neural_graph_collaborative_filtering

neural_graph_collaborative_filtering copied to clipboard

How to calcuate nDCG in this paper?

1

I'm wondering how to calcuate nDCG paper? :confused: **Question:** The nDCG calculation needs the link predictions of all unobserved items, but It is not tractable(because of sparse density). What is...

swyo

The embedding propagation code seems to be not consistent with the paper

1

temp_embed = [] for f in range(self.n_fold): temp_embed.append(tf.sparse_tensor_dense_matmul(A_fold_hat[f], ego_embeddings)) side_embeddings = tf.concat(temp_embed, 0) sum_embeddings = tf.nn.leaky_relu(tf.matmul(side_embeddings, self.weights['W_gc_%d' % k]) + self.weights['b_gc_%d' % k]) bi_embeddings = tf.multiply(ego_embeddings, side_embeddings) bi_embeddings = tf.nn.leaky_relu(tf.matmul(bi_embeddings,...

Dousia

Could you release the pre-processing codes?

1

Could you release the pre-processing codes? Because I found the statistics of Gowalla dataset (after 10-core preprocessing ) is different with yours. Thanks so much.

RileyLee95

a problem about matrix inner produce

3

  the result from second image is not equal sum(first image red mark). i don't understand , the code use multiply too , could you give some detail ?...

sunshinelium

adj_type and _create_ngcf_embed

5

hi, there is not a adj_type named "ngcf" in the code, only has "norm"、“plain” and “gcmc”. The default setting of adj_type is "norm", but the method of calculate the laplacian...

wushihu

An experimental phenomenon

2

I did an experiment that delete the propagation on graph (comment line:194-221), which means that only do dot product. It is amazing that it can get comparible results in gollaw...

wu6shen

Validation data

6

Hi, Thanks for making the code public! You mentioned the train/valid/test split in paper but I cannot find the validation data in this code. Could you provide the code that...

Songweiping

A787

wrong api used

1

``` def _create_ngcf_embed(self): ... norm_embeddings = tf.math.l2_normalize(ego_embeddings, axis=1) ``` will caused `AttributeError: module 'tensorflow.math' has no attribute 'l2_normalize'` in tensorflow-gpu 1.8.0, so it should be `norm_embeddings = tf.nn.l2_normalize(ego_embeddings, axis=1)`

KimRasak

How to deal with the score of the original dataset? Is scoring an interaction? In a the five-grade marking system, if the user gives an item a score of 1,...

inoryw