Tableau Desktop vs. TensorFlow

The best scenario is definitely to collect data from several sources and create dedicated dashboards for specific recipients. However, I miss the possibility of explaining these reports in more detail. Sometimes, we order a report, and after half a year, we don't remember the meaning of some data (I know it's our fault as an organization, but the tool could force better practices).

Incentivized

TensorFlow is great for most deep learning purposes. This is especially true in two domains: 1. Computer vision: image classification, object detection and image generation via generative adversarial networks 2. Natural language processing: text classification and generation. The good community support often means that a lot of off-the-shelf models can be used to prove a concept or test an idea quickly. That, and Google's promotion of Colab means that ideas can be shared quite freely. Training, visualizing and debugging models is very easy in TensorFlow, compared to other platforms (especially the good old Caffe days). In terms of productionizing, it's a bit of a mixed bag. In our case, most of our feature building is performed via Apache Spark. This means having to convert Parquet (columnar optimized) files to a TensorFlow friendly format i.e., protobufs. The lack of good JVM bindings mean that our projects end up being a mix of Python and Scala. This makes it hard to reuse some of the tooling and support we wrote in Scala. This is where MXNet shines better (though its Scala API could do with more work).

Incentivized

• An excellent tool for data visualization, it presents information in an appealing visual format—an exceptional platform for storing and analyzing data in any size organization.
• Through interactive parameters, it enables real-time interaction with the user and is easy to learn and get support from the community.

Incentivized

• A vast library of functions for all kinds of tasks - Text, Images, Tabular, Video etc.
• Amazing community helps developers obtain knowledge faster and get unblocked in this active development space.
• Integration of high-level libraries like Keras and Estimators make it really simple for a beginner to get started with neural network based models.

Incentivized

• Pricing should be more user-friendly and usage-driven
• Making edits to the production reports is fairly tough and has a vast scope of additional capabilities
• Tableau Desktop should be able to differentiate itself from the Tableau server else there is no major meaning of two different products being offered

• RNNs are still a bit lacking, compared to Theano.
• Cannot handle sequence inputs
• Theano is perhaps a bit faster and eats up less memory than TensorFlow on a given GPU, perhaps due to element-wise ops. Tensorflow wins for multi-GPU and “compilation” time.

Our use of Tableau Desktop is still fairly low, and will continue over time. The only real concern is around cost of the licenses, and I have mentioned this to Tableau and fully expect the development of more sensible models for our industry. This will remove any impediment to expansion of our use.

Incentivized

Tableau Desktop has proven to be a lifesaver in many situations. Once we've completed the initial setup, it's simple to use. It has all of the features we need to quickly and efficiently synthesize our data. Tableau Desktop has advanced capabilities to improve our company's data structure and enable self-service for our employees.

Support of multiple components and ease of development.

Incentivized

When used as a stand-alone tool, Tableau Desktop has unlimited uptime, which is always nice. When used in conjunction with Tableau Server, this tool has as much uptime as your server admins are willing to give it. All in all, I've never had an issue with Tableau's availability.

Tableau Desktop's performance is solid. You can really dig into a large dataset in the form of a spreadsheet, and it exhibits similarly good performance when accessing a moderately sized Oracle database. I noticed that with Tableau Desktop 9.3, the performance using a spreadsheet started to slow around 75K rows by about 60 columns. This was easily remedied by creating an extract and pushing it to Tableau Server, where performance went to lightning fast

Incentivized

Tableau support has been extremely responsive and willing to help with all of our requests. They have assisted with creating advanced analysis and many different types of custom icons, data formatting, formulas, and actions embedded into graphs. Tableau offers a weekly presentation of features and assists with internal company projects.

Incentivized

Community support for TensorFlow is great. There's a huge community that truly loves the platform and there are many examples of development in TensorFlow. Often, when a new good technique is published, there will be a TensorFlow implementation not long after. This makes it quick to ally the latest techniques from academia straight to production-grade systems. Tooling around TensorFlow is also good. TensorBoard has been such a useful tool, I can't imagine how hard it would be to debug a deep neural network gone wrong without TensorBoard.

Incentivized

It is admittedly hard to train a group of people with disparate levels of ability coming in, but the software is so easy to use that this is not a huge problem; anyone who can follow simple instructions can catch up pretty quickly.

I think the training was good overall, but it was maybe stating the obvious things that a tech savvy young engineer would be able to pick up themselves too. However, the example work books were good and Tableau web community has helped me with many problems

Incentivized

Again, training is the key and the company provides a lot of example videos that will help users discover use cases that will greatly assist their creation of original visualizations. As with any new software tool, productivity will decline for a period. In the case of Tableau, the decline period is short and the later gains are well worth it.

Incentivized

Use of cloud for better execution power is recommended.

Incentivized

I have used Power BI as well, the pricing is better, and also training costs or certifications are not that high. Since there is python integration in Power BI where I can use data cleaning and visualizing libraries and also some machine learning models. I can import my python scripts and create a visualization on processed data.

Incentivized

Keras is built on top of TensorFlow, but it is much simpler to use and more Python style friendly, so if you don't want to focus on too many details or control and not focus on some advanced features, Keras is one of the best options, but as far as if you want to dig into more, for sure TensorFlow is the right choice

Incentivized

Tableau Desktop's scaleability is really limited to the scale of your back-end data systems. If you want to pull down an extract and work quickly in-memory, in my application it scaled to a few tens of millions of rows using the in-memory engine. But it's really only limited by your back-end data store if you have or are willing to invest in an optimized SQL store or purpose-built query engine like Veritca or Netezza or something similar.

• Tableau was acquired years ago, and has provided good value with the content created.
• Ongoing maintenance costs for the platform, both to maintain desktop and server licensing has made the continuing value questionable when compared to other offerings in the marketplace.
• Users have largely been satisfied with the content, but not with the overall performance. This is due to a combination of factors including the performance of the Tableau engines as well as development deficiencies.

Incentivized

• Learning is s bit difficult takes lot of time.
• Developing or implementing the whole neural network is time consuming with this, as you have to write everything.
• Once you have learned this, it make your job very easy of getting the good result.

Incentivized