Case generators, case generation tools, and a solution-testing program for Tech with Tim's weekly challenge on Discord.
The abstract testing unit for any challenge.
Read More
- Bishop
- Easy Numbers
- T-Primes
- In 2120
- Balances the distribution of values in a random sample from range [
10^m,10^M). - Attenuates to the 90% skew towards the upper exponent by dividing the sampling into
k
- m (int/float): Lower bound for
xin10^x - M (int/float): Upper bound for
xin10^x - n (int): Sample size
- k (int): Number of bins to use
- **kwargs: Any keyword argument that is an iterable. Built into a generator that sends one value to each keyword on each iteration
The wrapper passes 3 arguments during each call in the following order that the generator function must be able to handle:
- The lower bound (absolute, not exponent) e.g. (100)
- The upper bound (absolute, not exponent) e.g. (10000.55)
- The sample size e.g. 1000
Basic implementation.
from gentools.wrap import binning
import random
# Generate 5 sub-samples in the range 10^0 to
# 10^10 and merge them so they sum to 1000 cases
@binning(m=0, M=10, n=100000, bins=10)
def generator(a, b, n):
return random.choices(range(int(a), int(b)), k=n)
cases = generator()Using iterables to map as arguments for each bin.
from gentools.wrap import binning
import random
# Random seeds for each bin's generation
seeds = [101, 392, 81]
# Shift random values
bias_map = [random.randint(-5, 6) for _ in range(4)]
# Text to print for each bin
def printable():
i = 0
while True:
yield f"Generating sample {i}"
i += 1
@binning(m=0, M=10, n=10000, bins=5, seed=seeds, stdout=printable(), bias=bias_map)
# Generate 5 sub-samples in the range 10^0 to
# 10^10 and merge them so they sum to 1000
def generator(a, b, n, seed=1, bias=0, stdout="..."):
random.seed(seed)
print(stdout)
print(f"\tSeed: {seed}")
print(f"\tBias: {bias}")
return [([x+bias], (x + bias) % 2 == 0) for x in random.choices(range(int(a), int(b)), k=n)]
cases = generator()Runtime output
Generating sample 1
Seed: 101
Bias: 4
Generating sample 2
Seed: 392
Bias: 2
Generating sample 3
Seed: 81
Bias: -1
Generating sample 4
Seed: 1
Bias: 2
Generating sample 5
Seed: 1
Bias: 0
Note the shift to the default when the iterable arguments have no more values to supply.
Like any other keyword argument, you can also package the iterables in a dictionary
from gentools.wrap import binning
import random
# Random seeds for each bin's generation
seeds = [101, 392, 81]
# Shift random values
bias_map = [random.randint(-5, 6) for _ in range(4)]
# Text to print for each bin
def printable():
i = 0
while True:
yield f"Generating sample {i}"
i += 1
kwarg_dict = {'seed': seeds, 'bias': bias_map, 'stdout': printable()}
@binning(m=0, M=10, n=10000, bins=5, **kwarg_dict)
# Generate 5 sub-samples in the range 10^0 to
# 10^10 and merge them so they sum to 1000
def generator(a, b, n, seed=1, bias=0, stdout="..."):
random.seed(seed)
print(stdout)
print(f"\tSeed: {seed}")
print(f"\tBias: {bias}")
return [([x+bias], (x + bias) % 2 == 0) for x in random.choices(range(int(a), int(b)), k=n)]
cases = generator()
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent