def generate(data):
sf = 2
# Matrix shape
M = 3
myNumber = 2.148233
# Generating the orthogonal matrix U
# (numbers rounded with 2 decimal digits)
X = np.random.rand(M, M)
Q,R = sla.qr(X)
U = np.around(Q, sf + 1)
b = np.random.rand(M)
bc = b.reshape((M, 1))
br = b.reshape(1, M)
data['params']['sf'] = sf
data['params']['M'] = M
data['params']['U'] = pl.to_json(U)
data['params']['myNumber'] = pl.to_json(myNumber)
data['params']['b'] = pl.to_json(br)
data['params']['c'] = pl.to_json(bc)
return data
def generate(data):
sf = 2
# Matrix shape
M = 3
myNumber = 2.148233
# Generating the orthogonal matrix U
# (numbers rounded with 2 decimal digits)
X = np.random.rand(M, M)
Q, R = sla.qr(X)
U = np.around(Q, sf + 1)
b = np.random.rand(M)
bc = b.reshape((M, 1))
br = b.reshape(1, M)
data['params']['sf'] = sf
data['params']['M'] = M
data['params']['U'] = pl.to_json(U)
data['params']['myNumber'] = pl.to_json(myNumber)
data['params']['b'] = pl.to_json(br)
data['params']['c'] = pl.to_json(bc)
return data
def generate(data):
# Dimensions
nInnerMin = 2
nMax = 5
nRows = np.random.randint(1,nMax+1)
nInner = np.random.randint(nInnerMin,nMax+1)
nCols = np.random.randint(1,nMax+1)
# Number of digits after the decimal
nDigits = 1
# Range of entries
numMax = 10
# The two matrices to be multiplied
A = np.random.random_integers(-numMax,numMax,(nRows,nInner))/(10**nDigits)
B = np.random.random_integers(-numMax,numMax,(nInner,nCols))/(10**nDigits)
# Product of these two matrices
C = A.dot(B)
# Modify data and return
data["params"]["A"] = pl.to_json(A)
data["params"]["B"] = pl.to_json(B)
data["correct_answers"]["C"] = pl.to_json(C)
def generate(data):
df = pd.io.parsers.read_csv("breast-cancer-train.dat", header=None)
data['params']['df'] = pl.to_json(df.head(15))
data['params']['matrix'] = pl.to_json(np.random.random((3, 3)))
data['params']['my_dictionary'] = {'a': 1, 'b': 2, 'c': 3}
return data
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get true answer
a_tru = pl.from_json(data['correct_answers'].get(name, None))
if a_tru is None:
return
a_tru = np.array(a_tru)
if a_tru.ndim != 2:
raise ValueError('true answer must be a 2D array')
else:
m, n = np.shape(a_tru)
A = np.empty([m, n])
# Create an array for the submitted answer to be stored in data['submitted_answer'][name]
# used for display in the answer and submission panels
# Also creates invalid error messages
invalid_format = False
for i in range(m):
for j in range(n):
each_entry_name = name + str(n * i + j + 1)
a_sub = data['submitted_answers'].get(each_entry_name, None)
if a_sub is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(No submitted answer)'
invalid_format = True
elif not a_sub:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(Invalid blank entry)'
invalid_format = True
else:
a_sub_parsed = pl.string_to_number(a_sub, allow_complex=False)
if a_sub_parsed is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(Invalid format)'
invalid_format = True
elif not np.isfinite(a_sub_parsed):
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(Invalid format - not finite)'
invalid_format = True
else:
data['submitted_answers'][each_entry_name] = pl.to_json(
a_sub_parsed)
A[i, j] = a_sub_parsed
if invalid_format:
with open('pl-matrix-component-input.mustache', 'r',
encoding='utf-8') as f:
data['format_errors'][name] = chevron.render(
f, {
'format_error': True
}).strip()
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(A)
def generate(data):
sympy.var('y b a m')
x = (y - b) / a
z = m * (sympy.cos(a) + sympy.I * sympy.sin(a))
c = a + sympy.I * b
data['correct_answers']['x'] = pl.to_json(x)
data['correct_answers']['z'] = pl.to_json(z)
data['correct_answers']['I'] = 'V / R'
data['correct_answers']['c'] = pl.to_json(c)
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get true answer
a_tru = pl.from_json(data['correct_answers'].get(name, None))
if a_tru is None:
return
a_tru = np.array(a_tru)
if a_tru.ndim != 2:
raise ValueError('true answer must be a 2D array')
else:
m, n = np.shape(a_tru)
A = np.empty([m, n])
# Create an array for the submitted answer to be stored in data['submitted_answer'][name]
# used for display in the answer and submission panels
# Also creates invalid error messages
invalid_format = False
for i in range(m):
for j in range(n):
each_entry_name = name + str(n * i + j + 1)
a_sub = data['submitted_answers'].get(each_entry_name, None)
if a_sub is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(No submitted answer)'
invalid_format = True
elif not a_sub:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(Invalid blank entry)'
invalid_format = True
else:
a_sub_parsed = pl.string_to_number(a_sub, allow_complex=False)
if a_sub_parsed is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(Invalid format)'
invalid_format = True
elif not np.isfinite(a_sub_parsed):
data['submitted_answers'][each_entry_name] = None
data['format_errors'][
each_entry_name] = '(Invalid format - not finite)'
invalid_format = True
else:
data['submitted_answers'][each_entry_name] = pl.to_json(
a_sub_parsed)
A[i, j] = a_sub_parsed
if invalid_format:
data['format_errors'][
name] = 'At least one of the entries has invalid format (empty entries or not a double precision floating point number)'
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(A)
def generate(data): data['params']['weight1'] = random.randint(1, 10) data['params']['weight2'] = random.randint(1, 10) mat = np.random.random((3, 3)) mat = mat / la.norm(mat, 1, axis=0) data['params']['labels'] = pl.to_json(['A', 'B', 'C']) data['params']['matrix'] = pl.to_json(mat) mat2 = np.random.binomial(1, 0.5, (3, 3)) data['params']['matrix2'] = pl.to_json(mat2)
def generate(data):
N = random.choice([2, 3])
data["params"]["N"] = N
for i in range(2):
X = np.random.rand(N, N)
name = "N" + str(i + 1)
data["params"][name] = pl.to_json(X)
for i in range(4):
M = random.choice([j for j in range(2, 5) if j not in [N]])
X = np.random.rand(M, M)
name = "M" + str(i + 1)
data["params"][name] = pl.to_json(X)
return data
def generate(data):
N = random.choice([2,3])
data["params"]["N"] = N
for i in range(2):
X = np.random.rand(N,N)
name = "N" + str(i+1)
data["params"][name] = pl.to_json(X)
for i in range(4):
M = random.choice([j for j in range(2,5) if j not in [N]])
X = np.random.rand(M,M)
name = "M" + str(i+1)
data["params"][name] = pl.to_json(X)
return data
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get true answer
a_tru = pl.from_json(data['correct_answers'].get(name, None))
if a_tru is None:
return
a_tru = np.array(a_tru)
if a_tru.ndim != 2:
raise ValueError('true answer must be a 2D array')
else:
m, n = np.shape(a_tru)
A = np.empty([m, n])
# Create an array for the submitted answer to be stored in data['submitted_answer'][name]
# used for display in the answer and submission panels
# Also creates invalid error messages
invalid_format = False
for i in range(m):
for j in range(n):
each_entry_name = name + str(n * i + j + 1)
a_sub = data['submitted_answers'].get(each_entry_name, None)
if a_sub is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(No submitted answer)'
invalid_format = True
elif not a_sub:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(Invalid blank entry)'
invalid_format = True
else:
a_sub_parsed = pl.string_to_number(a_sub, allow_complex=False)
if a_sub_parsed is None:
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(Invalid format)'
invalid_format = True
elif not np.isfinite(a_sub_parsed):
data['submitted_answers'][each_entry_name] = None
data['format_errors'][each_entry_name] = '(Invalid format - not finite)'
invalid_format = True
else:
data['submitted_answers'][each_entry_name] = pl.to_json(a_sub_parsed)
A[i, j] = a_sub_parsed
if invalid_format:
data['format_errors'][name] = 'At least one of the entries has invalid format (empty entries or not a double precision floating point number)'
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(A)
def generate(data):
# Number of digits after the decimal
nDigits = 1
# Dimension
n = np.random.randint(2,4)
# Matrix
A = np.round( np.random.rand(n,n), nDigits)
# Product
B = [email protected]
# Return data
data["params"]["nDigits"] = nDigits
data["params"]["sigfigs"] = nDigits + 1
data["params"]["A"] = pl.to_json(A)
data["correct_answers"]["B"] = pl.to_json(B)
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
if a_sub.strip() == '':
opts = {
'format_error': True,
'format_error_message': 'the submitted answer was blank.'
}
with open('pl-integer-input.mustache', 'r', encoding='utf-8') as f:
format_str = chevron.render(f, opts).strip()
data['format_errors'][name] = format_str
data['submitted_answers'][name] = None
return
# Convert to integer
try:
a_sub_parsed = pl.string_to_integer(a_sub)
if a_sub_parsed is None:
raise ValueError('invalid submitted answer (wrong type)')
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
except Exception:
with open('pl-integer-input.mustache', 'r', encoding='utf-8') as f:
format_str = chevron.render(f, {'format_error': True}).strip()
data['format_errors'][name] = format_str
data['submitted_answers'][name] = None
def generate(data):
df = pd.read_csv("clientFilesQuestion/properties.csv")
selected_columns = [
"Designation", "Depth (in)", 'Width (in)',
'Moment of Inertia - Ix (in^4)', 'Moment of Inertia - Iy (in^4)'
]
m = len(df)
pos1 = np.random.randint(0, m - 10)
# note that in general, you will need to do a check if m > 10.
# for simplicity, I am not doing any safety check because I know the length of my table is greater than 10
pos2 = pos1 + 10
df = df[selected_columns].iloc[pos1:pos2]
m = len(df)
select = random.sample(range(1, m), 4)
name_list = df['Designation'].iloc[select]
for i, name in enumerate(name_list):
data["params"]["name" + str(i + 1)] = name
Ix = df['Moment of Inertia - Ix (in^4)'].iloc[select].values
h = df['Depth (in)'].iloc[select].values
M = np.random.randint(2, 8)
data["params"]["M"] = M
data["params"]["df"] = pl.to_json(df)
for i in range(len(h)):
sigma = M * (h[i] / 2) / Ix[i]
data["correct_answers"]["sigma" + str(i + 1)] = sigma * 100
def generate(data):
# Generates number of events.
x_events = random.randint(4, 7)
# Control rounding
n_digits = 2
# Generate random integers up to a number
r = [float(random.randint(1, 100)) for i in range(x_events)]
# Sum events
s = sum(r)
# Divide each value by sum to bring into [0, 1]
# Round events
a = [ round(i/s, n_digits) for i in r ]
# Enforce probability summation constraint
prob_sum = sum(a)
capped_sum = round(1 - prob_sum, n_digits)
# Truncate
if not capped_sum.is_integer():
max_val = max(a)
max_ind = a.index(max_val)
a[max_ind] += capped_sum
d = {'P(X)': a}
df = pd.DataFrame(data = d)
data['params']['df'] = pl.to_json(df)
data['correct_answers']['p_sum'] = sum(a)
data['correct_answers']['big_event'] = a.index(max(a))
def generate(data):
website_list = [
"Google", "Wikipedia", "YouTube", "Instagram", "Stackoverflow",
"Twitter", "NYTimes", "The Guardian", "taobao", "Amazon", "Reddit",
"Netflix", "Linkedin", "Ebay", "GitHub"
]
npages = 4
max_n_links = 3
min_n_links = 1
# Getting the Markov matrix
M, page_names = create_markov_matrix(website_list, npages, max_n_links,
min_n_links)
data["params"]["M"] = pl.to_json(M)
data["params"]["page_names"] = page_names
# Getting the eigenvector
G = 0.85 * M + (0.15 / npages) * np.ones((npages, npages))
xstar = power_iteration(G, 1e-8)
probs = np.sort(xstar)[::-1]
x_index_sorted = np.argsort(xstar)[::-1]
first = page_names[x_index_sorted[0]]
# Setting up the dropdown list
dropdown_list = []
for name in page_names:
if name != first:
dropdown_list.append({'tag': 'false', 'ans': name})
else:
dropdown_list.append({'tag': 'true', 'ans': name})
data["params"]["sites"] = dropdown_list
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
allow_complex = pl.get_boolean_attrib(element, 'allow-complex',
ALLOW_COMPLEX_DEFAULT)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
# Convert to float or complex
try:
a_sub_parsed = pl.string_to_number(a_sub, allow_complex=allow_complex)
if a_sub_parsed is None:
raise ValueError('invalid submitted answer (wrong type)')
if not np.isfinite(a_sub_parsed):
raise ValueError('invalid submitted answer (not finite)')
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
except Exception:
if allow_complex:
data['format_errors'][
name] = 'Invalid format. The submitted answer could not be interpreted as a double-precision floating-point or complex number.'
else:
data['format_errors'][
name] = 'Invalid format. The submitted answer could not be interpreted as a double-precision floating-point number.'
data['submitted_answers'][name] = None
def generate(data):
df = pd.read_csv("clientFilesQuestion/properties.csv")
selected_columns = [
"Designation", "Depth (in)", 'Width (in)',
'Moment of Inertia - Ix (in^4)', 'Moment of Inertia - Iy (in^4)'
]
m = len(df)
pos1 = np.random.randint(0, m - 10)
# note that in general, you will need to do a check if m > 10.
# for simplicity, I am not doing any safety check because I know the length of my table is greater than 10
pos2 = pos1 + 10
df = df[selected_columns].iloc[pos1:pos2]
data["params"]["df"] = pl.to_json(df)
m = len(df)
select = np.random.randint(0, m)
name = df['Designation'].iloc[select]
data["params"]["name"] = name
Ix = df['Moment of Inertia - Ix (in^4)'].iloc[select]
h = df['Depth (in)'].iloc[select]
M = np.random.randint(2, 8)
data["params"]["M"] = M
sigma = M * (h / 2) / Ix
data["correct_answers"]["sigma"] = sigma * 100
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get allow-blank option
allow_blank = pl.get_string_attrib(element, 'allow-blank',
ALLOW_BLANK_DEFAULT)
normalize_to_ascii = pl.get_boolean_attrib(element, 'normalize-to-ascii',
NORMALIZE_TO_ASCII_DEFAULT)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
if normalize_to_ascii:
a_sub = unidecode(a_sub)
data['submitted_answers'][name] = a_sub
if not a_sub and not allow_blank:
data['format_errors'][
name] = 'Invalid format. The submitted answer was left blank.'
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(a_sub)
def parse(element_html, data):
# By convention, this function returns at the first error found
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
allow_complex = pl.get_boolean_attrib(element, 'allow-complex',
ALLOW_COMPLEX_DEFAULT)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = get_format_string('No submitted answer.')
data['submitted_answers'][name] = None
return
# Convert submitted answer to numpy array (return parse_error on failure)
(a_sub_parsed, info) = pl.string_to_2darray(a_sub,
allow_complex=allow_complex)
if a_sub_parsed is None:
data['format_errors'][name] = get_format_string(info['format_error'])
data['submitted_answers'][name] = None
return
# Replace submitted answer with numpy array
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
# Store format type
if '_pl_matrix_input_format' not in data['submitted_answers']:
data['submitted_answers']['_pl_matrix_input_format'] = {}
data['submitted_answers']['_pl_matrix_input_format'][name] = info[
'format_type']
def parse(element_html, data):
# By convention, this function returns at the first error found
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
allow_complex = pl.get_boolean_attrib(element, 'allow-complex', False)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
# Convert submitted answer to numpy array (return parse_error on failure)
(a_sub_parsed, info) = pl.string_to_2darray(a_sub, allow_complex=allow_complex)
if a_sub_parsed is None:
data['format_errors'][name] = info['format_error']
data['submitted_answers'][name] = None
return
# Replace submitted answer with numpy array
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
# Store format type
if '_pl_matrix_input_format' not in data['submitted_answers']:
data['submitted_answers']['_pl_matrix_input_format'] = {}
data['submitted_answers']['_pl_matrix_input_format'][name] = info['format_type']
def generate(data):
data['params']['weight1'] = random.randint(1, 10)
data['params']['weight2'] = random.randint(1, 10)
mat = np.random.random((3, 3))
mat = mat / la.norm(mat, 1, axis=0)
data['params']['labels'] = pl.to_json(['A', 'B', 'C'])
data['params']['matrix'] = pl.to_json(mat)
mat2 = np.random.binomial(1, 0.5, (3, 3))
data['params']['matrix2'] = pl.to_json(mat2)
# chosen by dice roll, guaranteed to be random
edge_mat = np.array([[-1, 0, 1, 0], [0, -1, 1, 0], [1, 0, 0, -1],
[0, 1, -1, 0]])
data['params']['edge-inc-mat'] = pl.to_json(edge_mat)
def generate(data):
# Dimensions
n = np.random.randint(3, 5)
m = np.random.randint(2, 4)
A = np.round(np.random.rand(m, n), 2)
b = np.round(np.random.rand(n), 2)
# Product of these two matrices
C = A @ b
# # Modify data and return
data["params"]["A"] = pl.to_json(A)
data["params"]["b"] = pl.to_json(b.reshape(n, 1))
data["correct_answers"]["C"] = pl.to_json(C.reshape(m, 1))
def generate(data):
N = 2
A = np.random.rand(N,N)
sf = 2
B = np.round(A,sf)
x = np.array([[1,2,3,4]])
data["params"]["sf"] = sf
data["params"]["in"] = pl.to_json(B)
data["params"]["x"] = pl.to_json(x)
data["correct_answers"]["out1"] = pl.to_json(B)
data["correct_answers"]["out2"] = pl.to_json(B)
data["correct_answers"]["out3"] = pl.to_json(x)
return data
def generate(data):
mu_o = random.choice([19.7, 19.8, 19.9, 20, 20.1, 20.2])
std_gen = random.choice([1.3, 1.4, 1.5])
sample_size = random.randint(8, 11)
# dataset
sample_data = np.round(np.random.normal(mu_o, std_gen, sample_size), 1)
data['params']['array'] = pl.to_json(sample_data)
# computing the correct answer
data['correct_answers']['std'] = sample_data.std()
def generate(data):
# Number of digits after the decimal point to display
d = 3
# Generate random vectors
u = np.round(np.random.random(4), d)
v = np.round(np.random.random(4), d)
a = np.random.randint(1, 10)
b = np.random.randint(1, 10)
w = np.outer(a * u, b * v)
# Release parameters
data["params"]["u"] = pl.to_json(u)
data["params"]["v"] = pl.to_json(v)
data["params"]["a"] = a
data["params"]["b"] = b
data["params"]["d"] = d
data["params"]["d2"] = d + 1
data["correct_answers"]["w"] = pl.to_json(w)
def generate(data):
data["params"]["names_for_user"] = []
data["params"]["names_from_user"] = [{
"name": "df",
"description": "dataframe as described above.",
"type": "Pandas DataFrame"
}]
data['params']['df'] = pl.to_json(
pd.DataFrame(data={
'a': [1, 2],
'b': [3, 4]
}))
def generate(data):
# Generate a random value
x = random.uniform(1, 2)
# Fill in the Blank Inputs
data["correct_answers"]["ans_rtol"] = x
data["correct_answers"]["ans_sig"] = round(x, 2)
data["correct_answers"]["int_value"] = 42
data["correct_answers"]["string_value"] = "Learn"
# Symbolic
x, y = sympy.symbols('x y')
data['correct_answers']['symbolic_math'] = pl.to_json(x + y + 1)
# Matrix Fill in the Blank
data['correct_answers']['matrixA'] = pl.to_json(np.matrix('1 2; 3 4'))
# Programming Variant of Supplying a Matrix
data['correct_answers']['matrixB'] = pl.to_json(np.matrix('1 2; 3 4'))
# Threejs
data['correct_answers']['robotC'] = [[1, 0, 0], [0, 45, 0]]
# Output elements
data['params']['matrixC'] = pl.to_json(np.matrix('5 6; 7 8'))
data['params']['matrixD'] = pl.to_json(np.matrix('-1 4; 3 2'))
# Display python variable contents
data_dictionary = { 'a': 1, 'b': 2, 'c': 3 }
data['params']['data_dictionary'] = pl.to_json(data_dictionary)
# Display a pandas data frame
d = {'col1': [1, 2], 'col2': [3, 4]}
df = pd.DataFrame(data=d)
data['params']['df'] = pl.to_json(df)
# Display a dynamically generated graph
mat = np.random.random((3, 3))
mat = mat / np.linalg.norm(mat, 1, axis=0)
data['params']['labels'] = pl.to_json(['A', 'B', 'C'])
data['params']['matrix'] = pl.to_json(mat)
# Overlay
data['correct_answers']['c'] = (2 * (3 ** 2)) ** 0.5
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
base = pl.get_integer_attrib(element, 'base', BASE_DEFAULT)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
if a_sub.strip() == '':
if pl.get_boolean_attrib(element, 'allow-blank', ALLOW_BLANK_DEFAULT):
a_sub = pl.get_integer_attrib(element, 'blank-value',
BLANK_VALUE_DEFAULT)
else:
opts = {
'format_error': True,
'format_error_message': 'the submitted answer was blank.',
'base': base,
'default_base': base == BASE_DEFAULT or base == 0,
'zero_base': base == 0
}
with open('pl-integer-input.mustache', 'r', encoding='utf-8') as f:
format_str = chevron.render(f, opts).strip()
data['format_errors'][name] = format_str
data['submitted_answers'][name] = None
return
# Convert to integer
try:
a_sub_parsed = pl.string_to_integer(str(a_sub), base)
if a_sub_parsed is None:
raise ValueError('invalid submitted answer (wrong type)')
if a_sub_parsed > 2**53 - 1 or a_sub_parsed < -((2**53) - 1):
data['format_errors'][
name] = 'correct answer must be between -9007199254740991 and +9007199254740991 (that is, between -(2^53 - 1) and +(2^53 - 1)).'
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
except Exception:
with open('pl-integer-input.mustache', 'r', encoding='utf-8') as f:
format_str = chevron.render(
f, {
'format_error': True,
'base': base,
'default_base': base == BASE_DEFAULT or base == 0,
'zero_base': base == 0
}).strip()
data['format_errors'][name] = format_str
data['submitted_answers'][name] = None
def generate(data):
n = 4
choice = np.random.choice(np.arange(2, dtype='int'), p=[.7, .3])
if choice == 0:
# random graph
correct = np.random.choice([0, 1], size=(n, n), p=[0.5, 0.5])
elif choice == 1:
# ring
shift = np.random.choice([1, -1]) * np.random.randint(1, n - 2)
correct = np.roll(np.diag(np.ones(n)), shift)
data["correct_answers"]["mat"] = pl.to_json(correct.T)
data["params"]["mat"] = data["correct_answers"]["mat"]
return data
def generate(data):
# Fill in the Blank Inputs
data["correct_answers"]["numericvalue"] = 3.14
data["correct_answers"]["integervalue"] = 42
data["correct_answers"]["stringvalue"] = "PrairieLearn"
# Symbolic
sympy.var('x y')
data['correct_answers']['mathexpvalue'] = pl.to_json(x + y + 1)
# Matrix Fill in the Blank
data['correct_answers']['matrixA'] = pl.to_json(np.matrix('1 2; 3 4'))
# Programming Variant of Supplying a Matrix
data['correct_answers']['matrixB'] = pl.to_json(np.matrix('1 2; 3 4'))
# Threejs
data['correct_answers']['robotC'] = [[1, 0, 0], [0, 45, 0]]
# Output elements
data['params']['matrixC'] = pl.to_json(np.matrix('5 6; 7 8'))
data['params']['matrixD'] = pl.to_json(np.matrix('-1 4; 3 2'))
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
allow_fractions = pl.get_boolean_attrib(element, 'allow-fractions',
ALLOW_FRACTIONS_DEFAULT)
# Get true answer
a_tru = pl.from_json(data['correct_answers'].get(name, None))
if a_tru is None:
return
a_tru = np.array(a_tru)
if a_tru.ndim != 2:
raise ValueError('true answer must be a 2D array')
else:
m, n = np.shape(a_tru)
A = np.empty([m, n])
# Create an array for the submitted answer to be stored in data['submitted_answer'][name]
# used for display in the answer and submission panels
# Also creates invalid error messages
invalid_format = False
for i in range(m):
for j in range(n):
each_entry_name = name + str(n * i + j + 1)
a_sub = data['submitted_answers'].get(each_entry_name, None)
value, newdata = pl.string_fraction_to_number(a_sub,
allow_fractions,
allow_complex=False)
if value is not None:
A[i, j] = value
data['submitted_answers'][each_entry_name] = newdata[
'submitted_answers']
else:
invalid_format = True
data['format_errors'][each_entry_name] = newdata[
'format_errors']
data['submitted_answers'][each_entry_name] = None
if invalid_format:
with open('pl-matrix-component-input.mustache', 'r',
encoding='utf-8') as f:
data['format_errors'][name] = chevron.render(
f, {
'format_error': True,
'allow_fractions': allow_fractions
}).strip()
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(A)
def generate(data):
mu_o = random.choice([19.7, 19.8, 19.9, 20, 20.1, 20.2])
std_gen = random.choice([1.3, 1.4, 1.5])
sample_size = random.randint(8, 11)
# dataset
sample_data = np.round(np.random.normal(mu_o, std_gen, sample_size), 1)
# providing the dataset as pandas frame
df = pd.DataFrame(data=sample_data)
df.columns = ['Volume (oz)']
data['params']['df'] = pl.to_json(df)
# computing the correct answer
data['correct_answers']['average'] = sample_data.mean()
def generate(data):
# Sample two complex numbers with real and imaginary parts that are
# random integers between -5 and 5 (inclusive)
a = random.randint(-5, 5) + random.randint(-5, 5)*1j
b = random.randint(-5, 5) + random.randint(-5, 5)*1j
# Put string representations of these two complex numbers into data['params']
data['params']['a'] = '{:.0f}'.format(a)
data['params']['b'] = '{:.0f}'.format(b)
# Compute the sum of these two complex numbers
c = a + b
# Put the sum into data['correct_answers']
data['correct_answers']['c'] = pl.to_json(c)
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get allow-blank option
allow_blank = pl.get_string_attrib(element, 'allow-blank', False)
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
if not a_sub and not allow_blank:
data['format_errors'][name] = 'Invalid format. The submitted answer was left blank.'
data['submitted_answers'][name] = None
else:
data['submitted_answers'][name] = pl.to_json(a_sub)
def parse(element_html, data):
element = lxml.html.fragment_fromstring(element_html)
name = pl.get_string_attrib(element, 'answers-name')
# Get submitted answer or return parse_error if it does not exist
a_sub = data['submitted_answers'].get(name, None)
if a_sub is None:
data['format_errors'][name] = 'No submitted answer.'
data['submitted_answers'][name] = None
return
# Convert to integer
try:
a_sub_parsed = pl.string_to_integer(a_sub)
if a_sub_parsed is None:
raise ValueError('invalid submitted answer (wrong type)')
if not np.isfinite(a_sub_parsed):
raise ValueError('invalid submitted answer (not finite)')
data['submitted_answers'][name] = pl.to_json(a_sub_parsed)
except Exception:
data['format_errors'][name] = 'Invalid format. The submitted answer was not an integer.'
data['submitted_answers'][name] = None
def generate(data):
N = 2
A = np.random.rand(N,N)
sf = 2
B = np.round(A,sf)
x = np.array([[1,2,3,4]])
long_matrix = np.array([[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]])
data["params"]["sf"] = sf
data["params"]["in"] = pl.to_json(B)
data["params"]["x"] = pl.to_json(x)
data["correct_answers"]["out1"] = pl.to_json(B)
data["correct_answers"]["out2"] = pl.to_json(B)
data["correct_answers"]["out3"] = pl.to_json(x)
data["correct_answers"]["out4"] = pl.to_json(long_matrix)
return data
