- 15 Views
- 1 replies
- 1 kudos
Resolved! Options sporadic (and cost-efficient) Model Serving on Databricks?
Hi all,I'm new to Databricks so would appreciate some advice.I have a ML model deployed using Databricks Model Serving. My use case is very sporadic: I only need to make 5–15 prediction requests per day (industrial application), and there can be long...
- 15 Views
- 1 replies
- 1 kudos
- 1 kudos
Hi @cbossi , You are right! A 30-minute idle period precedes the endpoint's scaling down. You are billed for the compute resources used during this period, plus the actual serving time when requests are made. This is the current expected behaviour. Y...
- 1 kudos
- 31 Views
- 1 replies
- 1 kudos
Model Registration and hosting
I have train & tested a model in databricks, now I want to register it and host it. But I am unable too do so. Please find attach snapshot of code & error
- 31 Views
- 1 replies
- 1 kudos
- 1 kudos
Hi @intelliconnectq The above code will fail with AttributeError: 'NoneType' object has no attribute 'info' on the line: model_uri = f"runs:/{mlflow.active_run().info.run_id}/xgboost-model" This happens because once the with mlflow.start_run(): bloc...
- 1 kudos
- 3393 Views
- 1 replies
- 0 kudos
Surprisingly sparse_logs and tensorboard logfiles in Databricks-Workspace
Hi, surprisingly we have found 2 new folders with some short logfiles in our Databricks workspace:ls -lFr sparse_logs/ tensorboard/tensorboard/:-rwxrwxrwx 1 root root 88 Sep 2 11:26 events.out.tfevents.1725275744.0830-063833-n68nsxoq-10-139-64-10.20...
- 3393 Views
- 1 replies
- 0 kudos
- 0 kudos
Hey @steve2 , short answer: these look like TensorBoard event files, likely created by a library that briefly initialized a TensorBoard logger or writer during one of your training/serving runs; the sparse_logs folder naming and “manager stage: Mode...
- 0 kudos
- 5572 Views
- 3 replies
- 0 kudos
Accessing Databricks Volumes from a Serving Endpoint Using a Custom Model Class in Unity Catalog
Hi everyone,I’m looking for accessing Unity Catalog (UC) Volumes from a Databricks Serving Endpoint. Here’s my current setup:I have a custom AI model class for inference, which I logged into Unity Catalog using mlflow.pyfunc.log_model.I’ve created a ...
- 5572 Views
- 3 replies
- 0 kudos
- 0 kudos
Greetings @VELU1122 , you’re correct that the Databricks Model Serving container is isolated, so you can’t rely on cluster-only affordances like mounts or executor-distributed file utilities. The reliable way to read from Unity Catalog (UC) Volumes ...
- 0 kudos
- 3542 Views
- 1 replies
- 1 kudos
Lakehouse Monitoring of Inference Table
All,I'm trying to setup a lakehouse monitoring process for the WineQuality model that is widely available. While setting up the Serving Endpoint, I enabled "Inference Table" option for which the inference table was created automatically. The columns ...
- 3542 Views
- 1 replies
- 1 kudos
- 1 kudos
Hello @grajee , I can see you're dealing with two separate issues here. Let me address both: Issue 1: The model_id column (request_metadata MAP type) You're correct that request_metadata is a MAP type and can't be directly used as the model_id colum...
- 1 kudos
- 3611 Views
- 1 replies
- 0 kudos
Custom AutoML pipeline: Beyond StandardScaler().
The automated notebook pipeline in an AutoML experiment applies StandardScaler to all numerical features in the training dataset as part of the PreProcessor. See below.But I want a more nuanced and varied treatment of my numeric values (e.g. I have l...
- 3611 Views
- 1 replies
- 0 kudos
- 0 kudos
Greetings @sharpbetty Great question! Databricks AutoML's "glass box" approach actually gives you several options to customize preprocessing beyond the default StandardScaler. Here are two practical approaches: Option A: Pre-process Features Before ...
- 0 kudos
- 3555 Views
- 2 replies
- 3 kudos
Resolved! AutoML master notebook failing
I have recently been able to run AutoML successfully on a certain dataset. But it has just failed on a second dataset of similar construction, before being able to produce any machine learning training runs or output. The Experiments page says```Mo...
- 3555 Views
- 2 replies
- 3 kudos
- 3 kudos
Hi @dkxxx-rc , Thanks for the detailed context. This error is almost certainly coming from AutoML’s internal handling of imbalanced data and sampling, not your dataset itself. The internal column _automl_sample_weight_0000 is created by AutoML when i...
- 3 kudos
- 3466 Views
- 1 replies
- 1 kudos
Machine Learning Model Deployment on Databricks with Unity Catalog
Hi everyone! I found it would be helpful to document and share my experiences navigating model deployment on Databricks with the recent changes to deploy models to Unity Catalog instead of the Workspace Model Registry. https://medium.com/p/7d04e85395...
- 3466 Views
- 1 replies
- 1 kudos
- 1 kudos
Thanks for sharing this in-depth piece, @ac10 . Your walkthrough of model deployment via Unity Catalog is clear and practical—especially the insight about handling model signatures when working with Spark DataFrames. This will definitely help practit...
- 1 kudos
- 3455 Views
- 2 replies
- 0 kudos
Error when creating model env using 'virtualenv' with DBR 14.3
We were trying to inference from a logged model but had the following errorPreviously, we had been using `conda` as the environment manager, but that is no longer supported. I tried to update pyenv as some suggested but didn't get anywhere. Any insig...
- 3455 Views
- 2 replies
- 0 kudos
- 0 kudos
Hello @drjb1010 , This is a known issue with DBR 14.3 where the `virtualenv` environment manager fails because it depends on `pyenv` to install specific Python versions, but `pyenv` is either not installed or not properly configured in the runtime e...
- 0 kudos
- 3331 Views
- 1 replies
- 0 kudos
Patient Risk Score based on health history: Unable to create data folder for artifacts in S3 bucket
Hi All,we're using the below git project to build PoC on the concept of "Patient-Level Risk Scoring Based on Condition History": https://github.com/databricks-industry-solutions/hls-patient-riskI was able to import the solution into Databricks and ru...
- 3331 Views
- 1 replies
- 0 kudos
- 0 kudos
Greetings @SreeRam , here are some suggestions for you. Based on the error you're encountering with the hls-patient-risk solution accelerator, this is a common issue related to MLflow artifact access and storage configuration in Databricks. The probl...
- 0 kudos
- 3650 Views
- 1 replies
- 1 kudos
Table-Model Lineage for models without online Feature Lookups
Hi community,I am looking for the recommended way to achieve table-model lineage in Unity Catalog for models that don't use Feature Lookups but only offline features. When I use FeatureEngineeringClient.create_training_set with feature_lookups + mlfl...
- 3650 Views
- 1 replies
- 1 kudos
- 1 kudos
Hey @ssequ sorry this fell through the cracks but I have some ideas for you to consider. You can get Unity Catalog table→model lineage without Feature Lookups by logging the training datasets to MLflow and registering the model in Unity Catalog. ...
- 1 kudos
- 4271 Views
- 1 replies
- 1 kudos
AutoGluon MLflow integration
I am working on a personalized price package recommendation and implemented an AutoGluon code integrating it with MLflow.The code has been created in a modular fashion to be used by other team members. They just need to pass the data, target column a...
- 4271 Views
- 1 replies
- 1 kudos
- 1 kudos
Hi @cleversuresh Thanks for sharing the code and the context. Here are the core issues I see and how to fix them so MLflow logging works reliably on Databricks. What’s breaking MLflow logging in your code Your PyFunc wrapper loads the AutoGluon mod...
- 1 kudos
- 1247 Views
- 7 replies
- 3 kudos
Resolved! What is the most efficient way of running sentence-transformers on a Spark DataFrame column?
We're trying to run the bundled sentence-transformers library from SBert in a notebook running Databricks ML 16.4 on an AWS g4dn.2xlarge [T4] instance.However, we're experiencing out of memory crashes and are wondering what the optimal to run sentenc...
- 1247 Views
- 7 replies
- 3 kudos
- 3 kudos
@excavator-matt I’d recommend a quick refresher on the Pandas API on Spark to understand the implementation details. This video breaks it down clearly: https://youtu.be/tdZDotqKtps?si=pcIzCUYs2s_TeQKx Hope this helps. — Louis
- 3 kudos
- 1597 Views
- 8 replies
- 1 kudos
Resolved! Importing sentence-transformers no longer works on Databricks runtime 17.2 ML
In Databricks Runtime 16.4 LTS for Machine Learning, I am used to be able to import sentence-transformers without any installation as it is part of the runtime with from sentence_transformers import SentenceTransformer.In this case I am running on a ...
- 1597 Views
- 8 replies
- 1 kudos
- 1 kudos
I now upgraded to the new 17.3 LTS ML and it now works. I didn't try 17.2 ML, but with 17.3 ML available, I don't see any reason to use it anymore.
- 1 kudos
- 4945 Views
- 6 replies
- 1 kudos
Resolved! Model Serving Endpoint Creation through API
Hello,I am trying to create a model serving endpoint via the API as explained here: https://docs.databricks.com/api/workspace/servingendpoints/createI created a trusted IAM role with access to DynamoDB for the feature store. I try to use this field,"...
- 4945 Views
- 6 replies
- 1 kudos
- 1 kudos
If you're using the databricks terraform provider, make sure the role's name matches the instance-profile name.If not, use the `iam_role_arn` attribute to explicitly set the role's arn when creating the databricks instance profileresource "databricks...
- 1 kudos
