Hi everyone,   My team have some problem with PROC GRADBOOST of VDMML package on SAS Viya 3.5. I think, it is better to separate the topics.   Monotonic Constraints - We read a blog about this propertie of the proc gradboost, but we are confused about this part   " Of note for this particular data set is that the target variable "BAD" is coded as 0 and 1 with our desired predicted event level being 1, representing a bad loan. Since SAS gradient boosting defaults its event level as the first value for nominal targets (0 in our case), we will need to reverse the relationships to match our desired state."  } Complete blog: https://blogs.sas.com/content/subconsciousmusings/2020/06/03/monotonic-constraints-with-sas/   It means that the positive event is 0 by default? It is possible to change it to 1 category? I think that's the reason, we have some non intuitive results. I would apprecciate the help of technical support of SAS.   Deterministic Results Models - We have a CAS with 1 Controller and 3 Workers in a MPP Arquitecture. We find that training models with the same person, the same data (prometed in global), and also same hyperparams, we got differents results when training and then scoring model and use some performance metric like AUC (Run 1, we got AUC 58.8 - Run 2, we got AUC 59,2, ...) . I have a hypotesis about the Arquitecture that make the result about models, but i would apprecciate that a technical of SAS or a Expert could explain that part. ¿How could it be resolved to get deterministic models?   Thanks,   Claudio ... View more

Hi all,   I'm making differents stages about make models to production. In my company is very mature about working with scala, pyspark and also python with data engineers. My data scientist work with SAS Viya in version 3.5 redhat 7.9 on onpremise. I would like to know if i can work my models in SAS and then publish or export to Databricks o Azure Machine Learning to make the data engineer continue the process to production.   I read about this and found articles and blogs, but i dont know if there is a disclaimer about the version of SAS Viya that i manage and others technical requirements that are not specified on blogs.   Please your help about this case.   thanks, Claudio   https://communities.sas.com/t5/SAS-Communities-Library/Running-SAS-Models-in-Azure-Synapse-and-Databricks-Without/ta-p/838418 https://odsc.medium.com/how-to-deploy-sas-and-open-source-models-to-azure-machine-learning-with-sas-model-manager-c4c977a3df52     ... View more

Hi everybody,    I'm working with SAS 9.4 connecting to Hadoop/Cloudera with SAS ACCESS TO. When i try a query with Pass-through,  it resolved fast in Hive Database, but the table is creating too slow in SAS. I think there are opportunities to improve the I/O perhaps.    I want some help to improve the I/O with huge table (>20 millions rows) on SAS 9.4, i have 3 compute server in SAS Grid with 8 cpu core and 64GB ram each one.   I know about options bufsize, bufno and blksize, but i dont have expertise to combine and get an efficient parameters to get better I/O. ... View more

Hi everyone, I have SAS Viya 3.5 om rhel 7.9, i try ussing sas model manager with model python published on CAS. When i try to score with Cas action Ds2.RunModel, it takes much time in contrast to python program. I'm ussing pickle model and lgbm algorithm I would like to know if there is a way to submit python program (.py) to execute on CAS. My idea is to write python program that load all data on python dataframe, predict and then write on path. This program should be execute with CAS. Thank, Claudio ... View more

Hello Everybody,   I would like to know if exist a guide to install Jupyterhub on a compute server of SAS Viya 3.5, the operative system we have is Redhat 7.9, also we need that the app will be accesible for all our users - about 50 users.    I try with several commands like "pip install jupyterhub", "pip install jupyter", but get some issues in the installation.    I need some guide for the install of Jupyterhub and also how to make accesible to my users on the active directory.   thanks a lot, Claudio   ... View more

Hi everyone,   I want to know if there is a way to limit how many scheduled process a user can have. In my company, there are users that want to schedule 2 o 3 process and i want to manage that they only can schedule this 2 or 3 process. That is because i understand more schedule process without any management could finish with a performance problem or unavailability on the platform.   Perhaps, there is an environment parameter on SAS 9.4 about this Schedule Manager and how to limit process scheduled, RAM usage, others..   Thanks, Claudio ... View more

Hi everbody,   In my company, we are looking for the solution SAS Risk Modelling on SAS Viya Azure, I understand that as the solution SAS 9.4 SAS Credit Scoring, this solution needs a database to work.    Reading the documentation of SAS Risk Modelling, it suppose to support Teradata and SAS FileSystem, i want to know if there is a update on the documentation.. because i know that the FoundationMart would be accesible by CASLIB so why does it just support 2 technology? there is a problem why other database perhaps..synapseSQL, Hive, others?     For my case, could be a problem because we are working with Azure ADLS Gen2, Databricks and SynapseSQL.   I expect your help.   thanks, CDJ   #riskmodelling #sasviyaazure ... View more

Hi everbody,   I'm developer on SAS Viya 3.5 Linux, i want to create customized zones on snippets to share with all users SAS.    For example, now i can create and modify just my snippets, but i can not handle a customized snippet zone for all users and add new fragments or codes. I would like to know the step by step to manage this requirements and also improve the coding experience with users.     thanks, Claudio     ... View more

Hi everybody,   I want to know if it is possible to create hdat format file without take the step of loading in memory the dataset first and then save as hdat. I'm a developer on SAS Viya 3.5 Linux.   Thanks all, Claudio ... View more

Hi everybody,   In my company we use SAS 9.4 Grid on Linux Platform with 3 computer server, also file storage on spectrum scale IBM.    The last week, we had several problems with performance on the platform, tasks that usually end around 1 minute start to take about 5-8 minutes to finish. I don't take a snapshot but one time takes about 15 minutes, same data, same code, same all.     Also, i see that the real time is bigger in contrast to the user cpu time..     I´m a developer, my team is looking for the problem in the platform, but i think i can get some help in this forum to start looking. What kind of test would you apply to see performance problems? How to identify if the problem is about storage, servers communication, network or perhaps application like SAS.   Thanks for all, Claudio       ... View more