Ankit Pal

@Meghna-Goyal you can use GAE ( Graph auto-encoder ) model for link prediction. Train the embedding on adj and feature matrix and then use that embedding to classify the edges.

Thanks for those files Thomas, But can you let me know how to get allx, ally, graph,index, tx, tx, x, y from load function ? Currently I am getting adj,...

I am working on a Graph network framework, where you can just import any graph network (GCN, GAT etc ) and preprocess the data simply and train the model. I...

@kdmsit if I am getting your question correctly then : **adj** is the adjacency matrix of the graph, with all the nodes (it doesn't matter if they are part of...

check this related issue #149

You can use this, if you are dealing with multi-label, ``` class Optimizers(object): @staticmethod def multilabel_optimizer(logits, ground_truth, learning_rate): cross_entropy = tf.nn.sigmoid_cross_entropy_with_logits(logits = logits, labels = tf.cast(ground_truth, tf.float32)) loss = tf.reduce_mean(tf.reduce_sum(cross_entropy,...

I also need the pre-processing script. Any update on this issue?

Hi, I am working on this but confused on dataset. There is no clear explanation how to convert original dataset for gcn. Can you provide raw data preprocessing code files?...

We are working on documentation and adding more detailed examples.

Hi, Are you referring to open-source LLM models or Hugginface transformer models such as Bert etc?