Eran Feit

Eran Feit

1670068102

What actually sees a deep neural network model ?

Hi,

 

How to visualize CNN Deep neural network model ? 

What is actually sees during the training ? 

What are the chosen filters , and what is the outcome of each neuron .

This is part no. 5 of Tensorflow tutorial for classify monkey’s species images using CNN and transfer learning 

In this part we will focus of showing the outcome of the layers.

Very interesting !!

I also shared the Python code in the video description.

 

Oreily  come up with this book . the best book for learning Deep learning based on Tensorflow-Keras. This is the link : https://amzn.to/3STWZ2N

 

You can find the link for the video tutorial here : https://youtu.be/yg4Gs5_pebY

 

Enjoy

Eran

#Python #Cnn #TensorFlow #Deeplearning #AI

What actually sees a deep neural network model ?
emily joe

emily joe

1668667804

7 Reasons why you should learn Python

Python is a multi-purpose, learner-friendly programming language that is a great fit for Data scientists, analysts, and developers to build an array of web applications, and so forth. Learn the top 7 reasons to learn Python. bit.ly/3EDiYac 

#Python #programming #datascience #programming-languages #PowerBI  #numpy #pandas 

7 Reasons why you should learn Python
Eran Feit

Eran Feit

1667508511

How to classify monkeys images using convolutional neural network , Ke

Hi,

This is part no. 4 of Tensorflow tutorial for classify monkey’s species images using several methods.

In this part we will focus of classify the monkeys images using Transfer learning .
So basically , we will take the pre-trained VGG16 model , get rid of the last layer , and replace and retrain it to our images and required classes. 

We will use the freeze and use the current weights, customize it to our needs, and improve the result of our CNN model 

I also shared the Python code in the video description.

Oreily  come up with this book . the best book for learning Deep learning based on Tensorflow-Keras. This is the link : https://amzn.to/3STWZ2N

You can find the link for the video tutorial here : https://youtu.be/p9l9AgiVsqI

Enjoy

Eran

 

#Python #Cnn #TensorFlow #Deeplearning

How to classify monkeys images using convolutional neural network , Ke
Eran Feit

Eran Feit

1664879908

How to classify monkeys images using convolutional neural network , Ke

 

Hi,

 

This is the third part of four parts Tensorflow tutorial that enables you to classify monkey’s species images using several methods.

 

So , how can we decide how many layers should we define ? how many filters in each convolutional layer ?

Should we use Dropout layer ? and what should be its value ?

Which learning rate value is better ? and more similar questions.

 

Keras tuner is the solution for looking those parameters , which called hyper parameters.

Now we will dive into the keras tuner. We will take the model from our previous lesson , and improve it. 

In the next, forth video, will you a learn how to use transfer learning to classify the images in a different way.

 

 

I recommend This graphics card: NVIDIA GeForce RTX 3060 Ti . I am using it to train my Tensorflow models. Get perfect results and performance:   https://amzn.to/3mTa7HX

 

This is the link for part 3: https://bit.ly/3BLpJEu

 

I shared the Python code in the video description.

 

 

Enjoy

Eran

 

#Python #Cnn #TensorFlow

How to classify monkeys images using convolutional neural network , Ke
Riyad Amin

Riyad Amin

1664267877

Change Button Background Color During Mouse Click Python Tkinter

In this tutoria we will learn how to change background color of Tkinter button during mouse Click . You can change the button’s background color, while the button is pressed using mouse, using activebackground property of the button.

Now let's go and examples to understand how to change background color of Tkinter button during mouse Click.

Example 1: Change Button Background Color during Mouse Click

In this example, we will change the color of button to red while it is in pressed state.

from tkinter import *   

tkWindow = Tk()  
tkWindow.geometry('400x150')  
tkWindow.title('PythonExamples.org - Tkinter Example')
  
button = Button(tkWindow, text = 'Submit', bg='#ffffff', activebackground='#00ff00')  
button.pack()  
  
tkWindow.mainloop()

Output: 


Example 2: Change Button Background Color during Mouse Click

In this example, we will change the color of button to red color while it is in pressed state.

from tkinter import *   

tkWindow = Tk()  
tkWindow.geometry('400x150')  
tkWindow.title('PythonExamples.org - Tkinter Example')
  
button = Button(tkWindow, text = 'Submit', bg='#ffffff', activebackground='#4444ff')  
button.pack()  
  
tkWindow.mainloop()

Output


Example 3: Change Button Background Color during Mouse Click

In this example, we will change the color of button to red color while it is in pressed state.

from tkinter import *   

tkWindow = Tk()  
tkWindow.geometry('400x150')  
tkWindow.title('PythonExamples.org - Tkinter Example')
  
button = Button(tkWindow, text = 'Submit', bg='#ffffff', activebackground='red')  
button.pack()  
  
tkWindow.mainloop()

For the standard colors, you can provided the name of the color. Like red, black, green, yellow, etc.

#Python 

Change Button Background Color During Mouse Click Python Tkinter
Kevin  Simon

Kevin Simon

1664252861

Call a Function on Button Click using Python Tkinter with Example

Tkinter is a cross-platform GUI framework that comes built into the Python standard library. Since it is a cross-platform framework, the same code can be used on any operating system, such as Linux, macOS, and Windows. Tkinter provides an object-oriented interface to the Tk GUI toolkit.


In this tutorial, we shall learn how to call function on Tkinter button click with example programs.

Example : Call Function on Button Click

from tkinter import *
from tkinter import messagebox

tkWindow = Tk()  
tkWindow.geometry('400x150')  
tkWindow.title('PythonExamples.org - Tkinter Example')

def showMsg():  
    messagebox.showinfo('Message', 'You clicked the Submit button!')

button = Button(tkWindow,
	text = 'Submit',
	command = showMsg)  
button.pack()  
  
tkWindow.mainloop()

Output

In this tutorial, we learned how to call a function when Tkinter Button is clicked.

#Python 

Call a Function on Button Click using Python Tkinter with Example

Jim Walsh

1664209118

Comparing Node.js VS Python is a hot topic these days.

Comparing #NodeJS VS #Python is a hot topic these days. Comparing these technologies, StackOverflow claims Python to be one of the most used programming languages while Node.js is the best environment for server development. Watch this video to know more. https://www.youtube.com/watch?v=2OP8ubKnyfo

Comparing Node.js VS Python is a hot topic these days.
Eran Feit

Eran Feit

1663935679

How to classify monkeys images using convolutional neural network , Ke

Hi,

 

This is the second part our of four parts Tensorflow tutorial that enables you to classify monkey’s species images using several methods.

Now we will dive into building a neural network model , using CNN to classify images.
We will cover all the steps of creating the model , train it and validate the outcome using the test images.

 

In the next video we will improve our model using Keras tuner , and learn to choose the right hyper parameters for several important argument , such as how many layers, filters, learning rates etc..

And last we will you a transfer learning to classify the images in a different way.

 

I recommend This graphics card: NVIDIA GeForce RTX 3060 Ti . I am using it to train my Tensorflow models. Get perfect results and performance:   https://amzn.to/3mTa7HX

 

This is the link for part 2: https://bit.ly/3L5oynt

 

I  shared the Python code in the video description.

 

 

Enjoy

Eran

 

#Python #Cnn #TensorFlow

How to classify monkeys images using convolutional neural network , Ke
Bits Lover's

Bits Lover's

1663040377

Crawler Listing - 12 Most Important Crawlers from the Internet

Crawler Listing – 12 Most Important Crawlers from the Internet

 

#cloud #web-development #web-service  #Python 

Crawler Listing - 12 Most Important Crawlers from the Internet
Eran Feit

Eran Feit

1662807897

How to classify monkeys images using convolutional neural network , Ke

Hi,

This is a four parts Tensorflow tutorial that enables you to classify monkey’s species images using several methods.

After learning the images data , planning and coding some initial functions , we will dive into building a neural network model , using CNN to classify images

Than we will improve our model using Keras tuner , and learn to choose the right hyper parameters for several important argument , such as how many layers, filters, learning rates etc..

And last we will you a transfer learning to classify the images in a different way.

 

I recommend This graphics card: NVIDIA GeForce RTX 3060 Ti . I am using it to train my Tensorflow models.

Perfect results and performance:   https://amzn.to/3mTa7HX

 

This is the link for part 1: https://bit.ly/3czlNOr

 

I also shared the Python code in the video description.

 

 

Enjoy

Eran

 

#Python #Cnn #TensorFlow

How to classify monkeys images using convolutional neural network , Ke
Cyril  Parisian

Cyril Parisian

1660932480

CPK Package Manager: Light and Fast Package Manager on C/C++

CPK Package Manager

CPK Package Manager

Very light and easy and fast native package manager to install C/C++ (priority), JS, Python, Rust packages and compile sources when it needed. Same package can be in muitple form for different languages.

Main purpose of this package manager to have simple C/C++ manager with possibility to post packages like npm, and also provide functionality to support same algorithms and code base for different programming languages, so if someone wanna quicksort package in their project on Python - they will get it, and if need the same one for C++ he will get it with simillar command.

Cross-platform. Implemented on C/C++ and provided for Linux, Mac OS, Windows arch.

Usage

Installing packages

cpk install package

Install 2 packages package=1.0 version and package2 latest version

cpk install package@1.0 package2

Publish own package

Publish your own package (inside directory of project):

cpk publish

List of available packages

List of available packages for install:

cpk packages

Update packages

Update all packages:

cpk update

Publish own package

CPK package manager created to distribute any kind of open source packages, but we want to guarantee that any package can be used in commertial software in any form of use. So we recomend to use licenses like BSD or MIT if it possible.

to publish your package you can create cpk.json with following very basic config with package name and list of dependencies:

{
    "name": "example",
    "dependencies": {
        "zlib": ""
    }
}

and use

cpk publish

command to publish you own package

Available commands

Create a new application with the following options:

  • install PACKAGE - Install package
  • publish - Publish current package
  • update - Update tree of packages
  • packages - Update tree of packages
  • -h - Help

Build CPK Client by own

mkdir build
cd build
cmake -DCMAKE_BUILD_TYPE=Release ../
make -j8

Author:  DEgITx
Source code: https://github.com/DEgITx/cpk
License: MIT license
#cpluplus #javascript #Rust #Python 

CPK Package Manager: Light and Fast Package Manager on C/C++
Eran Feit

Eran Feit

1660316027

How To classify cars using Tensorflow , Resnet50 and Python ?

Hi,

 

This is a Tensorflow tutorial that enables you to classify cars images using a transfer learning process.

In this tutorial, We will install the relevant Python libraries , and use  Resnet50 model , and learn how to classify and detect objects in images.

 

Moreover, I recommend This graphics card: NVIDIA GeForce RTX 3060 Ti . I am using it to train my Tensorflow models.

Perfect results and performance:   https://amzn.to/3mTa7HX

 

The link for the video tutorial is here : https://bit.ly/3BJvKmK

I also shared the Python code in the video description.

 

Enjoy

 

Eran

 

#Python #Cnn #TensorFlow

How To classify cars using Tensorflow , Resnet50 and Python ?
Muhammad  Price

Muhammad Price

1659563280

Themis: Easy to Use Cryptographic Framework for Data Protection

Themis provides strong, usable cryptography for busy people 
 


       

General purpose cryptographic library for storage and messaging for iOS (Swift, Obj-C), Android (Java, Kotlin), React Native (iOS, Android), desktop Java, С/С++, Node.js, Python, Ruby, PHP, Go, Rust, WASM.

Perfect fit for multi-platform apps. Hides cryptographic details. Made by cryptographers for developers 🧡

What Themis is

Themis is an open-source high-level cryptographic services library for securing data during authentication, storage, messaging, network exchange, etc. Themis solves 90% of typical data protection use cases that are common for most apps.

Themis helps to build both simple and complex cryptographic features easily, quickly, and securely. Themis allows developers to focus on the main thing: developing their applications.

Use cases that Themis solves

Encrypt stored secrets in your apps and backend: API keys, session tokens, files.

Encrypt sensitive data fields before storing in database ("application-side field-level encryption").

Support searchable encryption, data tokenisation and data masking using Themis and Acra.

Exchange secrets securely: share sensitive data between parties, build simple chat app between patients and doctors.

Build end-to-end encryption schemes with centralised or decentralised architecture: encrypt data locally on one app, use it encrypted everywhere, decrypt only for authenticated user.

Maintain real-time secure sessions: send encrypted messages to control connected devices from your app, receive real-time sensitive data from your apps to your backend.

Compare secrets between parties without revealing them (zero-knowledge proof-based authentication).

One cryptographic library that fits them all: Themis is the best fit for multi-platform apps (e.g., iOS+Android+Electron app with Node.js backend) because it provides 100% compatible API and works in the same way across all supported platforms.

Cryptosystems

Themis provides ready-made building blocks (“cryptosystems”) which simplify usage of core cryptographic security operations.

Themis provides 4 important cryptographic services:

  • Secure Cell: a multi-mode cryptographic container suitable for storing anything from encrypted files to database records and format-preserved strings. Secure Cell is built around AES-256-GCM, AES-256-CTR.
  • Secure Message: a simple encrypted messaging solution for the widest scope of applications. Exchange the keys between the parties and you're good to go. Two pairs of underlying cryptosystems: ECC + ECDSA / RSA + PSS + PKCS#7.
  • Secure Session: session-oriented encrypted data exchange with forward secrecy for better security guarantees and more demanding infrastructures. Secure Session can perfectly function as socket encryption, session security, or a high-level messaging primitive (with some additional infrastructure like PKI). ECDH key agreement, ECC & AES encryption.
  • Secure Comparator: Zero knowledge proofs-based cryptographic protocol for authentication and comparing secrets.

We created Themis to build other products on top of it - i.e. Acra and Hermes.

Installation

Refer to the Installation page to install Themis for your mobile, web, desktop, or server-side application. We highly recommend installation packages instead of building from source.

Languages

Themis is available for the following languages/platforms, refer to language howtos for each:

PlatformDocumentationExamplesVersion
⚛️ React Native (iOS, Android)React Native Howtodocs/examples/react-nativenpm
🔶 Swift (iOS, macOS)Swift Howtodocs/examples/swiftCocoaPods
📱 Objective-C (iOS, macOS)Objective-C Howtodocs/examples/objcCocoaPods
☕️ Java (Desktop)Java (Desktop) HowtoJava projects 
☎️ Java (Android)Java (Android) HowtoAndroid projectsmaven
📞 Kotlin (Android)Java (Android) HowtoAndroid projectsmaven
🔻 RubyRuby Howtodocs/examples/rubyGem
🐍 PythonPython Howtodocs/examples/pythonPyPI
🐘 PHPPHP Howtodocs/examples/php 
➕ C++CPP Howtodocs/examples/c++ 
🍭 Node.jsJavascript (Node.js) Howtodocs/examples/jsnpm
🖥 WebAssemblyJavascript (WebAssembly) Howtodocs/examples/jsnpm
🐹 GoGo Howtodocs/examples/gogo.dev
🦀 RustRust Howtodocs/examples/rustcrates
🕸 С++ PNaCl for Google Chrome WebThemis project 

Availability

Themis supports following CPU architectures: x86_64/i386, ARM, Apple Silicon (ARM64), various Android architectures.

We build and verify Themis on the latest stable OS versions:

  • Debian (9, 10), CentOS (7, 8), Ubuntu (16.04, 18.04, 20.04)
  • macOS (10.12–10.15, 11.*)
  • Android (7–12)
  • iOS (11–15)
  • Windows (experimental MSYS2 support)

We plan to expand this list with a broader set of platforms. If you'd like to help improve or bring Themis to your favourite platform or language — get in touch.

Documentation

Documentation for Themis contains the ever-evolving official docs, which covers everything from deployment guidelines to use cases, with brief explanations of cryptosystems and architecture behind the main Themis library.

Refer to the documentation to learn more about:

Cryptography

Themis relies on proven cryptographic algorithms implemented by well-known cryptography libraries such as OpenSSL, LibreSSL, BoringSSL. Refer to Cryptograhy in Themis docs to learn more.

This distribution includes cryptographic software. The country in which you currently reside may have restrictions on the import, possession, use, and/or re-export to another country, of encryption software. BEFORE using any encryption software, please check your country's laws, regulations, and policies concerning the import, possession, or use, and re-export of encryption software, to see if this is permitted. See http://www.wassenaar.org/ for more information.

The U.S. Government Department of Commerce, Bureau of Industry and Security (BIS), has classified this software as Export Commodity Control Number (ECCN) 5D002.C.1, which includes information security software using or performing cryptographic functions with asymmetric algorithms. The form and manner of this distribution make it eligible for export under the License Exception ENC Technology Software Unrestricted (TSU) exception (see the BIS Export Administration Regulations, Section 740.13) for both object code and source code.

Submitting apps to the App Store

If your application uses Themis and you want to submit it to the Apple App Store, there are certain requirements towards declaring use of any cryptography.

Read about Apple export regulations on cryptography for Themis to find out what to do.

Security

Each change in Themis core library is being reviewed and approved by our internal team of cryptographers and security engineers. For every release, we perform internal audits by cryptographers who don't work on Themis.

We use a lot of automated security testing, i.e. static code analysers, fuzzing tools, memory analysers, unit tests (per each platform), integration tests (to find compatibility issues between different Themis-supported languages, OS and x86/x64 architectures). Read more about our security testing practices in Themis security docs.

If you believe that you've found a security-related issue, please drop us an email to dev@cossacklabs.com. Bug bounty program may apply.

GDPR, HIPAA, CCPA

As a cryptographic services library for mobile and server platforms, Themis is a "state of the art" encryption tool, which provides secure data exchange and storage.

Using Themis, you can reach better compliance with the current data privacy regulations, such as:

Read more about Regulations in docs.

Community

Themis is recommended by OWASP as data encryption library for mobile platforms.

Themis is widely-used for both non-commercial and commercial projects, some public applications and libraries can be found here.

Want to be featured on our blog and on the list of contributors, too? Write us about the project you’ve created using Themis!

Contributing

If you're looking for something to contribute to and gain eternal respect, just pick the things in the list of issues. Head over to our Contribution guidelines as your starting point.

Supporting Themis for all these numerous platforms is hard work, but we try to do our best to make using Themis convenient for everyone. Most issues that our users encounter are connected with the installation process and dependency management. If you face any challenges, please let us know.

Commercial support

At Cossack Labs, we offer professional support services for Themis and applications using Themis.

This support includes, but is not limited to the library integration, with a focus on web and mobile applications; designing and building end-to-end encryption schemes for mobile applications; security audits, for in-house library integrations or high-level protocol; custom application development that requires cryptography; consulting and training services.

Drop us an email to info@cossacklabs.com or check out the Cossack Labs cybersecurity services.

Contacts

If you want to ask a technical question, report a bug or suggest a feature, feel free to start a discussion on GitHub, raise an issue in the issue tracker, or write to dev@cossacklabs.com.

To talk to the business wing of Cossack Labs Limited, drop us an email to info@cossacklabs.com.


Author: cossacklabs
Source code: https://github.com/cossacklabs/themis
License: Apache-2.0 license

#Java, #С #С++ #Node.js #Python #Ruby #PHP #Go #Rust #WASM #React 

Themis: Easy to Use Cryptographic Framework for Data Protection
Eran Feit

Eran Feit

1659083809

How to classify images using Tensorflow , Mobilenet and Python ?

Hi,

 

This is a Tensorflow tutorial that enables you to classify images using a transfer learning process.

In this tutorial, We will install the relevant Python libraries , and use  Mobilenet model , and learn how to classify and detect objects in images.

Moreover, I recommend This graphics card: NVIDIA GeForce RTX 3060 Ti . I am using it to train my Tensorflow models.

Perfect results and performance:   https://amzn.to/3mTa7HX

 

The link for the video tutorial is here : https://bit.ly/3owAuUX

I also shared the Python code in the video description.

 

Enjoy

Eran

 

#Python #Cnn #TensorFlow

How to classify images using Tensorflow , Mobilenet and Python ?
Lina  Biyinzika

Lina Biyinzika

1658752980

VoxelViz: Visualization tool for FMRI Data-sets using Plotly Dash

VoxelViz

Visualization tool for (f)MRI data-sets using Plotly Dash. Submitted to the TransIP VPS competition.

UPDATE: I won the competition! Thanks for TransIP/Tweakers for hosting this competition (and the 1st prize: a kick-ass gaming laptop and HTC Vive) -- this definitely motivated me to keep on developing this app!

To do

Couple of thing that I'd still like to implement and (relatively unimportant) bugfixes.

To fix:

  •  Fix update timeseries when changing brainplot
  •  Fix range of slice-scrolling (currently sometimes out of range)
  •  Fix automatic range in heatmap (colors are now variable)
  •  Fix location (and text-color) of model/voxel checkbox!
  •  Fix darkgrey text on Hover-box

To implement:

  •  Refactor to make standalone app!
  •  Add option to load in own data (with dcc.Load component)
  •  Add option to show separate regressors (not only model fit)
  •  Option to toggle between timeseries and frequency view (nice to show effect of e.g. highpass filter)
  •  Add download script to download data to check examples (teaching/showcase) locally

Intro

This tool was originally developed for the TransIP VPS competition ("come up with an original idea for a virtual private server"), but mainly because I was looking for an excuse to mess around with the new Plotly Dash framework. The app turned out better than I expected and I think I'll try to convert it into an open-source package for everyone to use. That is, I'll rewrite it such that any neuroimaging researcher (or teacher! more about that later) with a proper dataset can use VoxelViz to show/demonstrate it.

Usage

As of now, VoxelViz can be used in two ways, which correspond to the two apps in this repository (the one in the usecase folder and the one in the teaching folder).

Showcase results

First, it can be used to interactively visualize results from (group-level) fMRI analyses. Or, in layman terms: show pretty brain pictures. In the usecase folder, the app.py file runs an app that shows the results from my fMRI study (freely accessible!) about the representation of emotion in the brain. VoxelViz visualizes the brain images corresponding to the different analyses we did (left panel of the app), which can be interactively manipulated by, for example, adjusting the statistical threshold, orientation (X, Y, Z), and the current "slice" (specific 2D view; changing this is like "scrolling" through the brain image).

Additionally, VoxelViz interactively visualizes the "timeseries" from the voxel (3D cube, which represents the unit of measurement) that you hover over in the brain image from the app's left panel. So, the right panel (timeseries plot) updates according to where your cursor is in the app's right panel - cool huh?

Teaching neuroimaging

Next to the "showcase" use, I think this app could be quite helpful in a teaching context. It happens to be that I myself teach two "neuroimaging" (analysis of brain-data) courses at the University of Amsterdam, and I struggly to explain "dry" (often statistical) concepts during my lectures and computer labs. Of course, I try to make my lectures more "lively" by including a lot of brain images, but 2D representations just don't do the trick. How cool would it be to show students "boring" things, like the effect of a high-pass filter on fMRI analyses, interactively in a web-app? Well, the folder teaching of this repository contains an app for just that! It shows the results (brain images) from different preprocessing pipelines. This way, you can check out (or show students) what happens to the brain images when you apply a high-pass filter (or not!) or when you (spatially) smooth your data. This should be especially clear from the timeseries plot in the right panel!

Features

I've alluded to some of the features in the previous section already, but here I'll describe them in more detail.

Show different datasets/results

When you start the app, you'll see a dropdown menu in the top left corner of the brain image plot. In the showcase example app, this dropdown menu contains the "contrasts" (results) from different analyses, which are referred to by keywords such as "action", "interoception", "action>interoception", etc. (These results refer to analyes which tested how the brain activates in response to action-oriented emotional states, interoception-oriented emotional states, and the difference between action and interoception states, respectively.) Changing the "contrast" will update both the brain plot and the timeseries plot!

alt text

Viewing/scrolling options

The brain is of course a 3D dimensional object that is, here, visualized in 2D. I've included two options to configure which part (and in which orientation) you're viewing the brain image. Next to the "contrasts" dropdown menu, you can choose whether you want to view the brain in a saggital orientation (from the side; "X"), coronal orientation (from the front; "Y"), or axial orientation (from the top; "Z"). Next to the "X/Y/Z" option, there is a slides that allows you "scroll" through the brain.

alt text

Adjust the (statistical) threshold

Usually, results from (f)MRI analyses are thresholded, to remove insignificant (de)activations. Therefore, the results are by default thresholded at abs(Z) > 2.3. But sometimes its informative to check out the unthresholded results, so VoxelViz contains a slides to adjust the threshold! Adjusting the threshold automatically updates the brain plot (higher thresholds should show less red/blue and vice versa).

alt text

Visualize underlying timeseries

In the right panel of the app, there is a "timeseries" plot that shows the underlying signal of the highlighted voxel (3D equivalent of pixel) in the brain plot. In other words, if you hover over a voxel in the brain plot, the timeseries plot will automatically show the underlying signal of that voxel.

alt text

Visualize model fit

The "results" in the left panel (the brain plot) are derived from models fit to the timeseries data in the right panel. By clicking on the "Model" checkbox above the right plot, you can visualize the model fit to the data, which again updates when you change the position of your cursor. This is especially helpful in a teaching context when you want to show students the effect of preprocessing options (such as filtering, confound regression, etc.) on your model fit!

alt text

See for yourself!

VoxelViz runs as a standalone app on a VPS X8 BladeVPS from TransIP, which can be viewed at teaching.lukas-snoek.com and showcase.lukas-snoek.com.

If you want to mess around with the app yourself with your own data, you can clone this repo! First, you need to install the dependencies, which you can do through:

$ pip install -r requirements.txt

Then, put your own data in the usecase or teaching folder (and rename the folder if you want), change the config.json file, and run:

$ python app.py

Now, go to http://localhost:8050 in your browser to view and use the app!


Author: lukassnoek
Source code: https://github.com/lukassnoek/VoxelViz
License:  GPL-3.0 license

#dash #Python 

VoxelViz: Visualization tool for FMRI Data-sets using Plotly Dash