Shiny gives a good product for an affordable price
April 26, 2022

Shiny gives a good product for an affordable price

Anonymous | TrustRadius Reviewer
Score 8 out of 10
Vetted Review
Verified User

Overall Satisfaction with Shiny

We deliver dashboards from specific models generated in R for diverse researchers working in the Research and Development department.

There are models generated to model different diseases (called disease modeling) that include working with patient data (historical data).

This can include Rx, fluids, and medical notes. It involves different structures of data, such as text, columnar, time series, and images.
  • Interacting with ML models.
  • Deliver an online service.
  • Ability to select the best model.
  • Speed
  • Work in the cloud.
  • Working with images.
  • Quick model into production.
  • Operability
  • Good for researchers working in R.
  • We saw a good involvement to researchers when showing their models in shiny.
  • We can have a quicker review from the user when the model is in production.
  • False positives can be found easily and they help the retraining of the model.
Shiny allows easy and fast development of a product into production whereas Jupyter Notebook can be broken really easily by a user. The idea of having a specific server that works with that model is very practical and it's a good advantage.
In the contrary, the quantity of models you could apply to your data is boxed due to R restrictions, while a Jupyter Notebook has much less of a limit.

Do you think Shiny delivers good value for the price?

Yes

Are you happy with Shiny's feature set?

Yes

Did Shiny live up to sales and marketing promises?

I wasn't involved with the selection/purchase process

Did implementation of Shiny go as expected?

Yes

Would you buy Shiny again?

Yes

Shiny is well suited for classical ML models that involve 1 specific type of data and the model uses a specific R library or package.

When the problem needs the application of several tools and libraries, Shiny could improve the speed of the model in production.

I really like how you can play with your models using real data.