Avoid redoing long computation in a jupyter cell by pickling the result

cache_result.py

import pickle
import os

# Helper functions
def pickle_them(pickle_file, *objs):
    print(f"saving to {pickle_file}")
    with open(pickle_file, 'wb') as f:
        for obj in objs:
            pickle.dump(obj, f)

def read_pickle(pickle_file):
    if not os.path.exists(pickle_file):
        return None
    print(f"reading from {pickle_file}")
    data=[]
    with open(pickle_file, 'rb') as f:
        while True:
            try:
                data.append(pickle.load(f))
            except EOFError:
                break
    if len(data) == 1:
        return data[0]
    else:
        return tuple(data)

#
# In cell to apply
#
pickle_file="testpkl.pkl"
data = read_pickle(pickle_file)
if data is None: # if pickle file not exists, do computation and pickle result
    a = 123
    b = "abc"
    c = 0.12
    pickle_them(pickle_file, a, b, c)
else:
    a, b, c = data # if pickle exists, read values from it

print(a, b, c)