Create an Embedding Recorder

MarkovML uses Recorder to record and store data in your MarkovML account's database. To record embedding-related data, use the EmbeddingRecorderutility from MarkovML followed by registering it to MarkovML by using register() so you can view the custom embeddings on the MarkovML UI.

πŸ“˜

Note

Before registering custom embeddings for a particular dataset, the dataset must first be registered with MarkovML. Visit the Register Dataset page for more information.

Provide the following details to record embedding using theEmbeddingRecorder():

  1. name: Give a unique name to your custom embedding.
  2. dataset_id: Add the dataset_id of the registered dataset you wish to create custom embeddings for.
  3. notes: Notes for future reference. (optional)

Sample Code

import markov
from markov import EmbeddingRecorder

# get dataset by name 
data_set = markov.dataset.get_by_name("my_dataset_name")
dataset_id = data_set.ds_id  # get dataset id

embedding_recorder = EmbeddingRecorder(
  name="Custom embedding name", 
  dataset_id=dataset_id,
  notes="Optional description for this custom embedding "
)
embedding_recorder.register()


What’s Next