H2O.ai vs. IBM SPSS Modeler vs. TensorFlow

Most suited if in little time you wanted to build and train a model. Then, H2O makes life very simple. It has support with R, Python and Java, so no programming dependency is required to use it. It's very simple to use. If you want to modify or tweak your ML algorithm then H2O is not suitable. You can't develop a model from scratch.

Incentivized

Fast NLP analytics are very easy in SPSS Modeler because there is a built-in interface for classifying concepts and themes and several pre-built models to match the incoming text source. The visualizations all match and help present NLP information without substantial coding, typically required for word clouds and such. SPSS Modeler is good at attaining results faster in general, and the visual nature of the code makes a good tool to have in the data science team's repository. For younger data scientists, and those just interested, it is a good tool to allow for exploring data science techniques.

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

• Excellent analytical and prediction tool
• In the beginning, usage of H20 Flow in Web UI enables quick development and sharing of the analytical model
• Readily available algorithms, easy to use in your analytical projects
• Faster than Python scikit learn (in machine learning supervised learning area)
• It can be accessed (run) from Python, not only JAVA etc.
• Well documented and suitable for fast training or self studying
• In the beginning, one can use the clickable Flow interface (WEB UI) and later move to a Python console. There is then no need to click in H20 Flow
• It can be used as open source

Incentivized

• Combine text and data
• Provide facilities for all phases of the data mining process.
• Use a node and stream paradigm to easily and quickly create models.

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

• Better documentation
• Improve the Visual presentations including charting etc

Incentivized

• Has very old style graphs, with lots of limitations.
• Some advanced statistical functions cannot be done through the menu.
• The data connectivity is not that extensive.
• It's an expensive tool.

Incentivized

• 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.

The ability to do predictive modeling, text analytics for both structured & unstructured data, decision management, optimization, and support for various data sources

Incentivized

Support of multiple components and ease of development.

Incentivized

The overall experience I have with H2O is really awesome, even with its cost effectiveness.

Incentivized

The online support board is helpful and the free add ons are incredibly appreciated.

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

Use of cloud for better execution power is recommended.

Incentivized

Both are open source (though H2O only up to some level). Both comprise of deep learning, but H2O is not focused directly on deep learning, while Tensor Flow has a "laser" focus on deep learning. H2O is also more focused on scalability. H2O should be looked at not as a competitor but rather a complementary tool. The use case is usually not only about the algorithms, but also about the data model and data logistics and accessibility. H2O is more accessible due to its UI. Also, both can be accessed from Python. The community around TensorFlow seems larger than that of H2O.

Incentivized

When it comes to investigation and descriptive we have found SPSS Statistics to be the tool of choice, but when it comes to projects with large and several datasets SPSS Modeler has been picked from our customers.

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

• Positive impact: saving in infrastructure expenses - compared to other bulky tools this costs a fraction
• Positive impact: ability to get quick fixes from H2O when problems arise - compared to waiting for several months/years for new releases from other vendors
• Positive impact: Access to H2O core team and able to get features that are needed for our business quickly added to the core H2O product

Incentivized

• Positive - Ease of decision making and reduction in product life cycle time.
• Positive - Gives entirely new perspective with the help of right team. Helps expanding the portfolio.
• Negative - Needs to have good understanding about mathematical modelling, of which talent is rare and expensive. Hence, increase the costs for R&D and manpower.

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