Stack Overflow
Stack Overflow

Questions tagged [dgl]

50 questions
1
vote
1 answer

from dgl.data.chem import mol_to_bigraph, smiles_to_bigraph does not work

I am working on graph classification problem based on molecules. My data is SMILES, and I want to convert it into graph and build model using DGL package. I have installed the dgl package by using pip install dgl in spyder and ! pip install dgl in…
helen2020
  • 13
  • 2
0
votes
0 answers

How to fix the error ecross training "PROTEINS" dataset from DGL by network involved SuperGAT layers?

I want to train "PROTEINS" dataset from the DGL library. In order to do it, I chose a network that involved two SupreGAT layers. When The training process was implemented, I faced the following error. It is my pleasure to help me fix this error. #…
khatooon khedri
  • 1
0
votes
0 answers

Memory requirement of GNN SageConv

Sorry for my simple question. My question is that in GraphSage paper, they talked about graphsage being inductive. Is Graphsage needs all of its training time samples in memory during testing? If not, how does it performs test? Since in its…
user85361
  • 177
  • 1
  • 9
0
votes
0 answers

(Very Intriguing Trial) GraphSAGE: Feature Request | Feature Instruction: Exploring the Feasibility of Implementing Asymmetric Node Feature Slicing

Currently, I am working on implementing GraphSage with DGL for an industrial node classification application. Here is a brief description of my scenario: In my application, each node in the graph has around 21 features associated with it. However,…
billz
  • 11
  • 1
0
votes
0 answers

When Train on RGCN, loss did not change at all

I meet a problem! When I train my heterogenous graph, I found that r-gcn network’s accuracy and loss stay constant. I try to print the gradient, however, the gradient at the beginning is 0 matric, and the gradient during the training process still…
Stupidog
  • 1
0
votes
0 answers

Why am I not getting representations for both product and customer nodes (heterogenous graph)?

I have the following graph: cust_prod_graph is Graph(num_nodes={‘customer’: 8813, ‘product’: 157466}, num_edges={(‘customer’, ‘browsed’, ‘product’): 860771, (‘customer’, ‘purchased’, ‘product’): 68367}, metagraph=[(‘customer’, ‘product’, ‘browsed’),…
Vishal Chakravarthi
  • 3
  • 2
0
votes
0 answers

I am playing around with DGL. In this specific case, I have a heterogenous graph, but the output of the RGCN is not what I expect

This is what my graph looks like: cust_prod_graph = Graph(num_nodes={'customer': 8813, 'product': 157466}, num_edges={('customer', 'browsed', 'product'): 860771, ('customer', 'purchased', 'product'): 68367}, metagraph=[('customer',…
Vishal Chakravarthi
  • 3
  • 2
0
votes
0 answers

How to solve the OOM problem encountered when using dgl to process ultra large image data？

I am trying to generate mini batch data using a sampler with built-in support in DataLoader to support random batch training on large images, dgl.dataloading.DataLoader( graph, valid_nids, sampler, batch_size=1024, shuffle=False, drop_last=False,…
user10769790
  • 1
0
votes
0 answers

A question about mxnet - MXNetError: Check failed: delay_alloc:

I am studying GNN and trying to follow and study some code samples. What i am looking at is Detecting fraud in heterogeneous networks using Amazon SageMaker and Deep Graph Library | AWS Machine Learning Blog. I am trying to implement this by plain…
kcwoo
  • 1
  • 1
0
votes
1 answer

GATconv dimensions

When I use this class to build a GATConv network to predict the class of a graph I get a dimension error. How would I resolve this. class GraphClassifier(nn.Module): def __init__(self, in_feats, hidden_size, num_classes): …
Branco François
  • 51
  • 2
  • 6
0
votes
0 answers

Out-of-Memory when batch training GNN on Very large graph

I use the dgl library to batch train the graph neural network model on a very large graph, and the sampling method is MultiLayerFullNeighborSampler(2). But even if the batch size is set to 1, a certain subgraph may have tens of millions of edges due…
Xiaoyi
  • 1
0
votes
1 answer

ERROR during trainning model with dgl: recordStream only works when TensorAdaptor is available

I encountered strange error while training my MLP model. I have no clue what to change nor how to fix it. First I've ran it in conda env with follow packages: - cudatoolkit=11.3 - cudnn=7.6.5 - python=3.7.4 - python-dateutil=2.8.0 - pip=19.2.3 -…
szauniema
  • 1
0
votes
0 answers

Python - AttributeError: 'numpy.ndarray' object has no attribute 'device'

everyone. I am beginner in coding and non-major in it. I run the code by python on Window WSL(Ubuntu). In running the code, dgl was run and it has error like 'AttributeError: 'numpy.ndarray' object has no attribute 'device'' More…
beanSsong
  • 1
  • 1
0
votes
0 answers

Creating dataset with labels using networkx and dgl

I’m quite new to dgl, therefore I have a question. Imagine, having a graphs with weights implemented in networkx and also the corresponding labels for them (let’s say stored in a list). import networkx as nx import dgl import torch labels =…
Keithx
  • 2,994
  • 15
  • 42
  • 71
0
votes
0 answers

Error when following dgl library tutorial "CSVDataset"

I'm quite new to the library dgl so I was following tutorial on how to use their class dgl.data.CSVDataset() for reading your own datasets. My example -graph classification section chapter and I'm even using the same data for this example that was…
Keithx
  • 2,994
  • 15
  • 42
  • 71
1
2
3 4