-
## The problem
I built a 3-layer CNN in TF Learn and was unable to learn anything. It gets stuck at a loss of 0.69315 even after 17 iterations. Comparably, in Keras, if I train for 17 iterations, the…
-
As I trained, the result of this repository is more accurate than the original paper (SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS) in cora dateset.
What is different from the o…
-
Could you suggest where I can implement a graph convolutional network? I would like to take the features from Faster R-CNN and pass them through a graph conv net. I'm unsure if I can do this in a new …
-
-
We should add the section, so advanced users can skip certain notebooks / skip to certain sections.
Also rename the section (either "Learning Objectives" or "In this section you'll learn".
| Noteb…
-
Hello, In this paper 《Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition》,I want to ask how to modify the forward propagation part of the class GGCN in model.py. I don…
lulig updated
3 years ago
-
### Name
Gregory Wolfe
### Email
gw2338@nyu.edu
### Dataset name
Defect GNN
### Authors
Md Habibur Rahman, Prince Gollapalli, Panayotis Manganaris, Satyesh Kumar Yadav, Ghanshyam Pilania, Brian…
-
## 🚀 Feature
Please add implementation for HGCN (https://github.com/HazyResearch/hgcn) and HGNN (https://github.com/facebookresearch/hgnn), if feasible and at your convenience.
## Motivation
…
-
Structured data arises in a few of the domains discussed throughout the review. For example, CNNs perform tremendously for many tasks based on data from a regular graph like images ([subsection here](…
-
the cora_adj.npy shows cora dataset have 5278 edges
but the detail of this dataset(https://github.com/yueliu1999/Awesome-Deep-Graph-Clustering/blob/main/dataset/README.md)
shows this dataset hav…