#Notes for Sept. 3rd
Noah - using too much jargon, use less technical works and more images - what is the motivation link to pymks how do you connect this for someone who has machine learning skills
Mahdi Roozbahani - tortuosity is a structure measure need to connect structure measure and physical properties still needs some definition
Dipen Patel - didnt communicate well with general group include pictures and links to papers
Alicia and Geet - Include references and more images to communicate
**Include more short term goals - what can we do in 2 weeks
Nils - do you need 3D?
Ali - flickr needs to use jpg or png image not the link that they give you
Patxi - documenting where the literature came from need a better from Experimental data down - a little nebulus
David Brough - connection between images(structure) and graph(property) is the richer problem X-ray is an aggregate signal defects will be impossible to reconstruct - you will only get top level features reconstruct if you want other properties other than this
Alex Lohse- spacial statistics between crystalline and amorphous regions can we define local state and do spacial corolations
Ahmet - presenting what he is capable of pick one project and he will be able to make his tools available for others who need it
##Spacial Corolations
microstructure function
m(x,n) dxdn = doublesum(mhs
(notes taken on paper due to mathematical formulas and drawings)
##Communication Often times data scientists have to find information There is a lot of information in different places that make it hard to pull out the data. Have to read through a lot of text to know what is important Layers
- content - data ect.
- context - description of what content is
- layout to add style
Inspect element allows you to see the html of the page - you dont want to have to go look at all this info THe pages have a nice plane text format when using jeckyl
Need to know where data is to determine if assertions are corrrect but dont always have it Calls to content that we are using to make this report text becomes the context description
Allows us to share data via research pages ###Creating posts must have a certain naming scheme yyyy-mm-dd-name.html each post must contain Yaml front matter as defined by "---" "---" want to apply styling layout: post (can use post, page and default because these are what we have as layout)
post includes disqus - liquid enables this taking data - calling it raw
raw: link in ---
! [click to download] ( {{page.raw}})
goal is to put data in a place that can be shared as links
make information already availble and then if you want to share it you can easily
This is valuable because the things you see is not everything there is a lot of info behind presentation layers
Plotly gives you raw data access you can load it into python by adding .py to the url
it is versioned
by providing me with a say fliker link
sometime all I want to see is what does the data look like
We want to see projects in these formats Best way to help is to ask questions - i tried this and it didnt work For our projects we need to generate information and trasfer it quickly post for experiments posts for research papers you can automate posts your productivity depends on how you communicate with people