M3 - Deploying your Prototype Internally in Streamlit - "Compute Pool" field

Course Q&A Fast Prototyping of GenAI Apps with Streamlit
,
1

When trying to deploy my prototype internally in Snowflake in Streamlit I get the below warning when I try to create the Streamlit App.

In the video this “compute pool” field was not visible. Obviously the conditions for a trial user for Snowflake have changed slightly since the creation of the course. I’m not sure how best to proceed to step through the remaining videos of the course. Thanks

image
image761×745 49.2 KB

2

@jan.ravnik, can you look into this?

1 Like
3

Hi @kate_gould1

If you’re seeing no compute pool, here’s the solution below.

To fix this, you can create a compute pool by running the following SQL statement in a SQL worksheet.

  1. Click on “+” > “SQL File“
  2. Paste the following SQL statement into the SQL worksheet and run:
CREATE COMPUTE POOL tutorial_compute_pool
  MIN_NODES = 1
  MAX_NODES = 1
  INSTANCE_FAMILY = CPU_X64_XS;
  1. You should see the following output if the run is successful:
Compute pool TUTORIAL_COMPUTE_POOL successfully created. Please run 'DESCRIBE COMPUTE POOL TUTORIAL_COMPUTE_POOL' to check the compute pool state. NOTE the compute pool is not ready to deploy a service or job before reaching ACTIVE or IDLE state.
  1. Refresh the browser once or twice just to be sure that the new compute tool gets reflected
  2. Head back to creating a Streamlit app and you should see the recently created compute pool.

Hope this helps :slight_smile:

Best regards,
Chanin

cc: @TMosh @jan.ravnik

2 Likes
4

Thanks @dataprofessor

1 Like
5

Hi @kate_gould1 @jan.ravnik It looks like the field compute pool appears when we select “Run on Container” ; that is not the case when we select “run on warehouse” as we did in previous videos.