- 675 Views
- 1 replies
- 0 kudos
MLflow autolging is not registering my experiments
When training a any ML model in a Databricks notebook, after calling model.fit() and train the model, before the model was automatically saved, but now is giving me this error:WARNING mlflow.utils.autologging_utils: Encountered unexpected error durin...
- 675 Views
- 1 replies
- 0 kudos
- 0 kudos
Hi @espartaco,The error message shows that there's an issue with SSL certificate verification when trying to connect to the Azure storage endpointCheck network and firewall configurations: You need to ensure that the network and firewall configuratio...
- 0 kudos
- 502 Views
- 1 replies
- 0 kudos
Applyinpandas executed twice
Hi,I have a dataframe containing records (sales) over time for +- 1000 different items, so based on these records each item has its own timeseries. The goal is to make predictions for each of these items. Since the behaviour of these items is very di...
- 502 Views
- 1 replies
- 0 kudos
- 0 kudos
Hi @fh ,To avoid this double execution, you can try using the concurrent.futures module in Python to parallelize the training of your models. This module provides a high-level interface for asynchronously executing callables.
- 0 kudos
- 506 Views
- 1 replies
- 0 kudos
Databricks documentation for training a local LLM
Im in the process of training a chat-bot for my team to use to learn about databricks and relevant tools quickly. Is there a place that I can easily (and legally) grab learning material in PDF or text?
- 506 Views
- 1 replies
- 0 kudos
- 0 kudos
Hi @acdello,Could you check this doc if that helps in between?
- 0 kudos
- 525 Views
- 1 replies
- 1 kudos
Feature Store - lookback_window does not work with primary keys of "date" type
I just discovered what I believe is a bug in Feature Store. The expected value (of the "value" column) is 'NULL' but the actual value is "a". If I instead change the format to timestamp of the "date" column (i.e. removes the .date() in the generation...
- 525 Views
- 1 replies
- 1 kudos
- 1 kudos
Thank you for answering. Yes, that is also what I figured out. In other words the lookback_window argument only works when using timestamp format for the primary key. I cannot see that this behavior is described in the documentation.
- 1 kudos
- 4317 Views
- 2 replies
- 2 kudos
Resolved! How to search the run id of an experiment run created in another notebook?
Hello,I have created an experiment using with mlflow.start_run(run_name='experment_1'):in a notebook say 'notebook_1'. In the 'Experiments' tab if I click on 'notebook_1', I am able to see 'experiment_1'. Now I am trying to search the experiment in ...
- 4317 Views
- 2 replies
- 2 kudos
- 376 Views
- 0 replies
- 0 kudos
error tu run btyd model
I run the model in april and ok but today I need run the model and I have error and it is not possible continue I change the penalizer_coef and nothing # fit a model with a larger penalizer coefficientbgf_engagement = BetaGeoFitter(penalizer_coef=100...
- 376 Views
- 0 replies
- 0 kudos
- 540 Views
- 0 replies
- 0 kudos
Debugging using vscode & databricks connect
Hi allI'm facing some difficulties when I use DataBricks Connect to debug my ML solution. A long story short, I want to investigate a few variables after I've conducted training. With the debugger at hand, I can simply place a breakpoint on the line ...
- 540 Views
- 0 replies
- 0 kudos
- 650 Views
- 0 replies
- 0 kudos
large scale yolo inference
I have 50 Million Images sitting on s3 I have a Yolov8 model trained with ultralytics and want to run inference on those images. I suspect I should be running inference using ML flow, but I am confused on how. I don't need to track experiments/traini...
- 650 Views
- 0 replies
- 0 kudos
- 5060 Views
- 4 replies
- 1 kudos
Vector Search Index Sync fails in Initializing
Vector Search Index Sync fails in Initializing. This index table was already up and running, and when I tried to sync it, it failed in Initializing. See the attached.
- 5060 Views
- 4 replies
- 1 kudos
- 1 kudos
The issue for us was most likely that we used CPU compute for the deployed embedding model, switching to GPU (small) solved the issue.
- 1 kudos
- 2171 Views
- 3 replies
- 0 kudos
Resolved! Initializing Vector Search index Sync failes with Failed to resolve flow: '__online_index_view'
When setting up a vector search in databricks using the bge_m3 (Version 1) embedding model available in system.ai schema, the setup runs for 20 minutes or so and then fails. Querying the served embedding models from the browser works perfectly fine. ...
- 2171 Views
- 3 replies
- 0 kudos
- 0 kudos
The issue was most likely to use a CPU compute for the deployed model, switching to GPU (small) solved the issue.
- 0 kudos
- 3242 Views
- 8 replies
- 0 kudos
Feature Store Model Serving endpoint
Hi,I am trying to deploy my model which was logged by featureStoreEngineering client as a serving endpoint in Databricks. But I am facing following error: The Databricks Lookup client from databricks-feature-lookup and Databricks Feature Store clie...
- 3242 Views
- 8 replies
- 0 kudos
- 0 kudos
Hi @damselfly20 unfortunately I can't help much with that as I've never worked with RAGs. Are you sure it's the same error though? @NaeemS's and my errors seems to be Java related and yours MLflow related.
- 0 kudos
- 1136 Views
- 1 replies
- 1 kudos
Resolved! Vectorsearch ConnectionResetError Max retries exceeded
Hi,we are serving a unity catalog langchain model with databricks model serving. When I run the predict() function on the model in a notebook, I get the expected output. But when I query the served model, errors occur in the service logs:Error messag...
- 1136 Views
- 1 replies
- 1 kudos
- 1 kudos
downgrading langchain-community to version 0.2.4 solved my problem.
- 1 kudos
- 1664 Views
- 2 replies
- 1 kudos
Building a Data Quality pipeline with alerting
Hi there,My question is how do we setup a data-quality pipeline with alerting?Background: We would like to setup a data-quality pipeline to ensure the data we collect each day is consistent and complete. We will use key metrics found in our bronze JS...
- 1664 Views
- 2 replies
- 1 kudos
- 1 kudos
Hi Kash!I know it might be too late, but if you managed to create this by yourself and you are struggling to scale the solution you could take a look at Rudol Data Quality, it covers up pretty much everything you mentioned with a focus on enabling no...
- 1 kudos
- 3203 Views
- 4 replies
- 3 kudos
Passing parameters in Databricks workflows
Hi Databricks, we have created several Databricks workflows and the `json-definition.json` for the same is stored inside version control i.e. GitHub. There are several parameters which are referred from params.json inside this job definition but the ...
- 3203 Views
- 4 replies
- 3 kudos
- 3 kudos
Have you considered using Databricks Asset Bundles? Very easy to parameterize!
- 3 kudos
- 2111 Views
- 4 replies
- 1 kudos
Resolved! Model flavour using feature store model training log_model()
Hi I'm have succesfully registered my model using the feature engineering client with the following codes:with mlflow.start_run(): # Calculate the ratio of negative class samples to positive class samples ratio = (len(y_train) - y_train.sum()...
- 2111 Views
- 4 replies
- 1 kudos
- 1 kudos
Thanks for your reply @robbe - yes I have created a custom pyfunc model which I can now use fe.score_batch() to return probabilities. Here is the code:# Calculate the ratio of negative class samples to positive class samples ratio = (len(y_train) - y...
- 1 kudos
