Package conflict

[anilakash]

Hi,
Thanks for sharing the code for A*NET. While installing packages like torchdrug it is resulting in different types of errors. There is a conflict for the packages. Could you please specify the version for each of these packages.

Thanks

[KiddoZhu]

Hi. Thanks for interest in this project. The version specifications can be found in requirements.txt.

[anilakash]

Hi,
Thanks for your response. The conflicts were resolved when I used the latest PyTorch. However, it leads to a Torchdrug error for missing header file namely, SparseTensorUtils.h under ATen. In recent PyTorch this file has been moved to ATen/native. I have requested a pull after committing these changes. Please approve if that works.

[JuliaGast]

Hello @anilakash
I have the same problem, the torchdrug error for missing header file.

Could you please let me know your solution?
Best and Thanks

[anilakash]

Hi @JuliaGast
I have forked the official torchdrug implementation and updated the paths to ATen/native. The committed repository is available at https://github.com/anilakash/torchdrug

[Cranooooooo]

Hi @anilakash
I have problems when setting up the environment.
I guess it is because of the version mismatch between those packages.

May I get the version list of your packages?
Thanks a lot!

[anilakash]

Hi @anilakash I have problems when setting up the environment. I guess it is because of the version mismatch between those packages.

May I get the version list of your packages? Thanks a lot!

Hi @Cranooooooo,

Thanks for the query. Could you please confirm the specific package. On the other hand, you can try to install other packages as suggested in the original repository and install torchdrug from here https://github.com/anilakash/torchdrug . This works for me. Good luck.

[EauDeData]

Hi @anilakash , words don't allow me to thank you enough for your help with your fork.
For whomever struggling with something similar (in CUDA12.4, Pytorch 2.40) find here a short bash script that worked for me:

conda create -n astar python==3.10 -y
conda activate astar; conda clean -a -y; pip cache purge
pip install numpy==1.26.4
pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 --index-url https://download.pytorch.org/whl/cu124
pip install torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
pip install torch-cluster -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
pip install pyg-lib -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
echo "Now go and thank this brother git@github.com:anilakash/torchdrug.git"
git clone git@github.com:anilakash/torchdrug.git
cd torchdrug
pip install -r requirements.txt
python setup.py install
cd ..; rm -rf ./torchdrug

pip install ogb
pip install easydict
pip install PyYAML
pip install easydict

[anilakash]

Hi Adri, Many thanks for kind words. Happy to contribute.

…

On Mon, Nov 4, 2024 at 2:18 PM Adri ***@***.***> wrote: Hi @anilakash <https://github.com/anilakash> , words don't allow me to thank you enough for your help with your fork. For whomever struggling with something similar (in CUDA12.4, Pytorch 2.40) find here a short bash script that worked for me: conda create -n astar python==3.10 -y conda activate astar; conda clean -a -y; pip cache purge pip install numpy==1.26.4 pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 --index-url https://download.pytorch.org/whl/cu124 pip install torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.4.0+cu124.html pip install torch-cluster -f https://data.pyg.org/whl/torch-2.4.0+cu124.html pip install pyg-lib -f https://data.pyg.org/whl/torch-2.4.0+cu124.html echo "Now go and thank this brother ***@***.***:anilakash/torchdrug.git" git clone ***@***.***:anilakash/torchdrug.git cd torchdrug pip install -r requirements.txt python setup.py install cd ..; rm -rf ./torchdrug pip install ogb pip install easydict pip install PyYAML pip install easydict — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC7HDFWAQSP7NUJNMMEZQL3Z6564ZAVCNFSM6AAAAABRELR4TSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINJUHAZTQNRXGE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Thanks & Regards, Akash Anil, Ph.D. Research Associate, Computer Science and Informatics Cardiff University, Wales, UK <https://sites.google.com/site/akashanil2601/>

----------------------------------------------------------------------------------------------------------------------------------------------- “You don’t have to be great to start, but you have to start to be great.” – Zig Ziglar

---------------------------------------------------------------------------------------------------------------------------------- --------------

[rainbowww24]

Hi @anilakash , words don't allow me to thank you enough for your help with your fork. For whomever struggling with something similar (in CUDA12.4, Pytorch 2.40) find here a short bash script that worked for me:

conda create -n astar python==3.10 -y
conda activate astar; conda clean -a -y; pip cache purge
pip install numpy==1.26.4
pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 --index-url https://download.pytorch.org/whl/cu124
pip install torch-scatter torch-sparse -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
pip install torch-cluster -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
pip install pyg-lib -f https://data.pyg.org/whl/torch-2.4.0+cu124.html
echo "Now go and thank this brother git@github.com:anilakash/torchdrug.git"
git clone git@github.com:anilakash/torchdrug.git
cd torchdrug
pip install -r requirements.txt
python setup.py install
cd ..; rm -rf ./torchdrug

pip install ogb
pip install easydict
pip install PyYAML
pip install easydict

Hi @EauDeData , I really can’t thank you enough. I’m a first-year graduate student, and I’ve been running the code repeatedly for two weeks — it kept saying the C++ files weren’t compiled or some ATen-related files were missing. After seeing your method, I’ve decided to give it a try. I truly believe it’ll work. Thank you so much!