Related
I'd like to evaluate np.random.dirichlet with large dimension as quickly as possible. More precisely, I'd like a function approximating the below by at least 10 times faster. Empirically, I observed that small-dimension-version of this function outputs one or two entries that have the order of 0.1, and every other entries are so small that they are immaterial. But this observation isn't based on any rigorous assessment. The approximation doesn't need to be so accurate, but I want something not too crude, as I'm using this noise for MCTS.
def g():
np.random.dirichlet([0.03]*4840)
>>> timeit.timeit(g,number=1000)
0.35117408499991143
Assuming your alpha is fixed over components and used for many iterations you could tabulate the ppf of the corresponding gamma distribution. This is probably available as scipy.stats.gamma.ppf but we can also use scipy.special.gammaincinv. This function seems rather slow, so this is a siginificant upfront investment.
Here is a crude implementation of the general idea:
import numpy as np
from scipy import special
class symm_dirichlet:
def __init__(self, alpha, resolution=2**16):
self.alpha = alpha
self.resolution = resolution
self.range, delta = np.linspace(0, 1, resolution,
endpoint=False, retstep=True)
self.range += delta / 2
self.table = special.gammaincinv(self.alpha, self.range)
def draw(self, n_sampl, n_comp, interp='nearest'):
if interp != 'nearest':
raise NotImplementedError
gamma = self.table[np.random.randint(0, self.resolution,
(n_sampl, n_comp))]
return gamma / gamma.sum(axis=1, keepdims=True)
import time, timeit
t0 = time.perf_counter()
X = symm_dirichlet(0.03)
t1 = time.perf_counter()
print(f'Upfront cost {t1-t0:.3f} sec')
print('Running cost per 1000 samples of width 4840')
print('tabulated {:3f} sec'.format(timeit.timeit(
'X.draw(1, 4840)', number=1000, globals=globals())))
print('np.random.dirichlet {:3f} sec'.format(timeit.timeit(
'np.random.dirichlet([0.03]*4840)', number=1000, globals=globals())))
Sample output:
Upfront cost 13.067 sec
Running cost per 1000 samples of width 4840
tabulated 0.059365 sec
np.random.dirichlet 0.980067 sec
Better check whether it is roughly correct:
I am implementing a Kalman filter for the first time to get voltage values from a source. It works and it stabilizes at the source voltage value but if then the source changes the voltage the filter doesn't adapt to the new value.
I use 3 steps:
Get the Kalman gain
KG = previous_error_in_estimate / ( previous_error_in_estimate + Error_in_measurement )
Get current estimation
Estimation = previous_estimation + KG*[measurement - previous_estimation]
Calculate the error in estimate
Error_in_estimate = [1-KG]*previous_error_in_estimate
The thing is that, as 0 <= KG <= 1, Error_in_estimate decreases more and more and that makes KG to also decrease more and more ( error_in_measurement is a constant ), so at the end the estimation only depends on the previous estimation and the current measurement is not taken into account.
This prevents the filter from adapt himself to measurement changes.
How can I do to make that happen?
Thanks
EDIT:
Answering to Claes:
I am not sure that the Kalman filter is valid for my problem since I don't have a system model, I just have a bunch of readings from a quite noisy sensor measuring a not very predictable variable.
To keep things simple, imagine reading a potentiometer ( a variable resistor ) changed by the user, you can't predict or model the user's behavior.
I have implemented a very basic SMA ( Simple Moving Average ) algorithm and I was wondering if there is a better way to do it.
Is the Kalman filter valid for a problem like this?
If not, what would you suggest?
2ND EDIT
Thanks to Claes for such an useful information
I have been doing some numerical tests in MathLab (with no real data yet) and doing the convolution with a Gaussian filter seems to give the most accurate result.
With the Kalman filter I don't know how to estimate the process and measurement variances, is there any method for that?. Only when I decrease quite a lot the measurement variance the kalman filter seems to adapt. In the previous image the measurement variance was R=0.1^2 (the one in the original example). This is the same test with R=0.01^2
Of course, these are MathLab tests with no real data. Tomorrow I will try to implement this filters in the real system with real data and see if I can get similar results
A simple MA filter is probably sufficient for your example. If you would like to use the Kalman filter there is a great example at the SciPy cookbook
I have modified the code to include a step change so you can see the convergence.
# Kalman filter example demo in Python
# A Python implementation of the example given in pages 11-15 of "An
# Introduction to the Kalman Filter" by Greg Welch and Gary Bishop,
# University of North Carolina at Chapel Hill, Department of Computer
# Science, TR 95-041,
# http://www.cs.unc.edu/~welch/kalman/kalmanIntro.html
# by Andrew D. Straw
import numpy as np
import matplotlib.pyplot as plt
plt.rcParams['figure.figsize'] = (10, 8)
# intial parameters
n_iter = 400
sz = (n_iter,) # size of array
x1 = -0.37727*np.ones(n_iter/2) # truth value 1
x2 = -0.57727*np.ones(n_iter/2) # truth value 2
x = np.concatenate((x1,x2),axis=0)
z = x+np.random.normal(0,0.1,size=sz) # observations (normal about x, sigma=0.1)
Q = 1e-5 # process variance
# allocate space for arrays
xhat=np.zeros(sz) # a posteri estimate of x
P=np.zeros(sz) # a posteri error estimate
xhatminus=np.zeros(sz) # a priori estimate of x
Pminus=np.zeros(sz) # a priori error estimate
K=np.zeros(sz) # gain or blending factor
R = 0.1**2 # estimate of measurement variance, change to see effect
# intial guesses
xhat[0] = 0.0
P[0] = 1.0
for k in range(1,n_iter):
# time update
xhatminus[k] = xhat[k-1]
Pminus[k] = P[k-1]+Q
# measurement update
K[k] = Pminus[k]/( Pminus[k]+R )
xhat[k] = xhatminus[k]+K[k]*(z[k]-xhatminus[k])
P[k] = (1-K[k])*Pminus[k]
plt.figure()
plt.plot(z,'k+',label='noisy measurements')
plt.plot(xhat,'b-',label='a posteri estimate')
plt.plot(x,color='g',label='truth value')
plt.legend()
plt.title('Estimate vs. iteration step', fontweight='bold')
plt.xlabel('Iteration')
plt.ylabel('Voltage')
And the output is:
I am using Newton Raphson +successive Substitute algorithm to perform flash calculation(chemical process simulation).
The algorithm can converge well when the input is in low precision like 0.1, but when the number precision is increased to 0.11111 or 0.99999. The algorithm will not converge.
When I am using the quasi newton method with BFGS update, the same problem occurs again. How can we decrease the sensitivity of the code to the numerical precision?
Here is a simple example using matlab to solve the Rachford-Rice equation. When the comp_overall=[0.9,1-0.9], it converges well. However, when the number precision increase to like[0.99999,1-0.99999]. It will not converge.
K=[0.053154011443159 34.234731216532658],
comp_overall= [0.99999 1- 0.99999], phi=0.5; %initial values
epsilon = 1.0;
iter1 = 1;
while (epsilon >=1.e-05)
rc=0.0;
drc=0.0;
for i=1:2
% Rachford-Rice Equation
rc = comp_overall(i)*(K(i)-1.0)/(1.0+phi*(K(i)-1.0))+rc;
% Derivative
drc = comp_overall(i)*(K(i)-1.0)^2/(1.0+phiK(i)-1.0))^2+drc;
end
% Newton-Raphson
phi1 = phi +0.01 (rc / drc);
epsilon = abs( (phi1-phi)/phi );
% Convergence
phi = phi1;
iter1=iter1+1;
end
The Newton–Raphson method relies on the function being differentiable between any two consecutive approximations. Depending on the choice of the initial value, this may not be the case for z₁ = 0.99999. Let's look at the graph of the Rachford-Rice function:
The root of this function is φ₀ ≈ –0.0300781429 and the nearest point of discontinuity is –1/(K₂-1) ≈ –0.0300890052. They are close enough for the Newton–Raphson method to overshoot, to jump over that discontinuity.
For example:
φ₁ = –0.025
f(φ₁) ≈ -0.9229770571
f'(φ₁) ≈ 1.2416569960
φ₂ = φ₁ + 0.01 * f(φ₁) / f'(φ₁) ≈ -0.0324334302
φ₂ lies to the left of the discontinuity, so the following steps will be away from, not towards the root.
φ₃ = -0.0358986759 < φ₂
What can be done about it:
When the algorithm fails to converge, repeat it with smaller steps. For example, start with the coefficient 0.01 (as it is now) and decrease it 10 times after every failure.
Detect overshoots. On each iteration check if there is a discontinuity point (–1/(Kᵢ-1)) between the current approximation and the previous one. When it happens, discard the current approximation, decrease the coefficient and continue.
Limit the scope of the search. Are solutions outside of [0, 1] physically meaningful? If not, you can stop once the approximated value falls out of that range.
Use different method. The function is monotonic on any interval between two consecutive discontinuity points, so you can perform binary search on each such interval. It will be both faster and more robust than the Newton–Raphson method.
I was persuaded some time ago to drop my comfortable matlab programming and start programming in Julia. I have been working for a long with neural networks and I thought that, now with Julia, I could get things done faster by parallelising the calculation of the gradient.
The gradient need not be calculated on the entire dataset in one go; instead one can split the calculation. For instance, by splitting the dataset in parts, we can calculate a partial gradient on each part. The total gradient is then calculated by adding up the partial gradients.
Though, the principle is simple, when I parallelise with Julia I get a performance degradation, i.e. one process is faster then two processes! I am obviously doing something wrong... I have consulted other questions asked in the forum but I could still not piece together an answer. I think my problem lies in that there is a lot of unnecessary data moving going on, but I can't fix it properly.
In order to avoid posting messy neural network code, I am posting below a simpler example that replicates my problem in the setting of linear regression.
The code-block below creates some data for a linear regression problem. The code explains the constants, but X is the matrix containing the data inputs. We randomly create a weight vector w which when multiplied with X creates some targets Y.
######################################
## CREATE LINEAR REGRESSION PROBLEM ##
######################################
# This code implements a simple linear regression problem
MAXITER = 100 # number of iterations for simple gradient descent
N = 10000 # number of data items
D = 50 # dimension of data items
X = randn(N, D) # create random matrix of data, data items appear row-wise
Wtrue = randn(D,1) # create arbitrary weight matrix to generate targets
Y = X*Wtrue # generate targets
The next code-block below defines functions for measuring the fitness of our regression (i.e. the negative log-likelihood) and the gradient of the weight vector w:
####################################
## DEFINE FUNCTIONS ##
####################################
#everywhere begin
#-------------------------------------------------------------------
function negative_loglikelihood(Y,X,W)
#-------------------------------------------------------------------
# number of data items
N = size(X,1)
# accumulate here log-likelihood
ll = 0
for nn=1:N
ll = ll - 0.5*sum((Y[nn,:] - X[nn,:]*W).^2)
end
return ll
end
#-------------------------------------------------------------------
function negative_loglikelihood_grad(Y,X,W, first_index,last_index)
#-------------------------------------------------------------------
# number of data items
N = size(X,1)
# accumulate here gradient contributions by each data item
grad = zeros(similar(W))
for nn=first_index:last_index
grad = grad + X[nn,:]' * (Y[nn,:] - X[nn,:]*W)
end
return grad
end
end
Note that the above functions are on purpose not vectorised! I choose not to vectorise, as the final code (the neural network case) will also not admit any vectorisation (let us not get into more details regarding this).
Finally, the code-block below shows a very simple gradient descent that tries to recover the parameter weight vector w from the given data Y and X:
####################################
## SOLVE LINEAR REGRESSION ##
####################################
# start from random initial solution
W = randn(D,1)
# learning rate, set here to some arbitrary small constant
eta = 0.000001
# the following for-loop implements simple gradient descent
for iter=1:MAXITER
# get gradient
ref_array = Array(RemoteRef, nworkers())
# let each worker process part of matrix X
for index=1:length(workers())
# first index of subset of X that worker should work on
first_index = (index-1)*int(ceil(N/nworkers())) + 1
# last index of subset of X that worker should work on
last_index = min((index)*(int(ceil(N/nworkers()))), N)
ref_array[index] = #spawn negative_loglikelihood_grad(Y,X,W, first_index,last_index)
end
# gather the gradients calculated on parts of matrix X
grad = zeros(similar(W))
for index=1:length(workers())
grad = grad + fetch(ref_array[index])
end
# now that we have the gradient we can update parameters W
W = W + eta*grad;
# report progress, monitor optimisation
#printf("Iter %d neg_loglikel=%.4f\n",iter, negative_loglikelihood(Y,X,W))
end
As is hopefully visible, I tried to parallelise the calculation of the gradient in the easiest possible way here. My strategy is to break the calculation of the gradient in as many parts as available workers. Each worker is required to work only on part of matrix X, which part is specified by first_index and last_index. Hence, each worker should work with X[first_index:last_index,:]. For instance, for 4 workers and N = 10000, the work should be divided as follows:
worker 1 => first_index = 1, last_index = 2500
worker 2 => first_index = 2501, last_index = 5000
worker 3 => first_index = 5001, last_index = 7500
worker 4 => first_index = 7501, last_index = 10000
Unfortunately, this entire code works faster if I have only one worker. If add more workers via addprocs(), the code runs slower. One can aggravate this issue by create more data items, for instance use instead N=20000.
With more data items, the degradation is even more pronounced.
In my particular computing environment with N=20000 and one core, the code runs in ~9 secs. With N=20000 and 4 cores it takes ~18 secs!
I tried many many different things inspired by the questions and answers in this forum but unfortunately to no avail. I realise that the parallelisation is naive and that data movement must be the problem, but I have no idea how to do it properly. It seems that the documentation is also a bit scarce on this issue (as is the nice book by Ivo Balbaert).
I would appreciate your help as I have been stuck for quite some while with this and I really need it for my work. For anyone wanting to run the code, to save you the trouble of copying-pasting you can get the code here.
Thanks for taking the time to read this very lengthy question! Help me turn this into a model answer that anyone new in Julia can then consult!
I would say that GD is not a good candidate for parallelizing it using any of the proposed methods: either SharedArray or DistributedArray, or own implementation of distribution of chunks of data.
The problem does not lay in Julia, but in the GD algorithm.
Consider the code:
Main process:
for iter = 1:iterations #iterations: "the more the better"
δ = _gradient_descent_shared(X, y, θ)
θ = θ - α * (δ/N)
end
The problem is in the above for-loop which is a must. No matter how good _gradient_descent_shared is, the total number of iterations kills the noble concept of the parallelization.
After reading the question and the above suggestion I've started implementing GD using SharedArray. Please note, I'm not an expert in the field of SharedArrays.
The main process parts (simple implementation without regularization):
run_gradient_descent(X::SharedArray, y::SharedArray, θ::SharedArray, α, iterations) = begin
N = length(y)
for iter = 1:iterations
δ = _gradient_descent_shared(X, y, θ)
θ = θ - α * (δ/N)
end
θ
end
_gradient_descent_shared(X::SharedArray, y::SharedArray, θ::SharedArray, op=(+)) = begin
if size(X,1) <= length(procs(X))
return _gradient_descent_serial(X, y, θ)
else
rrefs = map(p -> (#spawnat p _gradient_descent_serial(X, y, θ)), procs(X))
return mapreduce(r -> fetch(r), op, rrefs)
end
end
The code common to all workers:
#= Returns the range of indices of a chunk for every worker on which it can work.
The function splits data examples (N rows into chunks),
not the parts of the particular example (features dimensionality remains intact).=#
#everywhere function _worker_range(S::SharedArray)
idx = indexpids(S)
if idx == 0
return 1:size(S,1), 1:size(S,2)
end
nchunks = length(procs(S))
splits = [round(Int, s) for s in linspace(0,size(S,1),nchunks+1)]
splits[idx]+1:splits[idx+1], 1:size(S,2)
end
#Computations on the chunk of the all data.
#everywhere _gradient_descent_serial(X::SharedArray, y::SharedArray, θ::SharedArray) = begin
prange = _worker_range(X)
pX = sdata(X[prange[1], prange[2]])
py = sdata(y[prange[1],:])
tempδ = pX' * (pX * sdata(θ) .- py)
end
The data loading and training. Let me assume that we have:
features in X::Array of the size (N,D), where N - number of examples, D-dimensionality of the features
labels in y::Array of the size (N,1)
The main code might look like this:
X=[ones(size(X,1)) X] #adding the artificial coordinate
N, D = size(X)
MAXITER = 500
α = 0.01
initialθ = SharedArray(Float64, (D,1))
sX = convert(SharedArray, X)
sy = convert(SharedArray, y)
X = nothing
y = nothing
gc()
finalθ = run_gradient_descent(sX, sy, initialθ, α, MAXITER);
After implementing this and run (on 8-cores of my Intell Clore i7) I got a very slight acceleration over serial GD (1-core) on my training multiclass (19 classes) training data (715 sec for serial GD / 665 sec for shared GD).
If my implementation is correct (please check this out - I'm counting on that) then parallelization of the GD algorithm is not worth of that. Definitely you might get better acceleration using stochastic GD on 1-core.
If you want to reduce the amount of data movement, you should strongly consider using SharedArrays. You could preallocate just one output vector, and pass it as an argument to each worker. Each worker sets a chunk of it, just as you suggested.
Let's say that I know the probability of a "success" is P. I run the test N times, and I see S successes. The test is akin to tossing an unevenly weighted coin (perhaps heads is a success, tails is a failure).
I want to know the approximate probability of seeing either S successes, or a number of successes less likely than S successes.
So for example, if P is 0.3, N is 100, and I get 20 successes, I'm looking for the probability of getting 20 or fewer successes.
If, on the other hadn, P is 0.3, N is 100, and I get 40 successes, I'm looking for the probability of getting 40 our more successes.
I'm aware that this problem relates to finding the area under a binomial curve, however:
My math-fu is not up to the task of translating this knowledge into efficient code
While I understand a binomial curve would give an exact result, I get the impression that it would be inherently inefficient. A fast method to calculate an approximate result would suffice.
I should stress that this computation has to be fast, and should ideally be determinable with standard 64 or 128 bit floating point computation.
I'm looking for a function that takes P, S, and N - and returns a probability. As I'm more familiar with code than mathematical notation, I'd prefer that any answers employ pseudo-code or code.
Exact Binomial Distribution
def factorial(n):
if n < 2: return 1
return reduce(lambda x, y: x*y, xrange(2, int(n)+1))
def prob(s, p, n):
x = 1.0 - p
a = n - s
b = s + 1
c = a + b - 1
prob = 0.0
for j in xrange(a, c + 1):
prob += factorial(c) / (factorial(j)*factorial(c-j)) \
* x**j * (1 - x)**(c-j)
return prob
>>> prob(20, 0.3, 100)
0.016462853241869437
>>> 1-prob(40-1, 0.3, 100)
0.020988576003924564
Normal Estimate, good for large n
import math
def erf(z):
t = 1.0 / (1.0 + 0.5 * abs(z))
# use Horner's method
ans = 1 - t * math.exp( -z*z - 1.26551223 +
t * ( 1.00002368 +
t * ( 0.37409196 +
t * ( 0.09678418 +
t * (-0.18628806 +
t * ( 0.27886807 +
t * (-1.13520398 +
t * ( 1.48851587 +
t * (-0.82215223 +
t * ( 0.17087277))))))))))
if z >= 0.0:
return ans
else:
return -ans
def normal_estimate(s, p, n):
u = n * p
o = (u * (1-p)) ** 0.5
return 0.5 * (1 + erf((s-u)/(o*2**0.5)))
>>> normal_estimate(20, 0.3, 100)
0.014548164531920815
>>> 1-normal_estimate(40-1, 0.3, 100)
0.024767304545069813
Poisson Estimate: Good for large n and small p
import math
def poisson(s,p,n):
L = n*p
sum = 0
for i in xrange(0, s+1):
sum += L**i/factorial(i)
return sum*math.e**(-L)
>>> poisson(20, 0.3, 100)
0.013411150012837811
>>> 1-poisson(40-1, 0.3, 100)
0.046253037645840323
I was on a project where we needed to be able to calculate the binomial CDF in an environment that didn't have a factorial or gamma function defined. It took me a few weeks, but I ended up coming up with the following algorithm which calculates the CDF exactly (i.e. no approximation necessary). Python is basically as good as pseudocode, right?
import numpy as np
def binomial_cdf(x,n,p):
cdf = 0
b = 0
for k in range(x+1):
if k > 0:
b += + np.log(n-k+1) - np.log(k)
log_pmf_k = b + k * np.log(p) + (n-k) * np.log(1-p)
cdf += np.exp(log_pmf_k)
return cdf
Performance scales with x. For small values of x, this solution is about an order of magnitude faster than scipy.stats.binom.cdf, with similar performance at around x=10,000.
I won't go into a full derivation of this algorithm because stackoverflow doesn't support MathJax, but the thrust of it is first identifying the following equivalence:
For all k > 0, sp.misc.comb(n,k) == np.prod([(n-k+1)/k for k in range(1,k+1)])
Which we can rewrite as:
sp.misc.comb(n,k) == sp.misc.comb(n,k-1) * (n-k+1)/k
or in log space:
np.log( sp.misc.comb(n,k) ) == np.log(sp.misc.comb(n,k-1)) + np.log(n-k+1) - np.log(k)
Because the CDF is a summation of PMFs, we can use this formulation to calculate the binomial coefficient (the log of which is b in the function above) for PMF_{x=i} from the coefficient we calculated for PMF_{x=i-1}. This means we can do everything inside a single loop using accumulators, and we don't need to calculate any factorials!
The reason most of the calculations are done in log space is to improve the numerical stability of the polynomial terms, i.e. p^x and (1-p)^(1-x) have the potential to be extremely large or extremely small, which can cause computational errors.
EDIT: Is this a novel algorithm? I've been poking around on and off since before I posted this, and I'm increasingly wondering if I should write this up more formally and submit it to a journal.
I think you want to evaluate the incomplete beta function.
There's a nice implementation using a continued fraction representation in "Numerical Recipes In C", chapter 6: 'Special Functions'.
I can't totally vouch for the efficiency, but Scipy has a module for this
from scipy.stats.distributions import binom
binom.cdf(successes, attempts, chance_of_success_per_attempt)
An efficient and, more importantly, numerical stable algorithm exists in the domain of Bezier Curves used in Computer Aided Design. It is called de Casteljau's algorithm used to evaluate the Bernstein Polynomials used to define Bezier Curves.
I believe that I am only allowed one link per answer so start with Wikipedia - Bernstein Polynomials
Notice the very close relationship between the Binomial Distribution and the Bernstein Polynomials. Then click through to the link on de Casteljau's algorithm.
Lets say I know the probability of throwing a heads with a particular coin is P.
What is the probability of me throwing
the coin T times and getting at least
S heads?
Set n = T
Set beta[i] = 0 for i = 0, ... S - 1
Set beta[i] = 1 for i = S, ... T
Set t = p
Evaluate B(t) using de Casteljau
or at most S heads?
Set n = T
Set beta[i] = 1 for i = 0, ... S
Set beta[i] = 0 for i = S + 1, ... T
Set t = p
Evaluate B(t) using de Casteljau
Open source code probably exists already. NURBS Curves (Non-Uniform Rational B-spline Curves) are a generalization of Bezier Curves and are widely used in CAD. Try openNurbs (the license is very liberal) or failing that Open CASCADE (a somewhat less liberal and opaque license). Both toolkits are in C++, though, IIRC, .NET bindings exist.
If you are using Python, no need to code it yourself. Scipy got you covered:
from scipy.stats import binom
# probability that you get 20 or less successes out of 100, when p=0.3
binom.cdf(20, 100, 0.3)
>>> 0.016462853241869434
# probability that you get exactly 20 successes out of 100, when p=0.3
binom.pmf(20, 100, 0.3)
>>> 0.0075756449257260777
From the portion of your question "getting at least S heads" you want the cummulative binomial distribution function. See http://en.wikipedia.org/wiki/Binomial_distribution for the equation, which is described as being in terms of the "regularized incomplete beta function" (as already answered). If you just want to calculate the answer without having to implement the entire solution yourself, the GNU Scientific Library provides the function: gsl_cdf_binomial_P and gsl_cdf_binomial_Q.
The DCDFLIB Project has C# functions (wrappers around C code) to evaluate many CDF functions, including the binomial distribution. You can find the original C and FORTRAN code here. This code is well tested and accurate.
If you want to write your own code to avoid being dependent on an external library, you could use the normal approximation to the binomial mentioned in other answers. Here are some notes on how good the approximation is under various circumstances. If you go that route and need code to compute the normal CDF, here's Python code for doing that. It's only about a dozen lines of code and could easily be ported to any other language. But if you want high accuracy and efficient code, you're better off using third party code like DCDFLIB. Several man-years went into producing that library.
Try this one, used in GMP. Another reference is this.
import numpy as np
np.random.seed(1)
x=np.random.binomial(20,0.6,10000) #20 flips of coin,probability of
heads percentage and 10000 times
done.
sum(x>12)/len(x)
The output is 41% of times we got 12 heads.