import operations as ops
import fit
import exptree
import random
import functools
import itertools
x = sympy.symbols('x')
''' Note: prototype functions CAN NOT use ci (for int i) constant names'''
j, k, l, m, n, o, p = sympy.symbols('j k l m n o p')
'''List of function prototypes and their constants'''
basic_functions = [x, x**2, ops.sqrt(x + j), ops.log(x + k), ops.exp(x*l), ops.sin(x*m +n), ops.atan(x*o + p)]
def cross_multiply(l, k):
'''Given a list, return a list of all possible product combinations
(Up to k)
'''
result = l
for i in range(2, k + 1):
if i > len(l): break
combos = list(itertools.combinations(l, i))
for j in range(0, len(combos)):
combos[j] = functools.reduce(lambda x, y: x*y, combos[j])
result = result + combos
if operator == 1:
result = addition(num1, num2)
elif operator == 2:
result = subtraction(num1, num2)
elif operator == 3:
result = multiplication(num1, num2)
elif operator == 4:
result = division(num1, num2)
elif operator == 5:
result = integer_division(num1, num2)
elif operator == 6:
result = power(num1, num2)
elif operator == 7:
result = modulo(num1, num2)
elif operator == 8:
result = log(num1, num2)
elif operator == 9:
result = sigmoid(num1 + num2)
elif operator == 10:
result = rand_between(num1, num2)
elif operator == 11:
result = hcf(num1, num2)
elif operator == 12:
result = factorial(num1)
elif operator == 13:
result = exponential(num1)
elif operator == 14:
result = Sine()
elif operator == 15:
result = Cosine()
elif operator == 16:
# Create a new graph
Graph().as_default()
X = Placeholder()
c = Placeholder()
# Create a weight matrix for 2 outout classes:
# One with a weight vector (1, 1) for blue and one with a
# weight vector (-1, -1) for red
W = Variable([[1, -1], [1, -1]])
b = Variable([0, 0])
p = softmax(add(matmul(X, W), b))
# Cross-entropy loss
J = negative(reduce_sum(reduce_sum(multiply(c, log(p)), axis=1)))
# Create red points centered at (-2, -2)
red_points = np.random.randn(50, 2) - 2 * np.ones((50, 2))
# Create blue points centered at (2, 2)
blue_points = np.random.randn(50, 2) + 2 * np.ones((50, 2))
session = Session()
print(
session.run(
J, {
X: np.concatenate((blue_points, red_points)),
c: [[1, 0]] * len(blue_points) + [[0, 1]] * len(red_points)
}))
def cross_entropy(logits: op.Tensor, label: op.Tensor):
assert logits.shape == label.shape
return op.const(-1) * reduce_sum(op.log(logits) * label)
def sparse_cross_entropy(logits: op.Tensor, labels: op.Tensor):
assert len(logits.shape) == 1, 'but now is %s' % logits.shape
assert labels.shape == tuple(), 'label.shape should be a number'
logits = op.log(logits)
return op.const(-1) * op.Index(logits, labels)
x_data = [1, 2, 3, 4, 5]
y_data = [3, 9, 19, 33, 51]
expr = x
print("'fit' just x (no free parameters)")
popt, err = sym_fit(expr, [], x_data, y_data)
print("Fit to y_data")
print(popt)
print(err)
print("---")
x_data = range(-10, 10)
y_data = range(-10, 10)
expr = ops.log(x + c0) + c1
print("Attempting to c1 * log(x + c0) + c2 (multiple guesses)")
popt, err = sym_fit(expr, [c0, c1], x_data, y_data, 12)
print("Fit to y_data")
print(popt)
print(err)
print("")
x_data = [-1, 1, 2, 3]
y_data = [1, 2, 3, 4]
expr = ops.log(x) + c0
print("Attempting to log(x) + c0 (should fail on this domain)")
