It snowed again! I think that’s more snow in one week than the past two years

GPT Agents

• Need to compare the performance of each model for each probe and compare to ground truth. One thing to point out is how little data there is to sample:

SBIRs

• Fixing the “find matching”
• Make node size log-based

Book

• Put together some more data. Need to change the maps a bit

One year ago things were pretty crazy here

GPT Agents

• From On the Reliability and Validity of Detecting Approval of Political Actors in Tweets (Section 4.2) as an example of keyword SOTA :
  • We evaluate OTS and custom methods on the following datasets. While some of these datasets have common targets, for example, Trump is present in four of them, they are all collected in different periods of time, with different keywords (c.f Appendix B). All datasets have stance labels of ‘favor’, ‘against’, and ‘none’ towards the targets. (EMNLP)
• Finished with generating the new data, now we get to see if it works!
• It’s pretty good. Here’s the two GPT models, one trained on the first 50k reviews of the American dataset (iso) and the other trained on the first 50k of the American dataset that do not contain the string “vegetarian options”. The probes are:
  • no vegetarian options
  • some vegetarian options
  • several vegetarian options
  • many vegetarian options

• Basically identical
• Now I need to compare the response vs the ground truth for each of the probes

Jamie Raskin just released a book that apparently has some overlap with my work? Trying to track it down. Here’s something from CBS

GPT Agents

• Creating unistar models from the corpora that have ‘vegetarian options’ removed. As they are trained, I’m also generating responses to the vegetarian prompts that I’ll do the star and unigram compares with. Then put that in a table and write the paper around it. Also, add the Floober part or something fanciful.
• Models are all created. Finished running the first two and am now adding sentiment to them

SBIRs

• Continue code cleanup and documenting. I managed to remove a good deal of code that had to do with handing raw text selection of topics, since that seems to be broken in tk
  • Finished commenting QueryFrame. Now I need to fix that listing problem in on_link_existing_clicked()
• Set up meeting to discuss LAIC dev plans – done

It got really cold last night and I had forgotten to turn the water off to the outside and lost the faucet on the deck. Could have been worse. At least the pipes didn’t burst

Thinking about submitting a writeup on Sanhedrin 17a (Section 10.4 of the dissertation. Mostly) for the We Robot conference

• Abstracts due: March 7
• Decisions: May 9
• Final papers due: August 8

Book

• Playing around with negative scalars to see how that works. This resulted in some code cleanup and a better color gradient. Not sure if it looks better though:

Still like this better:

SBIRs

• Sprint planning
• Working on code cleanup for MabBuilder. First, adding comments!
• Fixed the exit condition that happened when clicking the ‘X’ close icon in the text compare popup
• Next, check through all the button behavior in QueryFrame
  • Set Group
  • Add Topic/Seed
  • Add Topic
  • Add Seed
  • Find Closest (and dialog)
  • Add Group
  • Next Seed
  • Rerun Seed
  • Get Topic Details
  • Direct Prompt
  • Wikipedia
  • Link Existing (make this work with descending length topics)

GPT Agents

• 3:30 Meeting. Going to make some models that explicitly are missing the phrase ‘vegetarian options’ from the training corpora. I’ll then run those as to compare to ‘vegetarian options’ in the ground truth by star and the other GPT models

This looks interesting: www.oreilly.com/library/view/natural-language-processing/9781098103231/

Book

• After a few false starts, I have the terrain extended:

• I still need to:
  • add a ‘lit’ and ‘unlit’ node for terrain and labels – done
  • add a height scalar – done
  • toggle grids and axis – done
  • Shift keys to move the lights the other direction, plus lambda functions for the parameters – done
  • Maybe add fog? docs.panda3d.org/1.10/python/programming/render-attributes/fog – nope, can’t get the fog to be relative to the terrain center

Today’s progress:

GPT Agents

• Get the number of POSITIVE and NEGATIVE sentiment for each isolated model and compare to ground truth. Make a chart and add to the draft. This is the part that shows that creating models for a population captures that population’s patterns, and that this method is more accurate and reliable than assuming that one general model has all the information needed in an accessible way. Done

Happy New Year everyone! It’s been warm here in the Baltimore region. Working on terrain visualization.

Got lighting working. You attach the lighting node to the node you want it to move with and then set it to the node you want to light. Here’s the code:

def add_directional_light(self, name:str, root_node:NodePath,   target_node:NodePath, color:Tuple = (1,1,1,1)) -> NodePath:
    dlight = DirectionalLight(name)
    dlight.setColor(color)
    dlnp = root_node.attachNewNode(dlight)
    target_node.setLight(dlnp)
    self.light_dict[name] = dlnp
    return dlnp

I also added grid lines to emphasize the contours of the terrain. I’m liking the overall look:

The last thing I want to do is extent the terrain beyond the nodes so that everything rises from a flat surface