Code Monkey home page Code Monkey logo

resources's Introduction

GPSS : Labs and Resources


Labs

Notebooks

Notebooks and answers are in the lab directory. Some changes from last year:

Lab 1: Gaussian Process Regression

  • Support for Python 3.5+ only
    • This is not essential, but notebooks require some adaptation to change
    • Uses Python 3 specific styling, and have made use of @ for matrix multiplication, amongst others
  • Exercises:
    1. Effects of parameters on (RBF) kernel
    2. Modelling with and validity of combined kernels
    3. Identify the covariance function
    4. Fitting a GP, changing parameters and sampling posterior
    5. Extrapolating with a GP, embedding prior knowledge for a better
    6. Fitting combinations of kernels to the Mauna Loa dataset
    7. Improving the fit

Lab 2 : Non-Gaussian likelihoods / classification

  • Non-Gaussian Likelihoods introduction, with Student's t, Bernoulli examples and 2D classification example
  • Sparse GPs with small comparative example and longer sparse exercise
  • No SVI

Lab 3 : Bayesian optimisation

  • Written by Javier

Equipment

setup

Setup information available as pdf, png, and svg

Description

The speaker should have a laptop plugged into an HDMI cable, and powered. Any conversion between display adaptors should be down so that the HDMI cable can be plugged in. Anyone with a non-UK power lead should use the provided conversion adaptor (apparently there's some legality issues here around using standard ones, but I've identified a multi-adaptor that should be fine). The speaker should also be fitted with the microphone and audio transmitter – this should be battery powered, rather than by mains, due to reported feedback in the audio receiver. The batteries should be changed in the transmitter after every speaker, and in the receiver every day.

The laptop display is connected via HDMI to the splitter, which needs to be independently powered. It needs to be plugged into the middle port, marked "HDMI-IN". One out should be plugged into the projector via HDMI (I am presuming that an HDMI cable will be provided in the room to connect to the projector).

The second output from the splitter should be directly converted to VGA using the non-powered, short HDMI-VGI adaptor. On the other end of the VGA cable, there should be an gender change adaptor, and the powered VGA-HDMI converter to convert back to HDMI. The adaptor is powered via a micro USB port. The reason for the HDMI-VGA-HDMI conversion is to strip the digital signal (by converting to analogue and back again) of any encryption that may prevent us from recording the screen. This is a confirmed issue on laptops running recent versions of macOS, and the VGA conversion is a successful work-around. Conceivably this could be done before the splitter (is there any known reason to do it a particular way ?, the limitation currently is that we lose sound output from the laptop in the recording, but can pass a digital signal to the projector).

The camera is powered by a USB fixed in the handle, with a 1 m extender available. The output from the camera is passed via microHDMI (inside the screen cover), for which we have a microHDMI to HDMI cable. The camera has no memory, and all recording is performed on the recording box, so don't press record on the camera. The audio receiver has an aux output which needs to be converted to R/W component. Both the camera output and audio should be plugged into channel 1, via HDMI/component respectively. There are two component adaptors provided in the lecture recorder. The output from the splitter should be plugged into channel 2 via HDMI. Headphones can be plugged into the front, for the operator to check the quality of audio in the recording.

Inventory

# Item Notes
1 Video Camera Sony HDR-CX240E, needs USB charger
1 Tripod
2 Audio Transmitter / Receiver Battery powered
1 Microphone (clip-on) Aux (in)
1 Lecture Recorder PAT done 26/07/18
1 DC 19V Adaptor + lead for recorder
1 4-way HDMI splitter split laptop signal between projector and recorder
1 DC 5V adaptor for splitter
6 USB Flash Drives (2GB) need checking
20 Batteries (AA) Checked for charge
# Cable Lengths Notes
2 microHDMI [M] → HDMI [M] 20 cm, 1 m camera → recorder
1 USB-A [F] → USB-A [M] 1 m extends camera power cable
2 aux [M] → aux [M] 30 cm, 1.5 m non-essential
1 aux [M] → component [M,M] 1.5 m audio receiver → recorder
3 VGA [M] → VGA [M] 1.5 m used for analogue conversion in VGA/HDMI setup
1 microUSB [M] → USB-A [M] 1 m to power VGA/HDMI adaptor
2 HDMI [M] → HDMI [M] 2 m, 10 m laptop → splitter, splitter → recorder (via VGA)
1 HDMI [F] → HDMI [M] 2 m extender, non-essential
# Adaptors Notes
1 HDMI [F] → miniHDMI [M] non-essential
2 HDMI [F] → microHDMI [M] non-essential
1 HDMI [M] → VGA [F] VGA/HDMI conversion
1 VGA [F] → VGA [F] VGA/HDMI conversion
1 VGA [M] → HDMI/aux [F,F] VGA/HDMI conversion, requires power, needs USB charger
2 USB power chargers For camera and VGA/HDMI conversion
1 mini DisplayPort [M] → HDMI [F] adaptor Laptop output adaptor
1 DisplayPort [M] → HDMI [F] adaptor Laptop output adaptor
1 USB-C [M] → HDMI [F] adaptor Laptop output adaptor
1 DVI [M] → HDMI [F] adaptor Laptop output adaptor
1 Power passport If speaker does not have UK plug
1 Headphones To check recording audio is working
1 Laser Pointer For speakers if necessary
12 Assorted Marker Pens
1 Multiline power adaptor For recording setup, can try and get from support (TO GET)

Misc.

Orders (to be done each year)

  • T-shirts
  • Branded stationary
    • Pads of paper
    • Pens
  • Lanyards
    • Holds name badge of size : 8.5 x 5.5 cm / 3.35 x 2.2 inches

resources's People

Contributors

wilocw avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.