Coremetrics / IBM Digital Analytics (discontinued) vs. TensorFlow

IBM analytics has continued to improve upon the days of being the original core metrics. After using the updated version for quite some time, it has been great at providing the needed analytics to measure ROI and goal performance for our quarterly KPI's. It has resulted in a great increase in web engagements although we are a midsize company, smaller outfits may not need such an expensive option.

Incentivized

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

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

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

• IBM CXA comprises an acquisition called Tealeaf. This tool has deep heritage and this is evident in its present-day capabilities.
• The Universal Behaviour Exchange or UBX puts the concept of personalisation at the forefront. The ability to combine physical (analog) and digital transactions to create the complete picture of a customer journey, is a stand out benefit.
• The solution does not have to involve the purchase of software. IBM CXA can be sold as a service bundled with analytics as a service. This not only lowers the cost of ownership, it gets around one of the principal issues. Strong staff with design and analytical capability to drive the solution and deliver tangible benefits.
• The seamless integration of Watson AI services to help with the heavy lifiting. Watson reinforces the analytical focus this solution has and can learn to recognise situations specific to a company.

Incentivized

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

• 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

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

• The user interface is in Flash, which can be very frustrating and slow at times. Apparently, this is to be transitioned in a future release.
• Can only segment the last 93 days of data. Any historical segmentation beyond the 93 days must be run in Explore (which is credit based, and has its own limitations with the number of credits per month, based on the initial contract with IBM).
• Reports can only display 93 days of data at a given time for custom date ranges. There are pre-programmed date ranges setup with IBM during implementation (last week, last month, last quarter etc.), but are not flexible enough to answer more specific questions.
• Certain reports cannot have segments applied, making answering some simple questions a bit more tricky. For example, I can create a segment around mobile devices and apply it to the marketing channels report, but I can't create a marketing channel segment and apply it to the mobile reports.
• Built in API calls allows for nice report design and automation.

• 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.

IBM Digital Analytics is a great solution for our clients and I believe they offer the best solution for the retail space. We have access to IBM support via email or live chat and they can answer many of the reporting questions that come up. IBM is receptive to our feedback of the product so I am confident they will continue making improvements

Very easy to implement and use.

Support of multiple components and ease of development.

Incentivized

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

Never had any issues

As reports are templated, the system is pretty quick. Sometimes you have to wait a bit for a report to render. Or you might have to re-load the page. But there is no real issue here and the system is on par with other similar systems.

Overall, the level of support is very good and I would say it is a strong asset of the solution. However, you can sometimes feel that there is a difference of level among the support team.

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

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

Online training is really great. One of the best assets that they have. Lots of great videos, pop quizzes at the end of each module. Fantastic. Other tools have similar features, but not as good.

See previous comment: reading and understanding the encyclopedic implementation guide is a must.

Use of cloud for better execution power is recommended.

Incentivized

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

Much of the work we did in IBM Digital Analytics could have been answered through Google Analytics, a much simpler, agile and FREE solution set. Not mention, given the vast number of Google Analytics USERS, free and actionable support is simply a click away ... this compared to IBM Digital Analytics fractured and often absent support service.

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

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

This solution can support large amount of data and transaction. The way that user management features are built, it shows it is meant for large organizations.

• We spend too much time trying to work around bugs on the new UI.
• We spend too much time trying to figure out how to make certain segments work because support and the knowledge center are lackluster.
• Our sales rep is very unresponsive and leaves us searching for a lot of answers on our own, including what other products we may benefit from that IBM offers.

Incentivized

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info

• 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

Independently Invited by TrustRadius

A de minimis incentive was given to thank the reviewer for their time. The incentive was not used to bias or drive a particular response, nor was the incentive contingent on a positive endorsement. More Info