python_intro_prep_2024_edited.py

# -*- coding: utf-8 -*-
"""python_intro_prep_2024_edited.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/github/ZuckermanLab/coding2024/blob/main/python_intro_prep_2024_edited.ipynb

**It's OK to make a mistake**
"""

print ("my name is claire sherman")

"""**Printing words, numbers, math calculations**"""

print('hi there')
print('')  # leave a space
# by the way, a '#' symbol means the line is a comment - it won't get processed

a = 2
b = 4
c = a*b

print('a=', a, '  b=', b, '  c=', c)

# fancier way
print(f'a={a}  b={b}  c={c}')

# you try - print something by typing below and running the cell
a=5
b=10
c=30

d=a*(b-c)
print(d)

"""**Lists and loops**"""

dan_list = ['dog', 'cat', 'mouse']

for animal in dan_list:
  print(animal)

print('')

for x in dan_list:
  print(x)

print('')

for x in dan_list:
  print(animal)

# you try - make a list of numbers and print them.  then do a math operation on them too

list1 = [apple, banana, orage, pineapple]
x=[list1]
print(list1)

"""**A little more with loops**"""

int_list = range(5)  # range is a built-in python function
print(int_list)

print('')

for i in int_list:
  print(i)

print('')

my_name = 'Dan'
for i in int_list:
  print(f'My name is {my_name}')

print('')

my_name = 'Dan'
for i in int_list:
  print(i, i**2, i**3)  # raise to a power with **

"""**Functions**"""

# you try - make a loop over a range of integers and print i + i**2 for each
# or write a function to sum the numbers from 1 to 100

# functions are just lists of commands
# they can take multiple 'arguments' (in parentheses) as inputs
# they can 'return' an answer, which might be a number, string, list, ...

def text_repeater(text, reps):
  for i in range(reps):
    print(text)

text_repeater("oh it's you", 4)

text_repeater("No, it's me", 2)

def another_function():
  print('')
  print('this')
  print('that')
  print(2*3)

another_function()

def square_it(x):
  return x**2

print('')
z = 5
print( z, square_it(z))

# you try - make a function and call it (execute it)
# for example - write a function that prints some text 3 times, and then sums a list of numbers

"""**Math plots with random data using Numerical Python (numpy)**"""

# First, bring in the libraries of built-in functions we'll need
import numpy as np  # numpy (numerical python) - very useful for all things math
import matplotlib.pyplot as plt  # plotting tools

# numpy arrays
N = 15
print('\nNumpy arrays will be used many times in this lesson')
x_arr = np.arange(N)
y_arr = np.random.rand(N) # random
print(x_arr)

# simple plots
plt.scatter(x_arr, y_arr) # plots points, taking x from first array and y from second
plt.title('Sample plot')
plt.xlabel('index of array')
plt.ylabel('value of array element')
plt.show()  # show it now in output of cell

# plot the same data, but change labels

# new cell

# second set of arrays
# check if there are append commands in numpy and start from empty array?
x_arr = np.arange(0, 10, 0.1)
y_arr = np.zeros( len(x_arr) )  # initialize

print('x_arr:', x_arr) # print stuff to see what it looks like
print('')  # print a space to make it easy to see different arays
print('y_arr:', y_arr)

# initialize other arrays for comparison
y_arr_2 = np.zeros( len(x_arr) )  # initialize
y_arr_3 = np.zeros( len(x_arr) )
y_arr_4 = np.zeros( len(x_arr) )

i = 0
for x in x_arr:
  #print(x, i)
  rando = np.random.rand() # random number between zero and one
  y_arr[i] = 3*x + 5*( rando - 0.5 )  # line of slope 3 with random 'noise' scaled by 5
  y_arr_2[i] = 2*x  # a simple line of slope 2
  y_arr_3[i] = 3*x  # slope 3
  y_arr_4[i] = 4*x  # slope 4
  i = i+1

print('')  # print a space to make it easy to see different arays
print('y_arr:', y_arr)
print('')
print('y_arr_2:', y_arr_2)

# plot the various arrays

# plot new arrays
plt.scatter(x_arr, y_arr_4, label='slope=4')
plt.scatter(x_arr, y_arr_3, label='slope=3')
plt.scatter(x_arr, y_arr_2, label='slope=2')
plt.scatter(x_arr, y_arr, label='noisy data')
plt.title('Data with noise and some lines with different slopes')
plt.xlabel('x')
plt.ylabel('y')
plt.legend(loc="upper left")
plt.show()  # show it now in output of cell

"""**Fitting Data**"""

def error_of_fit_abs( x_array, y_array, slope): # function to compute error between data and straight line
  error = 0 # initialize, then add to this
  for i in range( len(x_array) ):
    x = x_array[i]
    y = y_array[i]
    error += abs(y - slope*x)  # the compound '+=' tells python to add whatever's on the right to the left variable
    #print(x, y, slope*x, error)
  return error

print(2, error_of_fit_abs(x_arr, y_arr, 2))
print(3, error_of_fit_abs(x_arr, y_arr, 3))
print(4, error_of_fit_abs(x_arr, y_arr, 4))
print('')

# make a loop for finer-grained 'grid search'
slopes_array = np.arange(1.0, 5.0, 0.1)  # for plotting
n_slopes = len(slopes_array) # number of elements in array

x_slopes = np.zeros( n_slopes )
y_fits = np.zeros( n_slopes )

# make arrays for plotting
i = 0
for slope in slopes_array:
  x_slopes[i] = slope
  y_fits[i] = error_of_fit_abs(x_arr, y_arr, slope)
  i = i+1

plt.scatter(x_slopes, y_fits, label='Error in fit')
plt.title("'Grid search' in one dimension")
plt.xlabel('Slope')
plt.ylabel('Error in Fit')
plt.legend(loc="upper left")
plt.show()  # show it now in output of cell



"""2D loop ... suggests 2D grid search"""