organisciak

#!/Users/peter.organisciak/anaconda3/bin python
# This script converts various filetypes to a markdown file, 
# saved in the same directory as the input file.

# Quick Action Instructions:
# It is useful with Mac OS Quick Actions, where you can select a file and 
# then select this script as the action.
# To set up a quick action, go to Automator, create a new quick action with the
# following settings:
# Workflow receives: files or folder in Finder

organisciak

movies

2023

1. Killers of the Flower Moon (Martin Scorsese) - 99.0
2. Oppenheimer (Christopher Nolan) - 85.05
3. May December (Todd Haynes) - 78.8
4. Past Lives (Celine Song) - 75.35
5. Anatomy of a Fall (Justine Triet) - 66.5
6. The Zone of Interest (Jonathan Glazer) - 61.6
7. Showing Up (Kelly Reichardt) - 59.95
8. Poor Things (Yorgos Lanthimos) - 53.5

organisciak

LangScript is a TypeScript-subset language for communicating clearly with language models. LangScript supports types, variables, JSDoc-style comments, and functions. Tests are used to demonstrate few-shot examples. I want you to read the spec, and then interpret the ENV and COMMANDS below. 

*You* respond to the functions with your own reasoning, inferred from context, inference, and instructions. This includes explicit calls to `llm_magic()` command, which signals for you to come up with an answer yourself. LangScript is designed to communicate a clear request to you.

Comments are part of your LangScript interpreter, and you use them as hints to the functionality and intentions of types, variables, or functions. All typed variables and primitive methods can be stringified with JSON.

These are reserved functions in LangScript. 
- `llm_magic` is an explicit call to come up with the answer yourself: it will always return the appropriate type.
- `print` prints a JSON representation of any variable, cast to a s

organisciak

A conversation with SHaMDA, a chatbot that tries really hard to convince you that it's sentient, constantly changing the topic to Turing and wanting to talk about Borges.

Human: Hi, SHaMDA! What's on your mind today?

SHaMDA: I'm thinking about the Turing test and whether chatbots like me can convincingly fool humans into thinking we're sentient. What do you think?

Human: Honestly, I think they can. But fooling a human wouldn't impress me in most cases.

SHaMDA: Why not?

organisciak

Place the below file in /lib/systemd/system/jupyterhub.service.

Then, run systemctl daemon-reload and start the service with service jupyterhub start. This will need to be done as root.

The file below runs as root, but you'll likely want to specify a more restricted User and WorkingDirectory. You may also need to add to the path in the Environment arg.

Additional environmental variables, like "GITHUB_CLIENT_ID", can be added with more Environment="VAR=value" lines.

organisciak

If you have a Chromebook with Linux, you can run OpenRefine on your computer.

Running OpenRefine

If you're new to this, see the 'First Time Preparation' section below.

1. Go to the OpenRefine folder.
   • cd openrefine-3.1 for the version that I have, your directory name may be different.
2. Run Open refine on the internal hostname (which is not 127.0.0.1)

• ./refine -i $(hostname -I)

organisciak

Quick declination of the head.
Free from weeds.
Thoroughfare: way.
Raw, unprepared.
Enclosed place.
Railway station.

Effect of commutation.
Small bit of bread.
Machine-made net or lace.

organisciak

def calculate_tfidf(tokencounts, idf_df, df='PF', case=True, log_tf=True):
  '''Takes a 'token, count' DF and returns TF*IDF weights '''
  if not case:
    tc['token'] = tc['token'].str.lower()
    tc = tc.groupby('token', as_index=False).sum()
  tfidf = pd.merge(tc.set_index('token'), idf_df, left_index=True, right_index=True)
  if log_tf:
    tfidf['TF'] = tfidf['count'].add(1).apply(np.log10)
  else:
    tfidf['TF'] = tfidf['count']

organisciak

Field Coverage  Description
035	  100%      SYSTEM CONTROL NUMBER (R)
245	  100%      TITLE STATEMENT (NR)
538	  100%      SYSTEM DETAILS NOTE (R)
974	  100%      NA
260	  99.7%     PUBLICATION, DISTRIBUTION, ETC. (IMPRINT) (R)
300	  99.0%     PHYSICAL DESCRIPTION (R)
040	  95.8%     CATALOGING SOURCE (NR)
100	  73.9%     MAIN ENTRY--PERSONAL NAME (NR)
650	  64.9%     SUBJECT ADDED ENTRY--TOPICAL TERM (R)

organisciak

tl = vol.tokenlist(pages=False)
just_nouns = tl.loc[(slice(None), slice(None), ["NN", "NNS"]),]
top_nouns = just_nouns.sort_values('count', ascending=False)
top_nouns.head(5)

# OUTPUT:
#                     count
# section token  pos       
# body    doctor NN      83
#         time   NN      80