IBM Data Observability by Databand
IBM SoftwareReviews from AWS customer
0 AWS reviews
-
5 star0
-
4 star0
-
3 star0
-
2 star0
-
1 star0
External reviews
67 reviews
from
External reviews are not included in the AWS star rating for the product.
A good platform for flow managament & visualization
What do you like best about the product?
The easy flow viewing ability, easy input & output inspection and the ability to integrate with various other tools, as well as the source code itself.
What do you dislike about the product?
Some UI elements are a bit clumsy. The left panel in "runs" is a good example where things feel a bit rough.
What problems is the product solving and how is that benefiting you?
Databand allows us to manage and visualize lots of long-running executions for machine learning operations, and understand each and every component and even restart failed runs in a breeze.
Great experience on all platforms
What do you like best about the product?
I like the UI and UX of Databand.
It was very comfortable to use the Basic and User params, and get the output of each operator in a very clear way.
I enjoyed the way Databand integrated into Python.
It was very comfortable to use the Basic and User params, and get the output of each operator in a very clear way.
I enjoyed the way Databand integrated into Python.
What do you dislike about the product?
Sometimes the logs were not informative enough, especially if the failure is from the pod Kubernetes side.
I think it can be improved by integrating better with the logs that came from the pods and nodes.
I think it can be improved by integrating better with the logs that came from the pods and nodes.
What problems is the product solving and how is that benefiting you?
It solves the need for a very comfortable view of the pipelines we make.
Filtering, searching, and monitoring, and also getting the needed statistics.
It was extremely comfortable to integrate into Python, and it made our Pipeline creation system much more user-friendly.
Filtering, searching, and monitoring, and also getting the needed statistics.
It was extremely comfortable to integrate into Python, and it made our Pipeline creation system much more user-friendly.
using dbnd for controlling the training process of our models
What do you like best about the product?
I like that I see all info on one screen,
and the graphs are very useful
and the graphs are very useful
What do you dislike about the product?
no colors in the log, for example, dates, error string, it hard to find the real issue msg.
by adding colors it would be easier to scan the log.
by adding colors it would be easier to scan the log.
What problems is the product solving and how is that benefiting you?
monitoring the state of the pipeline structure is easy to understand.
Reviewing databand for machine learning
What do you like best about the product?
the option to automate different pipelines
What do you dislike about the product?
sometimes the errors are not obvious and the errors are in some places
What problems is the product solving and how is that benefiting you?
can automate training processes and make a graph that is clear to understand
My experience with Databand has been great, it allows us to create complex ML based flow easily
What do you like best about the product?
Provides an easy way to design an efficient ML pipeline with a unified structure for all the pipelines we use in our systems.
User-friendly UI to help monitor everything we need.
User-friendly UI to help monitor everything we need.
What do you dislike about the product?
Sometimes it can be difficult to mimic rather easy behavior because some of the variables are dynamically created and we can rely on the output.
What problems is the product solving and how is that benefiting you?
We need an automatic way for deploying ML-based pipelines and allow different pipelines being ran based on different computer vision algorithms.
DataBand it's an awesome tool. to track the flow of your work
What do you like best about the product?
I like the clear UI/UX it has, the simplicity of adding visualisation for your code run is amazing
What do you dislike about the product?
Sometimes it throws unclear errors. nothing major though
What problems is the product solving and how is that benefiting you?
It helps me track our different model, develop new infrastructure and control the input and output in an outstanding way.
Had a good experience working with Databand at a very high scale
What do you like best about the product?
visibility of the experiments and their params and outputs is very good
very easy to filter through experiments and find the one you are looking for
also, the graph view is very nice and intuetive
very easy to filter through experiments and find the one you are looking for
also, the graph view is very nice and intuetive
What do you dislike about the product?
k8s errors are sometimes harder to detect but Databand team helps us make it better and better so it becomes less of an issue
What problems is the product solving and how is that benefiting you?
Databand helps us to orchestrate many tasks according to their dependencies and gives us visual representation of those tasks and a real-time view of whats going on
showing 61 - 67