superduper

superduper copied to clipboard

It can return either a DataFrame or a list of dict. This helps us easily view the data in the database. For example: ```python db['documents'].data db['documents'].data[:10] ``` ```python db['documents'].data(id='xxxx') ```...

jieguangzhou

For all `encode` and `decode` in datatype, we have accepted an `info` variable, but it has not been used so far. It is suggested to remove it, or change it...

jieguangzhou

For each module, Including: 1. predict_one 2. predict 3. to_listener

jieguangzhou

In our vector search interface, we currently call `like({indexing_key: x}, vector_index=vector_index_name)`. Should we consider removing the `indexing_key` in future iterations? `like({x, vector_index=vector_index_name)`. This field corresponds to `vector_index.listener.key`, which means we...

jieguangzhou

Manage a CONSTANT module to handle all special characters.

1

For example: ``` _SOURCE = '_source' _LEAVES = '_leaves' _DB_LOAD_PREFIX = '?db.load' ``` This allows for global management of field information and facilitates problem troubleshooting during development, enabling developers to...

jieguangzhou

- [ ] #2087 - [ ] #1622 - [ ] #1545 - [ ] #1654 - [ ] #1626

fnikolai

[DistEnv] All outputs loaded in memory before bulk write to database

1

When the model has predicted lets say for 1 million data points, in `components/model.py : predict method` model stores the outputs of this 1 million data points into a single...

kartik4949

Avoid blocking calls. If not connected in 20-30 seconds, just abort

fnikolai

For the moment, we need to define the env variables of the configuration at every services we instantiate on the kubernetes stack. That includes the data backend credentials. However, for...

fnikolai

For insertions and queries, use a unified interface

5

For MongoDB and Ibis Currently: - MongoDB uses insert_many for data insertion and find for queries. - Ibis uses insert for data insertion and select for queries. For these very...

jieguangzhou