When it comes to machine learning, the platform plays a pivotal role in successful implementation. Databricks offers a best-in-class machine learning platform with cutting-edge features such as MLflow, Model Registry, Feature Store, and MLOps, which ...
We are trying to train a predictive ML model using the XGBoost Classifier. Part of the requirements we have gotten from our business team is to implement feature weighting as they have defined certain features mattering more than others. We have 69 f...
Hello @sjohnston2 here is some information i found internally: Possible Causes Memory Access Issue: The segmentation fault suggests that the program is trying to access memory that it's not allowed to, which could be caused by an internal bug in XGBo...
Hi, I've got a question about deploying an endpoint for Llama 3.1 8b. The following code should create the endpoint without scale-to-zero. The endpoint is being created, but with scale-to-zero, although scale_to_zero_enabled is set to False. Instead ...
Thanks for the reply @Walter_C. This didn't quite work, since it used a CPU and didn't consider the max_provisioned_throughput, but I finally got it to work like this: from mlflow.deployments import get_deploy_client client = get_deploy_client("data...
Hi, I've encountered a problem of serving a langchain model I just created successfully on Databricks.I was using the following code to set up a model in unity catalog:from mlflow.models import infer_signatureimport mlflowimport langchainmlflow.set_r...
I suspected the issue is coming from this small error I got: Got error: Must specify a chain Type in config. I used the chain_type="stuff" when building the langchain but I'm not sure how to fix it.
I'm using a Single Node machine with g5-2x-large to fine tune a LLaMa-2 model. My Come Notebook runs very smoothly on Google Col but when I try to run it on `Databricks`, it throws me the exact error given below:RuntimeError: Expected to mark a varia...
Hello @saleem_shady! Have you tried including the parameter ddp_find_unused_parameters=False in your TrainingArguments? Here's an example of how to include it: https://github.com/databricks/databricks-ml-examples/blob/master/llm-models/llamav2/llamav...
error: not found: type XGBoostEstimator Spark & Scala
@amal15 - can you please include the below to the import statement and see if it works. ml.dmlc.xgboost4j.scala.spark.XGBoostEstimator
Hello everyone,I have an Azure Databricks subscription with my company, and I want to use external LLMs in databricks, like claude-3 or gemini. I managed to create a serving endpoint for Anthropic and I am able to use claude 3.But I want to use a Gem...
I am looking for the more detailed resources comparing RAG to fine-tuning methods in AI models to processing text data with LLM in laymen notes. I have found one resource but looking for the more detailed view https://www.softwebsolutions.com/resour...
Hi,I have a transformer model from Hugging Face I have logged to MLFlow.When I load in using mlflow.transformers.load_model I receive a bunch of warnings: WARNING:urllib3.connectionpool:Connection pool is full, discarding connection: xxxx. Connection...
We have deployed Dolly (https://huggingface.co/databricks/dolly-v2-3b) as a REST API endpoint on our infrastructure. The notebook we used to do this is included in the text below my question.The Databricks infra used had the following config - (13.2...
I had a similar problem when I used HuggingFacePipeline(pipeline=generate_text) with langchain. It worked to me when I tried to use HuggingFaceHub instead. I used the same dolly-3b model.
I upgraded Tensorflow on Databricks notebook using %pip command. Now when running the training job, I get this error: "DNN library initialization failed."
I saw that Foundation Model API is not HIPAA compliant. Is there a timeline in which we could expect it to be HIPAA compliant? I work for a healthcare company with a BAA with Databricks.
Hi @yhyhy3 Foundation Model API's HIPAA certification:AWS: e.t.a. March 2024Azure: e.t.a. Aug 2024 HIPAA certification is essentially having a third party audit report for HIPAA. That is not the date that a HIPAA product offering may/will necessari...
Hi,I'm trying to enable inference table for my llama_2_7b_hf serving endpoint, however I'm getting the following error:"Inference tables are currently not available with accelerated inference." Anyone one have an idea on how to overcome this issue? C...
From the information you provided, it seems like you are trying to enable inference tables for an existing endpoint. However, the error message suggests that this feature may not be supported with accelerated inference.If you have previously disabled...
Can we deploy our own Custom LLM model in Databricks? If anyone has any material or link, please share with me.
Hi,I'm testing out LLM/RAG Databricks demo here: https://notebooks.databricks.com/demos/llm-rag-chatbot/index.html?_gl=1*1nj8hq2*_gcl_au*MTcxOTY0MDY4LjE2OTQ2MzgwNDU.# As part of the demo, I'm trying to create a vector search with the line below. vsc....
Hi @m12, Thank you for posting your question in the Databricks community. The vector search feature is currently undergoing a private preview. If you wish to participate, kindly complete the form provided below for onboarding. https://docs.google.com...
