Owl - An OCaml Numerical Library

Owl - An OCaml Numerical Library

Owl is an OCaml library for scientific computing and machine learning using functional programming

Artificial Intelligence

Description

Owl is an OCaml library for scientific computing. It enables academic researchers to fast prototype machine learning algorithms and construct deep neural networks with very concise code. It also facilitates industry programmers to develop robust analytical applications using functional language at a large scale.

Currently, Owl supports N-dimensional arrays, both dense and sparse matrix operations, linear algebra, regressions, fast Fourier transforms, and many advanced mathematical and statistical functions (such as Markov chain Monte Carlo methods). Recently, Owl has also implemented algorithmic differentiation, on top of which we have been building a powerful deep neural network module. With its advanced underlying distributed computation engine, Owl is able to support distributed computing to take advantage of the computation power in a computer cluster.

Owl is under active development. It aims to provide a comprehensive set of advanced numerical functions to enable fast development of robust and efficient analytical applications.

Github: https://github.com/ryanrhymes/owl

Gallery

Links

Owl's Github Repository

Aaeaaqaaaaaaaanfaaaajdy4yjc2nmewltg0mtatndgznc1iyte5lwnkyzvjyzqwywfkoq

Fabrizio L. added photos to project BTM - Big trees monitor

Medium aa73f57b 2473 4604 8120 25d40bcc0d3c

BTM - Big trees monitor

The sensors for big trees monitoring are positioned and installed following a safety analysis to determine the danger and the area of fall in order to be able to provide well in advance the fall of the tree and therefore, the possible commissioning in security of the surrounding area.

The initial safety analyzes allow us to estimate the height to which the sensors can be positioned, which will allow the monitoring of vibrations and movements even of the minimum amount of the shaft.

Medium aaeaaqaaaaaaaanfaaaajdy4yjc2nmewltg0mtatndgznc1iyte5lwnkyzvjyzqwywfkoq

Fabrizio L. created project BTM - Big trees monitor

Medium aa73f57b 2473 4604 8120 25d40bcc0d3c

BTM - Big trees monitor

The sensors for big trees monitoring are positioned and installed following a safety analysis to determine the danger and the area of fall in order to be able to provide well in advance the fall of the tree and therefore, the possible commissioning in security of the surrounding area.

The initial safety analyzes allow us to estimate the height to which the sensors can be positioned, which will allow the monitoring of vibrations and movements even of the minimum amount of the shaft.

Medium pic

Abhishek N. created project Autonomous drones for monitoring tea plantation sites

Medium 7c37042d bb7f 49de ac5f 4eda2c15c8a4

Autonomous drones for monitoring tea plantation sites

Why this project India is one of the largest tea producers of the world we will be using smart technique using AI to determine the places for future tea plantations. This project will be very beneficial for easy access of information and finding a way to get places for tea plantation areas faster.

Abstract: Drones or unmanned aerial vehicles are being used widely for a wide scale of application. This abstract involves the use of autonomous drones for monitoring health of tea plants. Autonomous drones operating in a specific grid will be taking infrared scans of the plantation site, and based on the data of the infrared camera, the health can be determined. Design: The design will be of that a typical quadrotor. The drone will operate autonomously in an unknown environment. What need to be taken care of are the waypoints that the drone needs to navigate. The flight controller is based on the Atmega series of microcontroller with integrated sensors for orientation and position. The sensing element is in infrared camera which will provide us the raw input for further processing. Since the drone will be designed to function fully autonomously, therefore some key areas are being focussed on as developing the flight control algorithm which includes navigation and image sensing and processing for monitoring plant health. Some conflicts that may be encountered • Control algorithm for stable flight • Fully autonomous with auto take-off and landing • Waypoint or grid based navigation • Processing large chunks of data obtained from infrared cameras

Neural Networks The quadrotor will monitor lots of data with the camera and this data needs to be trained the captured image will be checked with lot of parameters for determining the health of the tea plantation site and create DNN as a trained model. Our motive of using Movidius Neural Stick is to determine the scans of the area and find out if the area is good for cultivation of tea leaves or not. Movidius Neural Stick We intend to use Tensorflow on the trained model using the capability of neural computer stick and modifying the inception v3 model of it.The trained model we get is a deep neural network of infrared images and comparing it with v3 we will try to find the feasibility of the plantation sites based on parameters.

Inference part

We will use Slim inception for the TensorFlow process our main criterion will also be making the infrared images v3 compliant so that we can easily use the Slim inception process in it.

Prototyping First of all we will train the model Then we will do profiling,tuning and compiling the model For the prototyping stage we will imply the following process as shown in the diagram below.

We will attach the quadrotor with RasberryPi3 and a infrared camera module attached to it. Then we will attach the neural compute stick to the rasberryPi3 usb port to get the prototyping process going with TensorFlow model.

Requirements Quadrotor(Already have) RasberryPi3 Movidius Neural Stick Infrared Camera module

Stabilization of Quadrotor We have already implemented Reinforcement learning for the takeoff and stability and have used python as the program language similarly for neural network too we will use Python and expose the APIs too with the help of Python itself. We don’t have Movidius neural stick but we prefer TensorFlow as our choice with v3 inception but if adaptability demand we will scale it with Caffe.

Medium 0 1c3b1 choltmzdn8yqmboqw3oqufgy9uyqmbv6qhitnmgmh8jqr6t9hh2ql o0m3jcmwyhw8fkwauvjuskxur6euckw7ususjkxkrfxtw5yilubavztokplkgalx srykf8wbzipftd

Sujata T. (Intel) added a comment on project Shared Sensor Network for Vehicles and Smart Roads

Medium 068cd49d bfc1 4abd bb1b 758120777361

Shared Sensor Network for Vehicles and Smart Roads

Idea

Create a large network with mobile sensor nodes installed inside vehicles connected to each other in order to detect, analyze and send data automatically and anonymously in data collection centers through check-points placed on smart roads. The aim is to have as much data as possible on air quality, fine dusts, km driven by vehicles, and on which roads mainly; Data collection centers will be checkpoints on smart roads. Smart roads connected with 5G technology allow vehicles to connect and send data, this idea will also allow to report data about vehicles that generally do not hit external roads (highways for example) and, therefore, remaining more in the city accumulate interesting amounts of routes citizen data.

Description

Continuous technological advances in miniaturization and in the construction of low energy consumption circuits and the continuous progress in the field of information coupled with the high level of efficiency achieved by radio communication devices, increasingly affirm a new technological perspective: Sensor networks (WSN - Wireless sensor network). This technology arise from the fact that electronic devices are becoming smaller and more complex, and the tendency is to distribute intelligence into objects with relatively lower computing power but strongly related each other, instead of bring it together to a single expensive unit, cumbersome and hardly manageable; The birth of WSN is due to a basic idea which involves the use of a large number of nodes, allowing to perform reliefs and elaborations with greater precision and frequency than the single sensor case of use, capable of providing better performances at detriment of higher costs. All those premise is the logic from which the project starts, distributing analysis and reliefs among multiple mobile nodes (related vehicles) that converge acquired data into defined checkpoints (points on the smart roads) connected to data collections center. Mobile nodes consist essentially of:

  • Processor (CPU): usually represented by a microcontroller that performs all management functions, including translation of transducers electrical signals, actuator management and communication control.

  • Memory: Represents a volatile memory block that acts as an aid to execution.

  • Transducers: can be more than one, transform some physical dimension (temperature, gps perpendicular paths, air quality, fine dust pad etc etc) into an electrical signal.

  • Actuators: they can be more than one and of a different kind (Leds, acoustic sonar etc..).

  • Communication Unit: Allows inter-node communication.

Network description

One of the primary features of this network type is the management of logical addressing and routing of packets. In a WSN, addressing can follow a particular technique, so the network is partitioned into node clusters and for each cluster it is assigned a unique address. In every cluster there is a coordinator that collects data from nodes and directs them to the collection center or possibly to the cluster group coordinator to which it belongs. In this way, the address space is safeguarded while retaining the possibility of referring to certain portions of the network. A typical application example is showed in the picture (down-below). Here, in fact, there is a practical application where clusters are grouped together in a large network. Red-circled knots act as coordinators for top-level clusters, while blue circled nodes act as coordinators for second-level clusters (the largest external ones) until they reach the data collection center. As for the routing algorithms, it was chosen a hybrid typology, that is an algorithm that allows to interface, create nodes whenever encounters a mobile node and to communicate and understand if that node will reach first the data collection center.

Medium 0 1c3b1 choltmzdn8yqmboqw3oqufgy9uyqmbv6qhitnmgmh8jqr6t9hh2ql o0m3jcmwyhw8fkwauvjuskxur6euckw7ususjkxkrfxtw5yilubavztokplkgalx srykf8wbzipftd

Sujata T. (Intel) added a comment on project Containerizing Deep Learning Workloads on Xeon Phi Cluster for AI Web Applications

Medium bae1ed0d 3a3a 40d6 8fba b5327e569eb3

Containerizing Deep Learning Workloads on Xeon Phi Cluster for AI Web Applications

Today's Web applications are data intensive and demand environments like tensorflow to execute the workloads. Hence, Containers are best suited to provide the framework and compute resources like CPU and memory for each workload. It decouples the app environment from the running machine/host and encapsulates all dependencies in a single portable unit. Nomad is a state of the art tool for scheduling Docker Containers. Test Model Workload is generated from Model Zoo for each framework.

See More

No users to show at the moment.

Bigger 0 17h0cidt6oijn3mpfqjj9z2q5bgjnt8sqljyqgatbbcgns v5djg wsn9fmjlsmpcpjy6f2a5gi4nkizixlkh4ut giznktmixljvik bbtlnfiri8tpx mfca
  • Projects 0
  • Followers 1

Suwon HAM

39 Rue Volta, 75003 Paris, France

Bigger 0 i7n9aslwif7qdm2alpm4cuxbqlgndouaxuhuc2uwmrpzo0oxqxcvvqslrge
  • Projects 0
  • Followers 3

Pablo Esteban Camacho

I am an Argentinean Master in Software Engineering, Specialist in Embedded Systems, and C / C++ Senior Software Development Engineer.

Guadalajara, Roma Nte., 06700 Ciudad de México, CDMX, Mexico

Bigger profilbild
  • Projects 1
  • Followers 1

Stephan Volkeri

Leonardo-Campus 3, 48149 Münster, Germany

Bigger confidancejp
  • Projects 0
  • Followers 0

Jalpesh Panchal

Kalptaru Society, Umiya Nagar Rd, Umiya Nagar, Anand Nagar, Ranip, Ahmedabad, Gujarat 382480, India

Bigger capture
  • Projects 0
  • Followers 0

Duy Vu Ngo

I am Electrical Engineering in UQ come from Vietnam

Brisbane QLD, Australia

Bigger color line sketch of ambika light background
  • Projects 0
  • Followers 1

Neo AI

Been in IT inudstry for more than 15 years. Held different positions, currently exploring AI and entrepreneurship.

Victorian Avenue Block II, 6th D Main Road, Eshwara Layout, Indiranagar, Bengaluru, Karnataka 560008, India

See More