Group Abstract Group Abstract

Message Boards Message Boards

[WSG20] New Machine Learning Basics Study Group begins Monday, November 16

POSTED BY: Jamie Peterson
35 Replies
Posted 5 years ago
POSTED BY: tsukisan
Posted 5 years ago
POSTED BY: rjehanathan
POSTED BY: Mickaël Bouvier
POSTED BY: Mickaël Bouvier
Posted 5 years ago
Attachments:
POSTED BY: r t
Posted 5 years ago
POSTED BY: tsukisan

Here are a few stories which I applied the sentiment classifier and generated matrix plots. SCF's 1,2, and 3 are the various stories. P 1,2, and 3 are the resulting probabilities extracted from the association returns from SCF's.

Attachments:
POSTED BY: Brad Button
Posted 5 years ago
ListLinePlot[
 Classify["Sentiment", 
  TextSentences[ExampleData[{"Text", "AliceInWonderland"}]], 
  "Probability" -> "Positive"]]
POSTED BY: rjehanathan
POSTED BY: Jon McLoone
POSTED BY: vincent feng
Posted 5 years ago
POSTED BY: tsukisan

Right. The neural net model is being used to extract features from the words here and those features are used to place the words in the 2-dimensional feature space.

Posted 5 years ago
POSTED BY: tsukisan

A final challenge

Create an deploy a simple machine learning app of your choice. Feel free to use one of our built-in classifiers or predictors.


To get you started, here is the example we looked at in the study group session:

CloudDeploy[
 FormPage[
  {{"myPicture", "Your picture"} -> "Image"},
  Module[{entity},
    entity = Classify["NotablePerson", #myPicture];
    Row[{#myPicture, Spacer[10], entity, Spacer[10], 
      entity["Image"]}]
    ] &,
  AppearanceRules -> <|
    "Title" -> "Which famous person do you look like?", 
    "Description" -> 
     "Enter your image and we'll show you your famous twin!!",
    "SubmitLabel" -> "Show me who"|>,
  PageTheme -> "Black"
  ],
 "FamousLookAlike",
 Permissions -> "Public"
 ]

Here's the simpler version for testing: https://wolfr.am/RddbHZR4

Thanks!!!! I totally didn't even look at the emails because I had the event saved in my calendar. My bad.

POSTED BY: Stephanie Meyer

@Stephanie Meyer, please check your Study Group email reminder that was sent earlier today. It includes a link to the quiz.

POSTED BY: Jamie Peterson

Where do we find the quiz for the course? I haven't seen an email with a link or a file for it. I want to try it before our final Q and A class tomorrow.

POSTED BY: Stephanie Meyer
Posted 5 years ago
POSTED BY: C Ellis

Let's bring your question to Friday's Q&A session with the development team.

Daily Challenge: Day 3

Today's challenge will lead nicely into tomorrow's topic "Neural Networks". Visit the GloVe 100-Dimensional Word Vectors Trained on Tweets available in the Wolfram Neural Net repository. Look at the example offered under "Feature Visualization". Try FeatureSpacePlot with FeatureExtractor set to this particular model to visualize a clustering of words in any text of your choice. Share any interesting results you find here.

POSTED BY: Kevin Hawekotte

I would attribute such results really to the training data. The performance of most classification algorithms is closely tied in with the quality and availability of the training data.

Posted 5 years ago
Attachment

Attachments:
POSTED BY: C Ellis

Just to give context: My approach could just be naïve (I'm a beginner here, so pardon me in advance) but using the Classifier function to analyze the sentiment is throwing some weird results my way. I wanted to just use a few sentences before I approached with a larger text file.

Here is what I have:

enter image description here

I was wondering if the word 'extremely' had some weight to it that made the output different? Thanks for the help!

POSTED BY: Kevin Hawekotte
Posted 5 years ago

Thank you! That worked perfectly!

I'm wondering about slide 7 of the presentation where it mentions automating the method choice by testing different approaches on subsets of data - do you have any examples of this?

Thanks!

POSTED BY: Ivan G

Daily Challenge: Day 2

The dataset at https://datarepository.wolframcloud.com/resources/State-of-the-Union-Addresses contains the complete text of State of the Union addresses from 1790 to 2019. Test the performance of a Predictor trained to predict the age of the president from the date and text of the address.

Posted 5 years ago

Hi,

I'm confused about how to import excel files into the "{parameters}->label" format. My excel file looks like the picture below, with some of the data being blank.

Input Data Format in Excel

POSTED BY: Ivan G
Posted 5 years ago
Attachment

Attachments:
POSTED BY: C Ellis
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard