For the deployment aspect, I have tried two more independent product approachs, first one deployed at

http://vfvalue.com/ai/image.html

this page needs authentication to login,

it features chinese input, The following example the input is dog in chinese, and then the result return "True"

Starting from this, I also tried APIFunction[],and embed the api call in node.js at the server side, that way I can independent get total control of the front end look like and pave a path for independent commercial projects using web interface.

The result:

Detecting the oldest and youngest in a picture.

https://www.wolframcloud.com/obj/tsukisan/LookingForExtremes

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.

FeatureSpacePlot[
 DeleteDuplicates@
  TextWords[
   "The Wolfram Language includes a wide range of state-of-the-art \
integrated machine learning capabilities, from highly automated \
functions like Predict and Classify to functions based on specific \
methods and diagnostics, including the latest neural net approaches. \
The functions work on many types of data, including numerical, \
categorical, time series, textual, image and audio. 
   The Wolfram Language has state-of-the-art capabilities for the \
construction, training and deployment of neural network machine \
learning systems. Many standard layer types are available and are \
assembled symbolically into a network, which can then immediately be \
trained and deployed on available CPUs and GPUs. 
   Using a variety of state-of-the-art methods, the Wolfram Language \
provides immediate functions for detecting and extracting features in \
images and other arrays of data. The Wolfram Language supports \
specific geometrical features such as edges and corners, as well as \
general keypoints that can be used to register and compare images."], 
 FeatureExtractor -> 
  NetModel["GloVe 100-Dimensional Word Vectors Trained on Tweets"]]

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.

Unfortunately it appears that PredictorMeasurements cannot deal with test data being in a Dataset. We will have to wrangle it into key-value pairs {feature1, feature2,...} -> target

For this specific dataset we can do the following:

modifiedTesting = Most[#] -> Last[#] & /@ Values[Normal[testing]]

which will set up the test data as: {date, text of address}-> age and allow you to use the PredictorMeasurements function.

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.

@ivang1 Perhaps the example in the attached notebook will help.

Attachments:

WranglingDataFor...nb

This is being automatically done by the Classify or Predict function. You will observe it in the training progress panel as the training happens. You will see it is trying different algorithms.

Ok. So I have one of those pesky theory questions... In yesterday's session, we saw several examples of "very small numbers" returned for (admittedly unlikely) probabilities within the classification function exercises. Returning to the concept of how Mathematica/WL symbolically evaluates the expression of SQRT(2)^2-1 from a previous seminar series... would it be safe to say that these "very small numbers" are themselves measurements of computational error?

I've attached screen captures with evaluation of the "Supervised Learning: Classification - Basic Example" from my own computer (64 bit PC) and from the lecture video. (Mr. McLoone's obviously superior computing cluster) So what result does everyone else get when evaluating this expression?

And if my hypothesis is correct (more processing cores = reduced computational error) is there a method for "forcing" WL to askew floating point arithmetic? Or does this even matter provided we askew "mission critical" uses as mentioned in the video and remain cognizant of error propagation?

Or... is there some form of symbolic approach to this issue? (for which I admittedly have absolutely no idea what such methods might even look like) Just curious. Seems like an emerging issue as we continue to measure more things more often with ever increasing levels of precision.

Attachments:

Capture - Comput...JPG

thank you. I'm just now beginning to see how esoteric this topic is typically considered, yet how clearly it was evident while immersed in learning R just after your Summer Boot Camp sessions.

I believe yet another example was provided in today's lecture when we were evaluating predictive probabilities for boolean operators (or something like that) with values of .99998 accompanied by mean error of 1.00058 (or something like that) note: it's been awhile since I've taken a stats class.

So... perhaps a more precise version of my question for Friday is: If we accept that these tiny values are, in fact, measurements of computational error with magnitude attributable to specifications of the system which had computed it; and we are willing to admit these values of something times 10 to the negative 50th power are meaningless aside from representations of computational processor error...

why not just round it zero? or perhaps some symbolic feature of irrelevance? continued use seems to create user confusion from inconsistent outputs and higher risk of error propagation than benefit.

looking through the documentation I saw some functions which appear to effectuate this under the heading of Precision & Accuracy Control, but why not make such settings default? or default for a particular user type? (like "student/learner") then everyone in a class would receive similar results regardless of what type of computer they are using.

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:

ML_Challenge_1.nb

ListLinePlot[
 Classify["Sentiment", 
  TextSentences[ExampleData[{"Text", "AliceInWonderland"}]], 
  "Probability" -> "Positive"]]

Looking forward to reviewing these topics along with you all. Please feel free to post any questions you may have, from the course or from the daily study group sessions, here. Also look out for the fun daily challenges to showcase your new machine learning skills.

See you on Monday.