We have heard a lot about GPT-3, OpenAI and their trained models to implement your own AI in a fast and pre-trained way. The results of your demos are fascinating, but… are they really like that? or are they “ready”?

Well, raising that same question, I have decided to make an implementation to integrate your Telegram user with said AI and dockerize it, so that by configuring your keys from docker-compose.yml you can easily start the bot, and have it start answering this intelligence for you.

One of the advantages of Python code is the large number of libraries that are available, in this case we are going to use QtWebKit, with which we can easily create a browser or embed it within our application. We edit a file, for example pybrowser.py with the following content

#!/usr/bin/env python

import sys
from PyQt4.QtCore import *
from PyQt4.QtGui import *
from PyQt4.QtWebKit import *

app = QApplication(sys.argv)

web = QWebView()
web.load(QUrl("http://www.thempra.net"))
web.show()

sys.exit(app.exec_())

Y ya solo nos queda lanzar nuestro mini-navegador:

python pybrowser.py

A partir de aquí se ha de trabajar con los botones, barra de navegación , ......

Next Saturday, August 18, I will be at MUNCYT in La Coruña, with a talk on how to take brain wave readings with low-cost Electroencephalograms (EEG).  I will present the Theeg project and then Jose Angel Mateo will give a presentation on the DotNetClub.

Agenda

10:00-10:15 Presentation of the event and speakers. 10:15-11:00 ‘Theeg: Reading other people’s minds’ Theoretical session. 11:00-11:45 ‘Theeg: Reading other people’s minds’ Practical session. 11:45-12:00 Coffee break

12:00-12:45 ‘IEEE-UNED Microsoft DotNet Club’. 12:45-13:00 Farewell and closing.