Comments (4)
Wow, thank you so much for your outstanding work! I genuinely believe it is incredibly helpful 👍👍👍.
from codef.
%cd /content
%env TF_CPP_MIN_LOG_LEVEL=1
# !git clone --recursive https://github.com/nvlabs/tiny-cuda-nn
# %cd /content/tiny-cuda-nn
# !cmake . -B build
# !cmake --build build --config RelWithDebInfo -j
# !pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch
!git clone -b dev https://github.com/camenduru/CoDeF
%cd /content/CoDeF
!pip install -q pytorch_lightning einops ipdb torch_optimizer kornia scikit-video
!pip install -q https://huggingface.co/camenduru/CoDeF/resolve/main/tinycudann-1.7-cp310-cp310-linux_x86_64.whl
!apt -y install -qq aria2
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/all_sequences.zip -d /content/CoDeF/ckpts -o all_sequences.zip
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/beauty_0.ckpt -d /content/CoDeF/ckpts -o beauty_0.ckpt
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/beauty_1.ckpt -d /content/CoDeF/ckpts -o beauty_1.ckpt
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/lemon_hit.ckpt -d /content/CoDeF/ckpts -o lemon_hit.ckpt
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/scene_0.ckpt -d /content/CoDeF/ckpts -o scene_0.ckpt
!aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/camenduru/CoDeF/resolve/main/white_smoke.ckpt -d /content/CoDeF/ckpts -o white_smoke.ckpt
!unzip /content/CoDeF/ckpts/all_sequences.zip -d /content/CoDeF/ckpts
!rm -rf /content/CoDeF/ckpts/__MACOSX
# !mkdir /content/CoDeF/ckpts/all_sequences/scene_0/base
# !cp /content/CoDeF/ckpts/scene_0.ckpt /content/CoDeF/ckpts/all_sequences/scene_0/base/scene_0.ckpt
# @title Generate Video (Step 3)
GPUS=0
NAME="scene_0"
EXP_NAME="base"
ROOT_DIRECTORY=f"ckpts/all_sequences/{NAME}/{NAME}"
LOG_SAVE_PATH=f"logs/test_all_sequences/{NAME}"
MASK_DIRECTORY=f"ckpts/all_sequences/{NAME}/{NAME}_masks_0 ckpts/all_sequences/{NAME}/{NAME}_masks_1"
CONFIG_DIRECTORY=f"configs/{NAME}/{EXP_NAME}.yaml"
CANONICAL_DIR=f"ckpts/all_sequences/{NAME}/{EXP_NAME}_control"
WEIGHT_PATH= f"ckpts/{NAME}.ckpt"
!python train.py --test --encode_w \
--root_dir $ROOT_DIRECTORY \
--log_save_path $LOG_SAVE_PATH \
--mask_dir $MASK_DIRECTORY \
--weight_path $WEIGHT_PATH \
--gpus $GPUS \
--canonical_dir $CANONICAL_DIR \
--config $CONFIG_DIRECTORY \
--exp_name $EXP_NAME
from codef.
i am facing issue with colab
Traceback (most recent call last):
File "/content/CoDeF/train.py", line 556, in
hparams = get_opts()
File "/content/CoDeF/opt.py", line 157, in get_opts
with open(args.config, 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: 'configs/serge/base.yaml'
from codef.
replace the block 1 with the following code and it should work. Issue is with variable names and relative paths.
# @title Train (Step 1)
GPUS=0
NAME="beauty_0"
EXP_NAME="base"
ROOT_DIRECTORY=f"ckpts/all_sequences/{NAME}/{NAME}"
MODEL_SAVE_PATH=f"ckpts/all_sequences/{NAME}"
LOG_SAVE_PATH=f"logs/test_all_sequences/{NAME}"
WEIGHT_PATH=f"ckpts/all_sequences/{NAME}.ckpt"
CONFIG_DIRECTORY=f"configs/{NAME}/{EXP_NAME}.yaml"
MASK_DIRECTORY=f"all_sequences/{NAME}/{NAME}_masks_0 all_sequences/{NAME}/{NAME}_masks_1"
FLOW_DIRECTORY=f"all_sequences/{NAME}/{NAME}_flow"
!python train.py --root_dir $ROOT_DIRECTORY \
--model_save_path $MODEL_SAVE_PATH \
--log_save_path $LOG_SAVE_PATH \
--gpus $GPUS \
--encode_w --annealed \
--config $CONFIG_DIRECTORY \
--exp_name $EXP_NAME
# --mask_dir $MASK_DIRECTORY
# --flow_dir $FLOW_DIRECTORY
https://github.com/desaiankitb/video-capabilities/blob/main/Copy_of_CoDeF_colab.ipynb - this is the working copy with "beauty_0" example.
from codef.
Related Issues (20)
- 错误似乎是在尝试使用skvideo库的ffmpeg模块写入视频帧时发生的。具体来说,您遇到了一个BrokenPipeError,这通常意味着子进程(在这种情况下是ffmpeg)在父进程尝试写入数据时已经终止或崩溃 HOT 1
- 重建视频很清楚,但是canonical图片很模糊 HOT 1
- No such file or directory: 'configs/StrangerThings/base.yaml'
- CUDA error: out of memory HOT 1
- 重建的transfomed视频是黑色的
- Question about the flow HOT 1
- How can I reduce the blur in my canonical image?
- Regarding pe_deformed_grid = (deformed_grid + 0.3) / 1.6 HOT 1
- 【BUG ?】 Problems about the background loss when `self_bg=False`
- a question about train code? HOT 1
- Some confusions about the code HOT 2
- why can't i run this application on windows10 OS? HOT 1
- 请问一个不同视频就得训练一个模型吗 HOT 2
- problem of reconstruction by canonical image HOT 1
- i met some errors when run train step 1 in colab
- video translation 测试这一步具体是怎么实现的,该对哪个图片使用controlnet。那这个视频风格迁移的效果是取决于controlnet生图的效果吗
- Codef
- 在哪里指定prompt?
- 如何改变transform后的视频的风格呢?
- 视频中人物动作变化太快,训练出来的 canonical image 非常的模糊
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from codef.