Databricks Lakehouse Platform vs TensorFlow

Likelihood to Recommend

Databricks has helped my teams write PySpark and Spark SQL jobs and test them out before formally integrating them in Spark jobs. Through Databricks we can create parquet and JSON output files. Datamodelers and scientists who are not very good with coding can get good insight into the data using the notebooks that can be developed by the engineers.

Read full review

Verified User

Team Lead in Engineering

Financial Services Company, 10,001+ employees

View all 7 answers on this topic

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 networks2. 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).

Read full review

Verified User

Engineer in Other

Computer Software Company, 201-500 employees

View all 12 answers on this topic

Pros

It supports all data science programming languages like Python and R
it takes very few minutes to deploy models into production
it has tools that ensures collaborations between developers

Read full review

Kofi Joshua

Machine Learning And Deep Learning Mentee

Runmila AI InstituteEducation Management, 201-500 employees

View all 7 answers on this topic

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.

Read full review

Nitin Pasumarthy

Web & Network Performance

LinkedInInternet, 5001-10,000 employees

View all 12 answers on this topic

Cons

Better Localized Testing
When they were primarily OSS Spark; it was easier to test/manage releases versus the newer DB Runtime. Wish there was more configuration in Runtime less pick a version.
Graphing Support went non-existent; when it was one of their compelling general engine.

Read full review

Verified User

Director in Information Technology

Financial Services Company, 201-500 employees

View all 7 answers on this topic

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.

Read full review

Nisha murthy

Senior product Manager @ CVS (Aetna)

Infosys ConsultingHospital & Health Care, 10,001+ employees

View all 12 answers on this topic

Usability

Databricks Lakehouse Platform 9.0

Based on 1 answer

This has been very useful in my organization for shared notebooks, integrated data pipeline automation and data sources integrations. Integration with AWS is seamless. Non tech users can easily learn how to use Databricks. You can have your company LDAP connect to it for login based access controls to some extent

Read full review

Verified User

Team Lead in Engineering

Financial Services Company, 10,001+ employees

View all 1 answers on this topic

TensorFlow 9.0

Based on 1 answer

Support of multiple components and ease of development.

Read full review

Anupam Mittal

Founder

BrontomindsInformation Technology and Services, 1-10 employees

View all 1 answers on this topic

Support Rating

No score

No answers yet

No answers on this topic

TensorFlow 9.1

Based on 3 answers

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.

Read full review

Verified User

Engineer in Other

Computer Software Company, 201-500 employees

View all 3 answers on this topic

Implementation Rating

No score

No answers yet

No answers on this topic

TensorFlow 8.0

Based on 1 answer

Use of cloud for better execution power is recommended.

Read full review

Anupam Mittal

Founder

BrontomindsInformation Technology and Services, 1-10 employees

View all 1 answers on this topic

Alternatives Considered

I also use Microsoft Azure Machine Learning in parallel with Databricks. They use different file formats which teach me to be flexible and able to write different programs. They are equally useful to me and I would like to master both platforms for any future usage. I do prefer Databricks because it could be free if I decided to go with the Databricks Community Edition only.

Read full review

Ann Le

Freelance Translator

ZOO Digital Group plcEntertainment, 501-1000 employees

View all 7 answers on this topic

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

Read full review

Verified User

Strategist in Information Technology

Package/Freight Delivery Company, 10,001+ employees

View all 12 answers on this topic

Return on Investment

Rapid growth of analytics within our company.
Cost model aligns with usage allowing us to make a reasonable initial investment and scale the cost as we realize the value.
Platform is easy to learn and Databricks provides excellent support and training.
Platform does not require a large DevOPs investment

Read full review

Verified User

Strategist in Engineering

Computer Hardware Company, 10,001+ employees

View all 7 answers on this topic

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.

Read full review

Shambhavi Jha

Software Engineer

Tata Consultancy ServicesInformation Technology and Services, 51-200 employees

View all 12 answers on this topic

Pricing Details

General

Free Trial

—

Free/Freemium Version

—

Premium Consulting/Integration Services

—

Entry-level set up fee?

No

Databricks Lakehouse Platform Editions & Modules

Edition

Standard	$0.07¹
Premium	$0.10¹
Enterprise	$0.13¹

Per DBU

Additional Pricing Details

—

General

Free Trial

—

Free/Freemium Version

—

Premium Consulting/Integration Services

—

Entry-level set up fee?

No

TensorFlow Editions & Modules

—

Additional Pricing Details

—