Python
ML / Models
Generative AI With Gemini
Neural Networks / Emergent Behavior / LTH
Miscellaneous
100

What is the correct output value?

print('AI Club')

AI Club

100

What does LLM Stand for?

Large Language Model

100

What is the key word to install google-generativeai on terminal? (Hint: ___ install)

pip

100

What is a neural network?

A large complex machine learning model with many nodes. 

100

What is the difference between good and bad data?

Good data is useful, relevant, accurate, and complete, while bad data is biased, not specific, and inaccurate.

Example:

Good data: "It was 72 degrees today."

Bad data: "It was hot today."

200

This code segment does not work as intended. IT is meant to print 11. What is the error in this code segment?

var num1 = 5

num2 = 6

print(num1 + num2)


Line 1. 

In python, to declare variables you simply type the name, and does not have a specific data type declaration. In this case, "var" is used in Javascript. 

The correct code would be:

num1 = 5

num2 = 6

print(num1 + num2)

200

What are the two types of ML models we learned? (Excluding Neural Networks).

Decision Trees and Random Forests

200

What is the thing we use to get access to a Gemini Model?

The API Key

200

Where might we see a large neural network be used?

In companies such as OpenAI, DeepSeek, etc.

200

What do machine learning models analyze to make predictions? 

Patterns

300

How do you declare a function in python?

def name():

      asdfasdfasdf

300

What is the difference between underfitting and overfitting? 

Underfitting - Model is not specific enough / not trained enough, meaning it is not accurate.

Overfitting - Model is too accurate for only a specific thing, being inaccurate for other things.

300

What is a system instruction?

It is a specific instruction given to a model to possible talk a certain way, respond a certain way, etc.

300

What is emergent behavior?

It is complex, unintended behavior emerging from your model, that was never directly coded in. 

300

What are some concerns with the use of AI?

- Dependency on it

- Job loss

- etc.

400

What is the correct key word for exiting a loop?

break

400

What are the use of the Pandas Library in ML?

We can use data frames from this library to make predictions, analyze/read data, and manipulate data off of a dataset. 

There are many more uses, but these are the basics we covered. 

400

chat = model.start_chat(...)


If we wanted to create a model that remembered what we told it, or have a history, what would need to go in between the parenthesis?

history=[]

400

What is weight in neural networks?

"Weight" refers to the connections between each node. A higher weight will be more prioritized by the model than a smaller weight. 

400

“Artificial intelligence is the science of making machines do things that would require intelligence if done by humans.”

“Machines are becoming more intelligent every day, but can they ever truly understand us?”

“Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks.”

Which one is human?

“Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks.”

500

Lets say that there is a large list of fruits. Your goal is to go through this list, and find every fruit that starts with the letter "a." You want to add all of these to another list, and find the fruits that end with letter "a," adding them to another list. 

Finally, you print the fruits added to this last list. How would you code this? Your answer does not have to be code, just concepts!

Begin with creating a loop iterating through this list. Create a conditional to check if each value in this list starts with the letter "a." Add all these elements to another list, and create another for loop. This one will iterate through the new list, and filter the ones that end with "a," adding them to another list. 

Finally, print this list. 

500

What is model validation? 

It is a method to check the accuracy of the model. 

For example, we might only use half the data to train, and the other half to test checking if the model can determine unseen patterns, and confirming that the model does not make inaccurate patterns. 

500

genai.configure(api_key)

model = genai.GenerativeModel('gemini-1.5-flash')

chat = model.start_chat(history=[])

response = chat.send_massage("hello")

This program does not work as intended. What is the error?

It should be chat.send_message, not _massage

500

What is the LTH?

The lottery ticket hypothesis refers to a claim that huge models do not need all the data they are trained on, in fact they would work the same 99 percent of the time if almost all data except ones with high weight were removed. 

It can be referred to as models only needing the lottery ticket winner, but must purchase millions of tickets just to get the winner.

500

What does IDE stand for?

Integrated development environment