"Connectors" from Activeloop webpage in documentation/source?

Hello!

I have been looking for a code like yours for a while, and I am interested in self-hosting datasets because of special characteristics of the I/O and interconnect of our supercomputers [1], plus the massive size of the datasets we are aiming for in our research institute [2].

However, your documentation as of Feb 23, 2021, does describe only S3, MinIO, GCS and Azure. However, your webpage describes (I’m pasting directly):

Connect your pipelines to any type of structured and unstructured data in the Powerful Cloud-Native Array Data Warehouse.
AMAZON S3  GOOGLE CLOUD STORAGE  BLOB  REDSHIFT  SNOWFLAKE  BIGQUERY  MYSQL  POSTGRESQL

I can’t find any reference to mysql or postgresql, for example, either on the documentation or on the source. The only reference to local data is “Create a dataset” on the README of Github repo - with no specification of underlying format or anything else. Am I missing something?

[1] https://www.fz-juelich.de/ias/jsc/EN/Expertise/Supercomputers/JUWELS/JUWELS_node.html [2] https://www.helmholtz.ai

[surak]

Thanks! Check your email. You might delete it from here to avoid the spam bots!

[mynameisvinn]

Hi @surak whats the best way to reach you? I’m at [deleted!] or @vinn on hub’s slack.