- 5039 Views
- 4 replies
- 0 kudos
Resolved! Spark Version Json code for LTS ML cluster
Hi! Any help will be greatly appreciated! Thank you so much! So I'm currently following this tutorial: https://docs.databricks.com/applications/mlflow/projects.htmlIn the tutorial, for the cluster specification, it has "spark_version": "7.3.x-scala2....
- 5039 Views
- 4 replies
- 0 kudos
- 26282 Views
- 5 replies
- 2 kudos
Resolved! Keep long-running notebook alive when closing browser
I am working with Azure Databricks jupyter notebooks and have time-consuming jobs (complex queries, model training, loops over many items, etc.).Every time I close the browser (or step away for a long time) of some running notebook, even before the c...
- 26282 Views
- 5 replies
- 2 kudos
- 2 kudos
Hey @Eric P​ Just wanted to check in if you were able to resolve your issue. If yes, would you be happy to mark an answer as best? If not, please tell us so we can help you.Thanks!
- 2 kudos
- 48956 Views
- 9 replies
- 6 kudos
Resolved! MLFlow error
I am running into an error within the Databricks notebook (on Databricks website) environment where MLFlow will not load: MLflow autologging encountered a warning: "/databricks/python/lib/python3.8/site-packages/mlflow/utils/autologging_utils/safet...
- 48956 Views
- 9 replies
- 6 kudos
- 6 kudos
Install the Databricks SQL Connector for Python library on your development machine by running pip install databricks-sql-connector .Query data. Insert data. Query metadata. Cursor and connection management. Configure logging.Regards,Willjoe
- 6 kudos
- 9476 Views
- 2 replies
- 1 kudos
Resolved! Host R Shiny App
Can I host a R Shiny App on a Databricks cluster?
- 9476 Views
- 2 replies
- 1 kudos
- 1 kudos
I’ll be asking my rep about the hosted RShiny server in private preview— our team didn’t know about that so we’ve struggled through putting our shiny app (developed on Databricks using RStudio, that part was fantastic) into a container and hosting it...
- 1 kudos
- 2556 Views
- 1 replies
- 1 kudos
UnimplementedError; Operation not supported
I am trying to fit a model with callbacks including Tensorboard, history and checkpoint.Then I am loading the model and when trying to fit it again on more epochs, I am receiving this error:UnimplementedError: /dbfs/<path_to_history dir>/history.csv;...
- 2556 Views
- 1 replies
- 1 kudos
- 2592 Views
- 1 replies
- 2 kudos
MLflow
What kind of features I can use with MLflow?
- 2592 Views
- 1 replies
- 2 kudos
- 2 kudos
https://docs.databricks.com/applications/mlflow/index.htmlwww.MyAARPMedicare.com
- 2 kudos
- 2144 Views
- 0 replies
- 0 kudos
Hubs make key management easier (natural keys from hubs can be converted to surrogate keys via Identity columns).
In this blog https://databricks.com/blog/2022/06/24/prescriptive-guidance-for-implementing-a-data-vault-model-on-the-databricks-lakehouse-platform.htmlIt is mention that "Data Vault modeling recommends using a hash of business keys as the primary key...
- 2144 Views
- 0 replies
- 0 kudos
- 2610 Views
- 2 replies
- 0 kudos
Resolved! How does ML Flow work with Databricks monitoring product recently released ?
Does the monitoring tool interact with MLFlow and Feature Store too ?
- 2610 Views
- 2 replies
- 0 kudos
- 2792 Views
- 2 replies
- 1 kudos
Resolved! Databricks high availability
Hi all, i’m are trying to define a HA model within databricks, cuz the lastest failure on Azure, I would like to know if there is a way to have an HA model multicloud, or what do you recommend about this?Regards
- 2792 Views
- 2 replies
- 1 kudos
- 1 kudos
Agree, there are added costs to multicloud. Don't forget to peer your vnets to keep latency low.
- 1 kudos
- 969 Views
- 0 replies
- 0 kudos
7EFCFA82-CC44-4603-A103-58CE6F90863B
Great session about NLP going on now
- 969 Views
- 0 replies
- 0 kudos
- 5726 Views
- 3 replies
- 1 kudos
How to add signature to model logged through feature store?
It seems that the current method log_model from the FeatureStoreClient class lacks a way to pass in the model signature (as opposed as doing it through mlflow directly). Is there a workaround to append this information? Thanks!
- 5726 Views
- 3 replies
- 1 kudos
- 1 kudos
Hello!You can log a model with a signature by passing a signature object as an argument with your log_model call. Please see here.Here's an example of this in action in a databricks notebook.Hope that helps!-Amir
- 1 kudos
- 1288 Views
- 0 replies
- 0 kudos
jupytemplate in databricks
Hi, Most of my notebooks follow the same structure (i.e. load data, preprocessing, learn ML model, evaluate, etc.). I came across the jupytemplate package which allows to define a template for your notebooks. However, I can't seem to make it work in ...
- 1288 Views
- 0 replies
- 0 kudos
- 1763 Views
- 1 replies
- 1 kudos
Accessing served models from Excel
Anyone know of an Excel plug-in (or code example) that would connect to a databricks serverd model for scoring and present results back in Excel?
- 1763 Views
- 1 replies
- 1 kudos
- 1 kudos
I would create a python program to convert excel into csv or whatever format your model takes as input and use that input for the model. And vice versa. I can make it for you if you want
- 1 kudos
- 2027 Views
- 1 replies
- 0 kudos
Resolved! Does MLFLOW support the storage of Spark Pipeline API
Does the pipeline api allow for saving an entire transformation pipeline and render the artifacts in MLFLOW.
- 2027 Views
- 1 replies
- 0 kudos
- 0 kudos
Yes the pipeline API allows pickling of a pipeline which can in fact be stored as an artifact. This allows for easy reproducibility of production pipelines!
- 0 kudos
- 4112 Views
- 2 replies
- 0 kudos
Access multiple .mdb files using Python
Hi, I wanted to access multiple .mdb access files which are stored in the Azure Data Lake Storage(ADLS) or on Databricks File System using Python. Can you please help me by guiding how can I do it? It would be great if you can share some code snippet...
- 4112 Views
- 2 replies
- 0 kudos
- 0 kudos
https://community.databricks.com/s/question/0D58Y00008rCmBySAK/access-multiple-mdb-files-using-pythonmyEHtrip Employee Login
- 0 kudos
