Microsoft Azure Machine Learning Studio Reviews

I use a combination of Microsoft Azure Machine Learning Studio and Azure Databricks. I mostly use Azure Databricks for building a machine learning system. There are several workflows for a machine learning tuning system that involves data pre-processing, quick modeling pipelines that execute within a couple of seconds, and complex model pipelines, such as hyperparameters. Additionally, there is a setting to set different AutoML parameters. 

For the training and evaluation phase of the whole machine learning system, I use MLflow, for a testing system and a model serving system, which is one core component of Databricks. I use it for Model Register and it allows me to do many things, such as registering model info, logs, and evaluation metrics.

The newer version of this solution has better integration with automated ML processes and different APIs. I feel like it is quite powerful in terms of general machine learning features, such as training data handily by having different sampling methods and has more useful modeling parameter settings. People who are not data scientists or data analysts, can quickly use the platform and build models to leverage the data to do some predictive models.

Azure Machine Learning Studio's most valuable features are the package from Azure AutoML. It is quite powerful compared to the building of ML in Databricks or other AutoMLs from other companies, such as Google and Amazon. It has the most sophisticated set of categories of parameters. The data encodings and options are good and it has the most detailed settings for specifics models.

I have found Databricks is a better solution because it has a lot of different cluster choices and better integration with MLflow, which is much easier to handle in a machine learning system.

The developers for this solution have not been as active in improving it as other solutions have had more improvements, such as Databricks.

Sometimes there might be some data drifting problems and this is what I am currently working on. For example, when our new data has a drift from the previous old data. I need to first work out a solution. Azure in Databricks or in Azure Machine Learning Studio both works fine. However, the normal data drifting solution is not working that well for the problem that I am facing. I am able to receive the distribution change and numerical metrics changes, but it will not inform me how to fix them.

I have been using this solution for approximately three months.

I use Databricks alongside this solution.

I rate Microsoft Azure Machine Learning Studio a seven out of ten.

We plan to use this solution for everything in business analytics including data harmonization, text analytics, marketing, credit scoring, risk analytics, and portfolio management.

It allows us to do machine learning experiments quickly.

We did not have machine learning solutions or platform earlier.

It's user-friendly, and it's a no-code model development. It's good for citizen data scientists, but also, other people can use Python, R or .NET code.

If you are on Microsoft Cloud, the development and implementation are super easy.

Every tool requires some improvement. They have already improved many things. They had added new features and a new pipeline.

They should have an on-premise version, other than Python and R Studio, which is only good for cloud-based deployments.

If they could have a copy of the on-premise version on Mac or Linux or Windows, it would be helpful.

It should have the flexibility to work o the desktop. They should have a desktop version to work on the platform.

I have been using Microsoft Azure Machine Learning Studio for almost five years.

It's a stable solution. Microsoft is very stable in general.

It's very scalable because it is using Microsoft cloud compute power.

We want to extend organization-wide, but currently, we are only working on a use case basis.

We have not required help from technical support, but Microsoft technical support comes with it when you subscribe.

Deployment of the tool is simple. Just one click on Microsoft. Once you have procured the license, you can just log in and use it. It's a ready-to-use tool.

When you deploy the solution after analytic development, it depends on the project but it can take anywhere from one month to six months.

Also, depending on the infrastructure, the initial deployment can take one week to a month.

In-house expertise.

The licensing cost is very cheap. It's less than $50 a month would costs for multiple users.

If you want to build a solution quickly without knowing any coding, it's pretty good to start with.

I will take a week to learn, from my experience. For anyone who is interested in trying it, they should start with the free version, which is free for up to 10 gigabytes of workspace.

Just log in and start developing and exploring the tool before onboarding.

I would rate Microsoft Azure Machine Learning a seven out of ten.

We primarily use this product for its price elasticity and the product mix on offer.

The solution is very easy to use, so far as our data scientists are concerned. 

There's an excellent self-developing capability that is provided that makes the product unique.

The solution is very stable. We haven't had any issues with its performance thus far.

We've found that, if you need to, you can scale the product.

The solution is very quick to implement.

We've found that the solution runs at a high cost. It's not cheap to utilize it.

Two additional items I would like to see added in future versions are software life cycle features and more security capabilities. There should be data access security, a role level security. Right now, they don't offer this.

I've only really been using the solution for the last few months. It really hasn't been too long at this point in time.

The solution is reliable. There are no bugs or glitches. We haven't experienced crashes or freezing. It's stable. It's very good in that sense.

If a company needs to scale the solution, they should have no problem doing so. I don't see any aspect of the solution that would stop a user from expanding it as needed.

Currently, we only have a handful of users. There are only about five to seven people on the product right now.

We do plan to continue to use the product and to increase usage in the future.

We've dealt with technical support in the past. We do, from time to time, have issues, which we work with the Microsoft team to resolve.

Overall, we've been satisfied with the level of support they have provided us.

We did not previously use a different product. This is the first type of solution that we've used.

The initial setup is quick and easy. It's not complex at all. There is no installation per se. It's simply that you plug into the cloud and start using it.

For deployment, you likely need a two or three-member team. You don't need a lot of people to get it up and running. Largely they are just managers, admins or engineers, or a combination of those three.

The solution is quite expensive. It's something the organization should work on improving.

We use this product on a pay-per-use basis, Therefore, there is no licensing fee. It's embedded in the cost of using the Studio.

We're just a Microsoft customer. We don't have a business relationship with Microsoft.

Currently, it is my understanding that we are using the latest version of the solution.

I'd recommend this product to other organizations.

Overall, I would rate the solution at an eight out of ten.

Developing and operationally implementing a powerful lead scoring model for a major Multiufamily developer and operator of apartment properties throughout major western states. The work included 3 years of data across over 60 properties with more than 500,000 leads and 3 million transactions.

Increased sales force productivity by permitting them to prioritize activity during peak leasing periods on those leads most likely to close

The ability to do the templating and be able to transfer it so that I can easily do multiple types of models and data mining is a valuable aspect of this solution. You only have to set up the flows, the templates, and the data once and then you can make modifications and test different segmentations throughout.

We were working across a number of internal departments as well as some outside departments and this solution made it extremely easy to communicate across functional area because it was all in flow chart and data form so that if somebody had an issue, like changing the data set or something like that, they could point right to it and we could get that handled and incorporated into the model. It's extremely efficient on the computer. We had to do a number of resets on the data in the model and to be able to turn things around and validate the model and the new set in two hours, was just incredible for me.

It was very robust. The ability to move the objects around so easily and then communicate is really its power. Then to be able to show it to the sales and senior management, in terms of what was employed and made it very easy to get my job done.

In terms of improvement, I'd like to have more ability to understand the detailed impact of the variables on the model and their interactions. Their algorithms are very powerful and they explain overall the net contribution of each of the variables to the solution. In terms of being able to say to people "If you did this, you'll get this much more improvement" Azure (at least my understanding of it) doesn't provide readily accessible tools to assess from a management perspective the impact of their changing a sinimized, the better.gle value - for instance in closing a lead, decreasing response time by 10%.

I recognize that the multivariate algorithms used from decision trees to neural nets do not readily provide the coefficients for each variable ala the older regression modeling approaches. My experience over my 50 years of developing and implementing predictive models has been that more than half the value of modeling lies in improving management's understanding of the process being modeled, often leading to major organization and operational structure changes. More ability to understand the variables impacting the end result being optimized would be very useful. 

I have worked extensively with this solution for the last three years. 

I haven't had any problems with stability. 

I didn't have any issues with the scale. we rapidly went from test to full implementation across all datasets.

I never had to use technical support.

I have used SPSS modeler (part of WATSON really) but because client was a Microsoft shop, I switched to Azure.

I found the setup to be very easy. I've been doing this type of work for 50 years so the modern terminology isn't always the same as what I grew up with. It took me a while to understand that, but the setups were very easy. As with anything, the hardest part is always getting the data together, but the outside consultants had built up a very, very good data warehouse. The ability to manipulate the data and create variables was very nice.

THIS IS THE ONLY MODELING APPROACH THAT EVER WORKED THE VERY TIME I RAN IT!!

Because client isa Microsoft shop, everything was Microsoft in terms of having solutions like Power BI and stuff like that. Azure is very useful and very inexpensive.

The major advice I give is that clients must get the user,somebody who understands the business issues, to be deeply involved with it and the data transformation. Most people don't. And that's true for data science applications. We don't just follow the data in a big pile and remodel, we advance the process that we're modeling. Consider what transformations of the data you need to make it workable and usable.

Remember, over half the initial value of modeling is the strategic understanding provided re the importance of different variables to the model and hence the organizaion's performance. Very often the modeling identifies opportunities for changing structures, decision rules, etc. even prior to the model's actual implementation technically.

I would rate it a nine out of ten.

Our organization employs people with diverse professional backgrounds. We have sociology, mathematics, and statistics backgrounds. We employ these people within our data science team. They require a certain amount of programming skills.

The good thing about Azure Machine Learning is they have a drag and drop feature. You can use Azure Machine Learning designer for all of your data science teams.

Any non-programmer can adopt it. All he needs is statistics and data analysis skills.                                                                                             

I used Azure Machine Learning in a free trial and I had a complete preview of the service. A problem that I encountered was that I had a model that I wanted to deploy and use on Azure Machine Learning, but there wasn't any option that that model can be used in the designer. I didn't find any option to upload my model, so that I can create my own block and use it in Azure Machine Learning designer.

I believe this is a problem because sometimes you have your model created on some other device and you just have a file that you think can be uploaded to Azure Machine Learning and can be tested through a simple drag and drop tool.

We have been using Azure for three months. We have been exploring it for different use cases. 

I haven't used it long enough to have found any bugs in our current system. If there were bugs I would definitely report it on their website.

We didn't have any problems with the setup. It was pretty straightforward.

It's an easy tool. They have a good level of resources and we are pretty low with resources as far as data science is concerned.

Azure Machine Learning offers an opportunity for those who haven't been introduced to Azure programming. You can use the data analytics and their statistics skills to build and deploy data science solutions that can be beneficial for society and for different organizations.

I would rate it a seven out of ten.

We used this solution for defining new predictive models, such as recommendation systems, but also price elasticity models for fraud detection, and the classification of customers.

We are not using this solution regularly. We are now using Azure Databricks.

The most valuable feature of this solution is the ability to use all of the cognitive services, prebuilt from Azure. You just have to drag and drop the services into your pipeline, and it can be applied through the pipeline. It's very helpful for data scientists. If you don't have any special knowledge in data science, just to know that you want to consume a service, that's all you need.

They have a tool for data gathering from some social networking sites such as Twitter and Facebook, which is great.

If you want to be able to deploy your tools outside of Microsoft Azure, this is not the best choice.

One of the problems that we had was that you could only execute the model inside the machine learning environment. Comparing this to Databricks, if you create a pipeline, it could be in a notebook and you have all the code and then you can export your notebook to some other tool directly, for example in Jupyter and Spark. If you change tools then you won't lose your assets.

I would like to see improvements to make this solution more user-friendly.

They need to have some tools, like Apache Airflow, for helping to build workflows.

Better tools are needed to bring the data from existing storage into the environment where they can play with it and start to analyze what they already have, on-site. This is what the majority of people would like to do.

A feature that would be useful is to have some standard data transportation functions. They have ADF, Azure Data Factory, but it's a little bit heavy to manipulate. If they could have something more user-friendly, like Apache Airflow, it would be very nice.

This is a stable solution, although we have had problems with JavaScript. When you have many JavaScripts running, sometimes you have something that freezes, but we didn't know whether it was based on our network, the configuration, or the tools. It is difficult to identify the precise cause.

In general, there are no major issues.

We never went into production because we switched to Azure Databricks. We did, however, try some performance testing and tried scaling some resources. The scalability of this solution is quite easy.

It is not difficult compared to some of the other tools that are available on Azure.

We have only five users including data engineers, data scientists, and one data DevOps engineer who was working with us on creating all of the DevOps pipelines for deploying all of our models.

I have been in touch with technical support many times. The client I work for is a first-year client for them and we received some very useful support. The showed great willingness to help and they provided a lot of support for free.

We also had meetings with some experts on their data side and we had some free consultancy days given by Microsoft. It is called FastTrack and it is only available for some kinds of clients.

We are completely satisfied with the technical support.

We did not use another solution prior to this one, but we now use Azure Databricks.

The initial setup of this solution is straightforward.

The client site that we were working at had a proxy, and we were having a lot of trouble managing the rules inside the proxy because the Machine Learning Studio was not showing on the screen, in the browser, as it should. There are a lot of JavaScripts and this is a heavy client. There is a lot of feature logic performed on the client-side, such as the drag-and-drop. We had a lot of problems.

Besides that, once we fixed our network problem, it was straightforward.

We implemented this solution on our own. The documents available on Microsoft Online made it quite easy.

When we started using this solution, our licensing fees were approximately €1,000 (approximately $1,100 USD) monthly, but it was fluctuating. When we got our first models and were ready for the user acceptance testing, our licensing fees were between €2,500 ($2,750 USD) and €3,000 ($3,300 USD) monthly. It was quite limited.

We expected the rate to be higher than this, at perhaps €10,000 (approximately $11,000 USD) per month, but it wasn't the case.

Microsoft has increased the usability and the features since we first implemented this solution.

If I had to start this process over again, I would involve Microsoft earlier because they were great for providing support, as well as guidance on the architecture and what kind of stuff you can do with the tool, and what you should do with it. This was very helpful to orient the team to the right documentation and tutorials.

The second thing I would do is to start working with DevOps activity as soon as you can. We found ourselves redoing the same things many times, instead of having a DevOps pipeline to implement the stuff that we already stabilized, for example, and then not losing time.

The third thing is involving an integrator to help put together the big picture.

I would rate this solution a seven out of ten.

Microsoft Azure Studio allows you to connect to multiple databases and do analysis.

The notebook feature allows you to write inquiries and create dashboards. These dashboards can integrate with multiple databases, such as Excel, HANA, or SQL Server. Connecting to various databases lets you link multiple dashboards or perform data analytics simultaneously. Additionally, the notebook feature supports version control, enabling you to commit code into a repository.

Performance is very poor.

I have been using Microsoft Azure Machine Learning Studio for the past year.

I worked with PowerBI.

The initial setup is straightforward. It is a .exe file that can be installed on your system. It is easily downloadable and open source solution. We can now easily download it from the Microsoft site and use it.

If performance is improved, it can provide a good return on investment because people often make mistakes when they are not familiar with their dataset. Microsoft Azure Machine Learning Studio can pull information from the database and summarize it effectively.

If you want to take design lessons, Azure Machine Learning Studio is the best tool.

The product can simplify some AI-driven projects because it currently has extensive database connectivity. For example, it can easily connect to various databases. However, the support for some other databases is presently limited and can be improved.

It pulls information from the database. Its good analytics capability makes integrations very simple.

Overall, I rate the solution an eight out of ten.

We develop products on the solution. It also provides fraud detection. We use it mainly for IoT to save on the electricity bill for heating in the warehouses.

The product supports open-source tools. The integration with data services is an important feature. We use it in case the data is already available.

The price could be improved.

The tool is mature.

The tool is scalable. We have four users in our organization. We have plans to increase the usage in the future.

Whatever we develop, we deploy from the GUI. The tool can be easily deployed.

We do the deployment in-house.

We have an enterprise contract.

We used Google in the past.

Overall, I rate the solution a seven out of ten.