chrsmrrs / tudataset Goto Github PK

Hi,

I m trying to train a QSAR model some GNNs. But what I have is a list of molecular SMILES and the corresponding labels, for example:

x = ['CCCC', 'CCCO', 'CCCN' ...]
y = [1,1,0, ...]

Many of GNNs are taking your datasets as examples, which are preprocessed graph described by a list of files. But I did not find any way to convert to this format from molecular SMILES. Is there any function to convert molecular SMILES to the same format?

Thanks

I have been playing with ENZYMES dataset and found that there are no feature names for node features provided in README or anywhere else (or I need help finding them). While in the referenced paper (Bogwardt et al. 2005) it could be found that features AA length, Total Waals etc. I do not know which feature name corresponds to which feature value in tudataset.

Please provide feature name for each feature dimension in the dataset. It would be a lot more easier to do anything with dataset, starting from data exploratory analysis, when features are stated in README, e.g., in MUTAG dataset.

Analogous to the TUDataset， I have now understood and made the original format consistent with TUDataset, how should I write the class code ???(for processing the data set so that it can be pytorch_geometric)
I have prepared the raw data like this:

(1) XX_A.txt (m lines)
sparse (block diagonal) adjacency matrix for all graphs,
each line corresponds to (row, col) resp. (node_id, node_id)

(2) XX_graph_indicator.txt (n lines)
column vector of graph identifiers for all nodes of all graphs,
the value in the i-th line is the graph_id of the node with node_id i

(3) XX_graph_labels.txt (N lines)
class labels for all graphs in the dataset,
the value in the i-th line is the class label of the graph with graph_id i

(4) XX_node_attributes.txt (n lines)
matrix of node attributes,
the comma seperated values ​​in the i-th line is the attribute vector of the node with node_id i

Hello blogger, I have made a preprocessing data set by myself and want to use TG for training. Do you know how to write your own dataset or use TUDataset to process it into a trainable data format? (72 graphs, 63 nodes each, with 2300 features for each node that is not hot-coded)>>>