Feat/worksheet nlq beta #1639
Feat/worksheet nlq beta #1639
Conversation
…ksheet-nlq-beta # Conflicts: # backend/requirements.txt
frontend/src/modules/Worksheets/services/analyzeTextDocument.js
Outdated
Show resolved
Hide resolved
tests_new/integration_tests/modules/s3_datasets/test_s3_dataset.py
Outdated
Show resolved
Hide resolved
| client2, prompt, persistent_env1.environmentUri, worksheet1.worksheetUri, '' | ||
| ).contains('UnauthorizedOperation', 'RUN_ATHENA_QUERY') | ||
|
|
||
|
|
There was a problem hiding this comment.
Commented code. I am not sure what we should do with this code block
There was a problem hiding this comment.
left it here to hear others opinions - not sure what is best approach is, will think on this and get back tomorrow with some thoughts
|
Some example views from the UI: Text To SQL:
Document Analyzer:
UserGuide Information in this PR: |
|
@dlpzx - added additional tests and believe I have addressed all comments The one pending item is when testing the happy paths for both
If you retry the test in isolation / after some time it will resolve itself. Looking at the service quota limits, Claude 3.5 Sonnet has a fairly low max invoke count per account per minute of 50 (note: we are still far under that per minute threshold but I think that or similar could be causing this throttling exception - other Claude 3 models have a much higher per minute tolerance of invocations) Not sure if we should opt for a Claude 3 version for the time being to remediate this testing case and also to improve UX for a user using these features, thoughts? |
backend/dataall/modules/worksheets/aws/bedrock_prompts/process_text_template.txt
Outdated
Show resolved
Hide resolved
| def __init__(self): | ||
| self._session = SessionHelper.get_session() | ||
| self._client = self._session.client('bedrock-runtime') | ||
| model_id = 'anthropic.claude-3-5-sonnet-20240620-v1:0' |
There was a problem hiding this comment.
I am facing errors of the type: An error occurred (ValidationException) when calling the InvokeModel operation: Invocation of model ID anthropic.claude-3-5-sonnet-20240620-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model. I think it is related to the enforced cross-region inference
There was a problem hiding this comment.
I wonder if you have faced something similar when using this model
Feature or Bugfix
Detail
Relates
Security
Please answer the questions below briefly where applicable, or write
N/A. Based onOWASP 10.
fetching data from storage outside the application (e.g. a database, an S3 bucket)?
evalor similar functions are used?By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.