Relationship prior

[AmosDinh]

Hello,

would you be so kind as to explain what the relationship prior is? I looked into the code, but I do not understand that part.

Thank you

[Saladin58]

Hello,

Could you run the model successfully? do you mind to share the python version and the packages that you've used, please?

Thanks, Saladin

[AmosDinh]

Hello, no I haven't run the model yet. You can use the implementation of the python package Pytorch Geometric though

Saladin58 @.***> schrieb am Fr., 13. Okt. 2023, 12:09:

Hello,

Could you run the model successfully? do you mind to share the python version and the packages that you've used, please?

Thanks, Saladin

— Reply to this email directly, view it on GitHub https://github.com/UCLA-DM/pyHGT/issues/62#issuecomment-1762055944, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJQ5FRAUZB4GAIYPFH2QXQDX7GGW7AVCNFSM6AAAAAA57QZN52VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRSGA2TKOJUGQ . You are receiving this because you authored the thread.Message ID: @.***>

[Saladin58]

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

[RolandTate]

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

I just run the code successfully, what problem u meet. Maybe i can help u

[Saladin58]

Thanks for your email. I always get this error:

Traceback (most recent call last): File "/home/saakdr/OAG/train_paper_field.py", line 72, in graph = renamed_load(open(os.path.join(args.data_dir, 'graph%s.pk' % args.domain), 'rb')) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/saakdr/OAG/pyHGT/data.py", line 272, in renamed_load return RenameUnpickler(file_obj).load() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/storage/hpc/data/saakdr/miniconda/envs/HGT/lib/python3.11/site-packages/dill/_dill.py", line 442, in load obj = StockUnpickler.load(self)

Thanks, Best regards. Salhuldin Alqarghuli (Saladin)

---

From: RolandTate @.> Sent: Thursday, October 26, 2023 9:22 PM To: UCLA-DM/pyHGT @.> Cc: Alqarghuli, Salhuldin (MU-Student) @.>; Comment @.> Subject: Re: [UCLA-DM/pyHGT] Relationship prior (Issue #62)

WARNING: This message has originated from an External Source. This may be a phishing expedition that can result in unauthorized access to our IT System. Please use proper judgment and caution when opening attachments, clicking links, or responding to this email.

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

I just run the code successfully, what problem u meet. Maybe i can help u

— Reply to this email directly, view it on GitHubhttps://github.com/UCLA-DM/pyHGT/issues/62#issuecomment-1782185807, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A5NSBL442JAHJWSD4HMRLV3YBMLE7AVCNFSM6AAAAAA57QZN52VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBSGE4DKOBQG4. You are receiving this because you commented.Message ID: @.***>

[Saladin58]

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

I just run the code successfully, what problem u meet. Maybe i can help u

I finally solved the previous error by downgrading the dill package. Now I am getting the follwing error Have you ever had it before, please?

python train_paper_field.py --data_dir /home/saakdr/OAG/dataset --model_dir ./ --conv_name hgt --sample_depth 1 --sample_width 1 --n_pool 1 --repeat 1

multiprocessing.pool.RemoteTraceback: """ Traceback (most recent call last): File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 121, in worker result = (True, func(*args, **kwds)) File "train_paper_field.py", line 128, in node_classification_sample to_torch(feature, times, edge_list, graph) File "/home/saakdr/OAG/pyHGT/data.py", line 232, in to_torch node_num += len(feature[t]) KeyError: 'affiliation' """

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 657, in get raise self._value KeyError: 'affiliation'

Thanks, Saladin

[RolandTate]

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

I just run the code successfully, what problem u meet. Maybe i can help u

I finally solved the previous error by downgrading the dill package. Now I am getting the follwing error Have you ever had it before, please?

python train_paper_field.py --data_dir /home/saakdr/OAG/dataset --model_dir ./ --conv_name hgt --sample_depth 1 --sample_width 1 --n_pool 1 --repeat 1

multiprocessing.pool.RemoteTraceback: """ Traceback (most recent call last): File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 121, in worker result = (True, func(*args, **kwds)) File "train_paper_field.py", line 128, in node_classification_sample to_torch(feature, times, edge_list, graph) File "/home/saakdr/OAG/pyHGT/data.py", line 232, in to_torch node_num += len(feature[t]) KeyError: 'affiliation' """

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 657, in get raise self._value KeyError: 'affiliation'

Thanks, Saladin

Sorry, I reply u late. I'm glad to see u sovle the first problem. I run the code based on Windows, and I met a problem about the 'train_data = [job.get() for job in jobs[:-1]]', my error is not like urs. Here I share my expericence with u. I changed the code: 1) in 'preprocess_OAG.py', when open files, add one more parameter about encoding, like 'line_count = sum(1 for line in open(filename, encoding='utf-8'))'. 2)in 'utils.py' some parameter about dype is too old, like 'np.srt'. So u need to check this. Above is how i change the code to make the code successfully run based Windows. I hope I can help u. Good luck.

[Saladin58]

I have used these packages as recommended but still get errors. Python==3.7.9 dill==0.3.0 numpy==1.22.0 pandas==0.24.2 torch==1.13.1 torch-scatter==1.3.2 torch-cluster==1.4.5 torch-sparse==0.4.3 torch-spline-conv==1.1.1 torch-geometric==1.3.2 torchvision==0.4.1 tqdm==4.31.1 seaborn==0.9.0 matplotlib==3.0.3 transformers==4.30.0

I just run the code successfully, what problem u meet. Maybe i can help u

I finally solved the previous error by downgrading the dill package. Now I am getting the follwing error Have you ever had it before, please? python train_paper_field.py --data_dir /home/saakdr/OAG/dataset --model_dir ./ --conv_name hgt --sample_depth 1 --sample_width 1 --n_pool 1 --repeat 1 multiprocessing.pool.RemoteTraceback: """ Traceback (most recent call last): File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 121, in worker result = (True, func(*args, **kwds)) File "train_paper_field.py", line 128, in node_classification_sample to_torch(feature, times, edge_list, graph) File "/home/saakdr/OAG/pyHGT/data.py", line 232, in to_torch node_num += len(feature[t]) KeyError: 'affiliation' """ The above exception was the direct cause of the following exception: Traceback (most recent call last): File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "train_paper_field.py", line 224, in train_data = [job.get() for job in jobs[:-1]] File "/home/saakdr/.conda/envs/HGT/lib/python3.7/multiprocessing/pool.py", line 657, in get raise self._value KeyError: 'affiliation' Thanks, Saladin

Sorry, I reply u late. I'm glad to see u sovle the first problem. I run the code based on Windows, and I met a problem about the 'train_data = [job.get() for job in jobs[:-1]]', my error is not like urs. Here I share my expericence with u. I changed the code: 1) in 'preprocess_OAG.py', when open files, add one more parameter about encoding, like 'line_count = sum(1 for line in open(filename, encoding='utf-8'))'. 2)in 'utils.py' some parameter about dype is too old, like 'np.srt'. So u need to check this. Above is how i change the code to make the code successfully run based Windows. I hope I can help u. Good luck.

Thank you so much, I really appreciate that!

[AmosDinh]

For future readers which are interested in the original question: It was explained to me that the prior is a learnable weight.