def cosine_similarity(v, w):
return dotProduct(v, w) / math.sqrt(dotProduct(v, v) * dotProduct(w, w))
def transformVector(v, components):
return [dotProduct(v, w) for w in components]
def matrix_product_entry(A, B, i, j):
return dotProduct(getRow(A, i), getCol(B, j))
def __directionalVarianceGradiant(xi, w):
projectionLength = dotProduct(xi, direction(w))
return [2 * projectionLength * xij for xij in xi]
def project(v, w):
projectionLength = dotProduct(v, w)
return scalarMultiply(projectionLength, w)
def matrix_product_entry(A, B, i, j):
return dotProduct(getRow(A, i), getCol(B, j))
def __directionalVarianceI(xi, w):
return dotProduct(xi, direction(w)) ** 2
def transformVector(v, components):
return [dotProduct(v, w) for w in components]
def covariance(x, y):
n = len(x)
return dotProduct(de_mean(x), de_mean(y)) / (n - 1)
def project(v, w):
projectionLength = dotProduct(v, w)
return scalarMultiply(projectionLength, w)
def __directionalVarianceGradiant(xi, w):
projectionLength = dotProduct(xi, direction(w))
return [2 * projectionLength * xij for xij in xi]
def __directionalVarianceI(xi, w):
return dotProduct(xi, direction(w))**2
def cosine_similarity(v, w):
return dotProduct(v, w) / math.sqrt(dotProduct(v, v) * dotProduct(w, w))
def covariance(x, y):
n = len(x)
return dotProduct(de_mean(x), de_mean(y)) / (n - 1)
