Neum AI provides a managed cloud enviornment to run and manage your pipelines. It offers capabilities like pipeline management, run scheduling, automatic synchronization as well as a high scale distributed architecture to run large jobs. Learn more about the difference between Local and Cloud enviornments for Neum AI.
Prerequisite: Make sure you have already configured you Neum AI pipeline in our
quickstart.
You will need a Neum AI API Key. Get one by going to
dashboard.neum.ai and creating an account. It’s free!
Test pipeline locally
Before we deploy the pipeline to Neum AI Cloud, lets run a quick local test to make sure things are working as expected.
Validate
Run a local validation of the pipeline
print("Pipeline validation successful? " + pipeline.validate())
Test pipeline
Run a test to make sure the pipeline runs smoothly end to end
result = pipeline.run()
info = pipeline.sink.info()
print("Vectors written in pipeline run: " + result)
print(info)
Test search
Finally, lets test the stored vectors
results = pipeline.search(query="What are the challenges with scaling RAG?", number_of_results=3)
for result in results:
print(result.metadata)
Deploy pipeline to Neum AI Cloud
To deploy the pipeline, we will leverage the built-in functions.
You will need a Neum AI API Key. Get one by going to
dashboard.neum.ai and creating an account.
from neumai.Client.NeumClient import NeumClient
neumClient = NeumClient(api_key = '<INSERT NEUMAI API KEY>')
neumClient.create_pipeline(pipeline=pipeline)
Once deployed, you can go to Neum AI Cloud dashboard to see the status and pipeline information. Alternatively, you can programmatically query:
from neumai.Client.NeumClient import NeumClient
neumClient = NeumClient(api_key = '<INSERT NEUMAI API KEY>')
neumClient.get_pipeline(pipeline_id='<pipeline_id from above>')
Search pipeline
from neumai.Client.NeumClient import NeumClient
neumClient = NeumClient(api_key = '<INSERT NEUMAI API KEY>')
neumClient.search_pipeline(pipeline_id='<pipeline_id from above>', query=<search query>)
