3:4", "4:3", "1:1"],value="9:16",label="Aspect Ratio (H:W)", )length = gr.Radio(choices=["2s", "4s", "8s", "16s"], value="2s",label="Video Length (only effective for video generation)", info="8s may fail as Hugging Face ZeroGPU has the limitation of max 200 seconds inference time.") with gr.Row():seed = gr.Slider(value=1024,minimum=1,maximum=2048,step=1,label="Seed") sampling_steps = gr.Slider(value=100,minimum=1,maximum=200,step=1,label="Sampling steps")cfg_scale = gr.Slider(value=7.0,minimum=0.0,maximum=10.0,step=0.1,label="CFG Scale")reference_image = gr.Image(label="Reference Image (Optional)",)with gr.Column():output_video = gr.Video(label="Output Video",height="100%") with gr.Row():image_gen_button = gr.Button("Generate image")video_gen_button = gr.Button("Generate video")image_gen_button.click(fn=run_image_inference, inputs=[prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale], outputs=reference_image)video_gen_button.click(fn=run_video_inference, inputs=[prompt_text, resolution, aspect_ratio, length, reference_image, seed, sampling_steps, cfg_scale], outputs=output_video) # launchdemo.launch(share=True, inbrowser=True) if __name__ == "__main__":main()
Writing gradio/app-ref.py
运行Gradio应用，运行成功后点击
Running on public URL
后的网页链接即可体验！
!python gradio/app-ref.py
/home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/torch/_utils.py:776: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()return self.fget.__get__(instance, owner)() /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/transformers/models/marian/tokenization_marian.py:197: UserWarning: Recommended: pip install sacremoses.warnings.warn("Recommended: pip install sacremoses.") /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/colossalai/shardformer/layer/normalization.py:45: UserWarning: Please install apex from source (https://github.com/NVIDIA/apex) to use the fused layernorm kernelwarnings.warn("Please install apex from source (https://github.com/NVIDIA/apex) to use the fused layernorm kernel") Loading checkpoint shards: 100%|██████████████████| 2/2 [00:32<00:00, 16.15s/it] /home/ma-user/anaconda3/envs/python-3.10.10/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.warnings.warn( Running on local URL: http://127.0.0.1:7860 Running on public URL: https://64147712240bbb3753.gradio.live This share link expires in 72 hours. For free permanent hosting and GPU upgrades, run `gradio deploy` from Terminal to deploy to Spaces (https://huggingface.co/spaces)
我们也准备了一些提示词以供参考：
一只穿着紫色长袍的胖兔子穿过奇幻的风景
海浪冲击着孤零零的灯塔，不祥的灯光
一个神秘的森林展示了旅行者的冒险经历
一个蓝头发的法师在唱歌
一个超现实的景观，漂浮的岛屿和空中的瀑布
一只蓝鸟站在水里
一个年轻人独自走在海边
粉红色的玫瑰在玻璃表面滴，特写
驱车远眺，一列地铁正从隧道中驶出
太空中所有的行星都是绿色和粉色的，背景是明亮的白色恒星
一座漂浮在星体空间的城市，有星星和星云 高楼顶上的日出
粉色和青色粉末爆炸 树林里的鹿在阳光下凝视着相机
一道闪电，一个巫师从稀薄的空气中出现了，他的长袍在风中翻腾
夜晚的未来赛博朋克城市景观，高耸的霓虹灯照亮的摩天大楼
在这里，树木、花朵和动物聚集在一起，谱写出一曲大自然的交响乐
一艘幽灵般的船在云层中航行，在月光下的天空中航行 日落和美丽的海滩
一个年轻人独自走在森林里
生成好的视频也可以使用MusicGen进行配乐，使用AI进行短视频创作。
5. 视频效果展示
提示词：一个极端的特写一个头发花白的胡子的男人在他的60年代,他在思想深处思考宇宙的历史,他坐在一家咖啡馆在巴黎,他的眼睛关注人私生活方面大多像他坐在他们走不动,他穿着一件羊毛外套西装外套和一件衬衫,他穿着一件棕色的贝雷帽,眼镜和有一个非常专业的外表,和结束他提供了一个微妙的封闭式的笑容好像找到了答案的神秘生活,灯光非常电影化，金色的灯光和巴黎的街道和城市作为背景，景深，电影化的35mm胶片。
提示词：无人机拍摄的海浪冲击着大苏尔加雷角海滩上崎岖的悬崖。蓝色的海水拍打着白色的波浪，夕阳的金色光芒照亮了岩石海岸。远处有一座小岛，岛上有一座灯塔，悬崖边上长满了绿色的灌木丛。从公路到海滩的陡峭落差是一个戏剧性的壮举，悬崖的边缘突出在海面上。这是一幅捕捉到海岸原始美景和太平洋海岸公路崎岖景观的景色。
提示词：一段高耸的无人机镜头捕捉到了海岸悬崖的雄伟之美，它的红色和黄色分层岩石表面色彩丰富，映衬着充满活力的绿松石般的大海。可以看到海鸟在悬崖峭壁上飞翔。当无人机从不同角度缓慢移动时，变化的阳光投射出移动的阴影，突出了悬崖的崎岖纹理和周围平静的大海。水轻轻地拍打着岩石基座和附着在悬崖顶部的绿色植物，这一场景给人一种宁静的感觉，在海洋的边缘孤立。这段视频捕捉了未受人类建筑影响的原始自然美的本质。
提示词：雄伟美丽的瀑布从悬崖上倾泻而下，进入宁静的湖泊。瀑布，以其强大的流量，是视频的中心焦点。周围的景色郁郁葱葱，树木和树叶增添了自然美景。相机角度提供了瀑布的鸟瞰图，让观众欣赏瀑布的全部高度和壮观。这段视频令人惊叹地展现了大自然的力量和美。
提示词：夜晚熙熙攘攘的城市街道，充满了汽车前灯的光辉和街灯的环境光。场景是一个模糊的运动，汽车飞驰而过，行人在人行横道上穿行。城市景观是高耸的建筑和照明标志的混合，创造了一个充满活力和动态的氛围。视频的视角是高角度的，提供了街道及其周围环境的鸟瞰图。整个视频的风格充满活力和活力，捕捉到了夜晚城市生活的精髓。
提示词：森林地区宁静的夜景。第一个画面是一个宁静的湖泊，倒映着繁星满天的夜空。第二帧展示了美丽的日落，在风景上投下温暖的光芒。第三帧展示了夜空，充满了星星和充满活力的银河系。这段视频是延时拍摄的，捕捉了从白天到夜晚的过渡，湖泊和森林作为恒定的背景。视频的风格是自然主义的，强调夜空的美丽和森林的宁静。
点击关注，第一时间了解华为云新鲜技术~
![疯狂的麦克斯:狂暴女神迅雷BT下载[百度云AVI/1.26G]高清版[HD720p国语中字](https://img2024.cnblogs.com/blog/3068268/202406/3068268-20240606104032147-227073077.jpg)
