My favorite part about Model Manager was the ability to automate processes that would otherwise take a lot more time & energy in just R or Python. For example, the train and validation split of the data is automatically taken care of when running the model pipeline. Same goes for comparing the model outputs (Accuracy, area under ROC, etc.). Review collected by and hosted on G2.com.
Parameter tuning wasn't as intuitive (at least from the perspective of someone who is more familiar with coding this whole process). I thought it was difficult to understand many of the parameter tuning options, as well as hard to search for the specific tweaks I was looking to do. Review collected by and hosted on G2.com.
I love how simple Model Manager made working with classical ML models. Traditionally these models are made at the start of every task when working with data as an elementary step, but manual coding takes up a decent amount of time. With Model Manager I was pleased to see how simple the process was and how it even provided comparisons and tons of accuracy metrics. Review collected by and hosted on G2.com.
I would have preferred more documentation. With every new tool there is a learning curve but figuring out the small details relied on my own efforts of exploring the site for a while. There is a lot of documentation and instruction overall with SAS but it is hard to navigate and find the specific information I need. Review collected by and hosted on G2.com.
What I like best about SAS Visual Analytics is how it empowers users to explore data visually without needing deep technical skills. The drag-and-drop interface and interactive dashboards make it easy to uncover insights quickly and share them across teams. Review collected by and hosted on G2.com.
Sometimes, performance can be an issue when working with large datasets or complex reports—especially in cloud environments. Optimizing for speed and responsiveness could make a big difference. Review collected by and hosted on G2.com.
Visual modeling without code: The user interface allows the creation of machine learning models via drag and drop. This is particularly useful for analysts who are not necessarily programmers but still want to create complex models.
AutoML functionality: SAS VML can automatically create multiple models, compare them, and suggest the best model - including transparency in evaluation and selection. Review collected by and hosted on G2.com.
Since VML is designed for no-code usage, the integration of user code, especially in non-native platform languages, is not particularly well supported (e.g., no syntax highlighting in the code editor). Review collected by and hosted on G2.com.
It's great to be able to build machine learning models in the same environment where dashboards are created. Review collected by and hosted on G2.com.
Models created in this environment cannot have the same complexity as models created in Model Studio. Review collected by and hosted on G2.com.
SAS Visual Analytics is so easy to implement and start building visualizations. The drag and drop visuals are easy to use and look great. I use SAS Visual Analytics daily for a wide range of use cases. Review collected by and hosted on G2.com.
Sometimes data load times can be slow (depending on size of datasets being used). Review collected by and hosted on G2.com.
SAS Visual Machine Learning lets you easily build your ML models, various algorithms being available to deploy. You can build a new model by just copying the existing model and changing the algorithm as needed. Review collected by and hosted on G2.com.
The entire workflow is not as intuitive as it was in SAS Enterprise Miner. Review collected by and hosted on G2.com.
Model manager allows me to view all my models in a singular place. Allows me to share and collaborate with others in my team Review collected by and hosted on G2.com.
. But better functionality in viewing who owns the model. Who has done updates to the model in an audit capability would be good Review collected by and hosted on G2.com.
All the infinite possibilities you can create as it is a white page where you can work and create amazing thing according to your necessities. Review collected by and hosted on G2.com.
The slow reaction as it increases the objects in the page Review collected by and hosted on G2.com.
You may get insights in few clicks. Easy to use. Review collected by and hosted on G2.com.
Documentation from the current Viya version is not so good as the previous one. Also, I was not able to find the Case study PDF, the one that solves the same problem in 3 ways, according to the user's skills. Review collected by and hosted on G2.com.