GDELT/chain_results.json missing some nodes

[Kewth]

The GDELT dataset contains 6,786 nodes, but the pre-processed file GDELT/chain_results.json only contains entries for 5,849 nodes.

When running the model, the code attempts to access features for node IDs that exist in the graph but are missing from the JSON file, causing the following crash:

File ".../TemporalChainModules.py", line 39, in find_temporal_chain_features
    temporal_chain_timestamp, temporal_chain_feature = self.temporal_chain_features[node_id]
KeyError: np.int64(5910)

Is the provided chain_results.json outdated/incomplete?

[SiweiPro]

Hi, thanks for following.

The maximum entity ID is 5849 in the GDELT dataset, so please check that you are using the right dataset.

Thanks again and good luck!