tomthetrainer / june26class Goto Github PK

Introduction/01.md has an outline page, since the outline is in flux this will have to be edited for final release

@turambar @bpark738

If you could answer these questions that would come from a student then I can incorporate that into the class.

What is a Sequence?

Are my log files a sequence, are my financial transactions a sequence? Why and how do they differ?

RNN gets sequence in, what does it emit as output?

Univariate, multivariate, resampling: What are these and when do I need to resample?

What is the relationship of the time series to the input layer? In a MLP I have 4 features then I have 4 inputs, in an RNN what is the relationship. Input to MLP is a vector, value, tensor, what is the shape of that input, and compared to an RNN what is the shape of the RNN's input?

If I have log files with events that occur randomly with millisecond granularity, how do I handle that? Split on the second? Summarize per second? Split on the millisecond? What is the effect of those decisions on the complexity of my NN?

Processing input seems a lot like "expert feature engineering" is it? Or can it be simplified in some way?

What about Video content? Feels like a RNN, sequences of images, but perhaps a CNN because they are images?

If my sequence size is fixed, say temperature reading every minute from some lab experiment and I split that on day, would an MLP be as useful or do I need an RNN?

This repo was slightly out of date when cloned, check with the content I have in

/Users/tomhanlon/SkyMind/TRAINING/New_Courses/Standard_Courses/Intro_to_Neural_Networks_with_DeepLearning4J/Intro_to_Neural_Networks_with_DeepLearning4J

To verify it is latest version

@turambar had some nice titles for his labs, mine are dull, please advise.

We have

• LSTM for Character Sequence Generation

This is Shakespeare but with weather forecasts
Is that name good, or should it have some more technical terms in it?

• LSTM for Univariate Sequence Classification

This is Susans UCI signal classification, is the title good.

Below are the Labs Dave and Briton are providing.

• multivariate time series classification example using physionet (irregularly sampled + time features)
• multivariate time series classification example using physionet (irregularly sampled, no time features)
• multivariate time series classification example using physionet (resampled)

I created a sample directory that demonstrates what you need to know.

1. Clone the repo
2. switch to "Time_Series_Analysis_With_DL4J/Dave_Briton_Demo"
3. Run "showoff serve" in that directory
4. Point a browser at http://localhost:9090/presenter#1
5. Point another browser window at http://localhost:9090/
6. The Presenter window will drive the display window

Authoring content

1. Showoff.json controls the Table of contents
2. showoff.css configure the logo and some styles for a few items
3. In general I create one file for directory but that is not required
4. Create your content in one directory or multiple directories
5. Point to it in the main showoff.json
6. Do not stress out about order of content, just give me good stuff and I can move it around as needed.
7. Comment here if you need further advice

@turambar @bpark738

I like to start with an early Lab, I will have them do simplest network that is basically the simplest NN you could imagine, that Lab is really just to verify they have a working environment and can use IntelliJ.

The second Lab I have is a DataVec Lab.

I demo a spark analysis from the demos section
AbaloneDataTransform

For this class it might be nice to demo something more appropriate to time series.

After the demo I have a basic DataVec Lab,

The code is DataVecLab, basic ingest of iris csv data with a prebuilt Neural Network.

It might be nice to have a basic Sequence oriented DataVec Lab.

I could write it but if either of you wanted to summarize the issues with reading time series data and the datavec classes involved that might make a nice intro.

• @turambar: brainstorm course outline