Hi all. I've been having a lot of trouble uploading some data from a .conllu file into a nebula graph space managed by the docker extension. My project is a python project. After failing to properly import the data based on my schema and model I tried using the AI importer in nebula studio on my dataset (a .conllu file) From there I downloaded the schema dll. But I'm still struggling to figure out how to import the data into nebula graph while having it fit my model. Any help or guidance would be greatly appreciated. I will want to be importing big datasets so need some batching capabilities too. Should I be using nebula-carina to simplify my life? Is that ORM layer still in active development? And would it slow down my graph operations (this code is part of a GNN).

Hi, NebulaGraph does not support importing

.conllu

files directly. NebulaGraph requires some node files and some edge files, and their file format is best as CSV. The AI importer is an experimental feature. For more information, you can refer to: https://docs.nebula-graph.io/3.8.0/import-export/write-tools/

Thanks. I’ve got my conllu importer working now. Slowly getting there. Is there an ideal batch size for importing tags, edges and indexes? Do I need to add delays in between the batches?

You can start with a batch size of

1024

for importing. Typically, you don't need to add delays between batches.