send-to-squey API documentation

class send_to_squey.SqueyInstance(endpoint, auth, disable_server_verification, compression_codec)[source]

Bases: object

import_data(data, dataset_name=None)[source]

Uploads a dataset to the cloud Squey instance and imports it.

Parameters:
send_to_squey.connect(endpoint, auth, disable_server_verification=False, compression_codec='lz4')[source]

Connect to Squey server and returns a SqueyInstance object.

Parameters:
  • endpoint (string) – the connection endpoint.

  • auth ((string, string)) – a tuple composed of username and password.

  • port (int) – the apache arrow flight server port.

  • disable_server_verification (bool) – disable SSL server verification.

  • compression_codec (string) – compression codec used (“lz4”, “zstd” or None; defaults to “lz4”)

Returns:

a SqueyInstance object

send_to_squey.start_instance(instance_id, cloud_provider=CloudProvider.AWS, profile_name=None, region_name=None)[source]

Start a cloud instance.

Parameters:
  • instance_id (string) – the id of the instance

  • cloud_provider – a value of the CloudProvider enum

  • profile_name (string) – the profile name

  • region_name (string) – the region code

Examples

Connect and send a dataset

from squeylab import send_to_squey
import pandas as pd

squey = send_to_squey.connect(
    endpoint="3.222.243.61.aws.squeylab.com",
    auth=("squey", "p@$$w0rd!")
)

df = pd.DataFrame(data={'col1': [1, 2], 'col2': [3, 4]})

squey.import_data(df, dataset_name="my_dataset")

Note

import_data data parameter can be any object compatible with pyarrow.record_batch data parameter, or a path toward a parquet file.

Start the instance

from squeylab import send_to_squey

send_to_squey.start_instance(instance_id="i-0eb3fbba537abfa95")

Note

Only supporting AWS for the moment. Needs a properly configured credential file.