TensorFlow
Stable Version 1.0.0
Published on 15 Mar by 
Created on 15 Mar
Details
Connector and sample on how to use Tensorflow to integrate machine learning models into OutSystems applications. Sample uses a pre-trained inception model for image recognition.
Read More

What is TensorFlow?

TensorFlow is an open source library for numerical computation and large-scale machine learning. TensorFlow bundles together a slew of machine learning and deep learning (aka neural networking) models and algorithms and makes them useful by way of a common metaphor. TensorFlow can train and run deep neural networks for handwritten digit classification, image recognition, word embeddings, recurrent neural networks, sequence-to-sequence models for machine translation, natural language processing, and PDE (partial differential equation) based simulations. Best of all, TensorFlow supports production prediction at scale, with the same models used for training.


TensorFlow allows developers to create dataflow graphs—structures that describe how data moves through a graph, or a series of processing nodes. Each node in the graph represents a mathematical operation, and each connection or edge between nodes is a multidimensional data array or tensor.


Inception model?

The inception model is a deep neural network which has already been trained for classifying images into thousands of different categories. This means that by using this model, we can start classifying images immediately without further training.



How does this module work?


- DownloadModel


The DownloadModel action is used to download and apply the inception model from Google (https://storage.googleapis.com/download.tensorflow.org/models/inception5h.zip). This is a pre-trained model equivalent do Inception V1.


- Image Recognition


The ImageRecognition receives a record list of Images (structure, binary only) and for each one of them creates a new Tensor, which is then processed and calculated against the pre-trained values.


Please refer to the demo module within this application for additional information.

Reviews (0)
Category
Connectors, Machine Learning
Support Options
This component is not supported by OutSystems. You may use the discussion forums to leave suggestions or obtain best-effort support from the community, including from André Vieira who created this component.
Dependencies
TensorFlow has no dependencies.
Requirements
Platform
10.0.0.402
Database
All
Stack
.NET
Component Consumers
TensorFlow has no consumers.
Weekly Downloads 
Related Components
Google Drive Connector
OutSystems R&D
Google Drive Connector allows your OutSystems Applications to use the Google Drive REST API to interact with your files.
1006
Google Maps
Labs
Google Maps is a Google Maps Javascript V3 API component, providing Web Blocks and Actions to add interactive maps, markers and directions, along with event driven functionality.
10689
JIRA Connector
OutSystems R&D
JIRA is a issue tracking product, developed by Atlassian. It provides bug tracking, issue tracking, and project management functions. This connector gives you the power to integrate JIRA with your apps to automate your work.
597
More from André Vieira
SortRecordList
SortRecordList Team
This extension is a generic record list sorter. It sorts any record list in ascending or descending order by any of its attributes (string, integer, decimal, etc). It can also return a distinct list by comparing any of its attributes. 
4124
ZXing Services
OutSystems Technical Communication
This component provides services for barcode encoding and decoding in different formats: CODE 39, CODE 128, QR, EAN13, EAN8 throughZXing library (http://code.google.com/p/zxing/)
2696
App Icon Forge
André Vieira
Forge your OutSystems Applications icons in an instant.
134