Python Numbers
Learn Python number types: int, float, and complex. Then see how they show up in AI / ML.
Why it matters: Scores, learning rates, batch sizes, and seeds are all numbers.
Basics
Python Numbers
There are three numeric types in Python:
intfloatcomplex
Variables of numeric types are created when you assign a value.
To check the type of any object, use the type() function.
Tiny example
Create one int, one float, and one complex number.
x = 1 # int
y = 2.8 # float
z = 1j # complex
print(type(x))
print(type(y))
print(type(z))Basics
Int
Int, or integer, is a whole number, positive or negative, without decimals.
In Python, integers can be very large. You do not need to worry about a fixed size limit.
Tiny example
Integers are whole numbers. No decimal point.
x = 1
y = 35656222554887711
z = -3255522
print(type(x))
print(type(y))
print(type(z))Basics
Float
Float, or floating point number, is a number with one or more decimals.
Float can also use scientific notation with e for power of 10:
x = 35e3
y = 12E4
z = -87.7e100
print(type(x))
Common mistake: writing a learning rate as the string "2e-5" instead of the float 2e-5.
Tiny example
Floats have a decimal point.
x = 1.10
y = 1.0
z = -35.59
print(type(x))
print(type(y))
print(type(z))Basics
Useful Number Operators
Useful operators:
/normal divide (always gives float)//floor divide (whole number result)%remainder**power
Also useful: abs(), round(), min(), max(), sum().
Tiny example
Try divide, floor divide, and remainder.
print(7 / 2)
print(7 // 2)
print(7 % 2)
print(2 ** 10)Then AI / ML
AI / ML examples
Same idea, used in real AI work. Press Try in lab to run it.
AI / ML example
AI / ML: learning rate and batch size
Same number types, now used like a tiny training config.
In AI / ML: Learning rates are floats. Batch sizes and epochs are ints.
example.py
Python
learning_rate = 2e-5 # float
batch_size = 32 # int
epochs = 3 # int
print(type(learning_rate))
print(1000 // batch_size)AI / ML example 2
AI / ML: keep or drop a score
A float score plus a simple comparison.
In AI / ML: Classifiers often keep predictions only when score >= threshold.
example.py
Python
score = 0.82
threshold = 0.7
if score >= threshold:
print("keep")
else:
print("drop")Takeaways
- 1Python has three number types: int, float, and complex.
- 2Use type() to check what kind of number you have.
- 3In AI / ML, scores and rates are floats; counts are ints.