James Nguyen

ray_param=RayParams( num_actors=10, cpus_per_actor=1) config = { "input_data":args.input_folder, "booster":'gbtree', "verbosity":1, "eta":0.1, #0.1 "gamma":0, #0 "max_depth": tune.choice([5, 6]), "colsample_bytree":0.7, #0.7, "min_child_weight":1, 'objective':'multi:softprob', ### binary:logistic 'eval_metric':"mlogloss", ### mlogloss, error 'ray_param':ray_param } analysis =...

model_xgb = train( config, dtrain_svm, evals_result=progress, evals=watchlist, verbose_eval=True, num_boost_round=5, ray_params=config['ray_param'])

@Yard1 thank you for offering to help. I think it's the placement group that could not obtain the resource. Let's connect next week to dig further.

Hi, the library moved to https://github.com/microsoft/ray-on-aml So in your code in compute instance, can you do pip install --upgrade ray-on-aml then restart the kernel? Follow the example here: https://github.com/microsoft/ray-on-aml/blob/master/examples/quick_start_examples.ipynb

Did you run the code to tell AML to use a compute cluster to build docker image? `` ws.update(image_build_compute = 'mycomputecluster') ``

It was not updated to work with new version of the SDK. We have another multi-agent example here: https://github.com/microsoft/OpenAIWorkshop/tree/main/scenarios/incubations/copilot/travel_leisure It was written on 1+ version of Open AI SDK

I think it's because Azure AI Search recently updated their SDK. If your goal is to experience with multi-agent, then you can check out this newer demo: https://github.com/microsoft/OpenAIWorkshop/tree/main/scenarios/incubations/copilot/travel_leisure If you...

That lab is deprecated, can you try https://github.com/microsoft/OpenAIWorkshop/blob/main/scenarios/incubations/automating_analytics/README.md?

You can review the bicep template in this, and use azd to deploy: https://github.com/microsoft/OpenAIWorkshop/tree/main/infra

@rcy0228 @amitlals the secrets.env should be directly under [copilot ](https://github.com/microsoft/OpenAIWorkshop/tree/main/scenarios/incubations/copilot) directory. It should have following values if all features are to be used. AZURE_OPENAI_ENDPOINT="" AZURE_OPENAI_API_KEY="" USE_AZCS="False" AZURE_SEARCH_SERVICE_ENDPOINT="" AZURE_SEARCH_INDEX_NAME="payroll_hr" CACHE_INDEX_NAME="payroll_hr_cache" AZURE_SEARCH_ADMIN_KEY=""...