def calculate_total_grade():
import numpy as np
csv_file = open('combined.csv')
headers_list = csv_file.next().strip().split(',')
headers_list.append('total_grade')
data_lines = []
scores = []
for line in csv_file:
splits = [f.strip() for f in line.split(',')]
splits = [int(s) if get_type(s) is int else s for s in splits]
ph_score = numeric_or_zero(
splits[6]) #int(splits[6]) if get_type(splits[6]) is int else 0
q_score = numeric_or_zero(
splits[7]
) #float(splits[7]) if get_type(splits[7]) is float else 0
a_score = numeric_or_zero(
splits[11]
) #float(splits[11]) if get_type(splits[11]) is float else 0
total_score = ph_score + q_score + a_score
scores.append(total_score)
data_lines.append(splits)
final_lines = [headers_list]
for key, split_line in enumerate(data_lines):
split_line.append(round(scores[key], 2))
final_lines.append(split_line)
csv_file.close()
with open('ccc2.csv', 'w') as fp:
a = csv.writer(fp, delimiter=',')
a.writerows(final_lines)
def read_multiple_csv(files, key_header):
"""
takes a list of filesnames and a key_header string to merge the files.
Can merge if there are different columns in the files, different records in
the files.
Assumption - all the files should have the column named key_header
example : files = ['grades.csv', 'previous_grades.csv']
key_header = 'email'
merged_data = read_multiple_csv(files, key_header)
"""
data = {}
# files = ['assignments.csv', 'discuss.csv', 'questions.csv']
# key_header = 'email'
for filename in files:
file_generator = open(filename)
headers = file_generator.next().split(',')
headers = [h.strip() for h in headers]
for line in file_generator:
splits = line.split(',')
inner_dict = {}
for k,header in enumerate(headers):
#assumes header is a string. probably it is
if header is not '':
inner_dict[header] = int(splits[k]) if get_type(splits[k]) is int else splits[k]
data_key = inner_dict.pop(key_header)
try:
z = data[data_key].copy()
except KeyError:
z = {}
z.update(inner_dict)
data[data_key] = z
return data
def calculate_total_grade():
import numpy as np
csv_file = open('combined.csv')
headers_list = csv_file.next().strip().split(',')
headers_list.append('total_grade')
data_lines = []
scores = []
for line in csv_file:
splits = [f.strip() for f in line.split(',')]
splits = [int(s) if get_type(s) is int else s for s in splits]
ph_score = numeric_or_zero(splits[6]) #int(splits[6]) if get_type(splits[6]) is int else 0
q_score = numeric_or_zero(splits[7])#float(splits[7]) if get_type(splits[7]) is float else 0
a_score = numeric_or_zero(splits[11]) #float(splits[11]) if get_type(splits[11]) is float else 0
total_score = ph_score + q_score + a_score
scores.append(total_score)
data_lines.append(splits)
final_lines = [headers_list]
for key, split_line in enumerate(data_lines):
split_line.append(round(scores[key], 2))
final_lines.append(split_line)
csv_file.close()
with open('ccc2.csv', 'w') as fp:
a = csv.writer(fp, delimiter=',')
a.writerows(final_lines)
def calculate_assignment_grade():
import numpy as np
with open('aaa.csv') as fo:
headers_list = fo.readline().strip().split(',')
headers_list.append('assignment_grade')
data_lines = []
scores = []
for line in fo:
splits = [f.strip() for f in line.split(',')]
splits = [0 if s == "" else s for s in splits]
splits = [
get_type(s)(s) if get_type(s) in (int, float) else s
for s in splits
]
scores.append(sum(splits[2:]))
data_lines.append(splits)
import pdb
pdb.set_trace()
non_zero_scores = [s for s in scores if s > 0]
std_dev = np.std(non_zero_scores)
grades = [s * 1.0 / std_dev for s in scores]
max_grade = max(grades)
grades = [g * 10 / max_grade for g in grades]
averaging_factor = 7 / np.mean(grades) # converts the average to 7
graded_lines = [headers_list]
for key, split_line in enumerate(data_lines):
try:
split_line.append(round(grades[key], 2))
except:
import pdb
pdb.set_trace()
graded_lines.append(split_line)
with open('aaa2.csv', 'w') as fp:
a = csv.writer(fp, delimiter=',')
a.writerows(graded_lines)
def calculate_assignment_grade():
import numpy as np
with open('aaa.csv') as fo:
headers_list = fo.readline().strip().split(',')
headers_list.append('assignment_grade')
data_lines = []
scores = []
for line in fo:
splits = [f.strip() for f in line.split(',')]
splits = [0 if s == "" else s for s in splits]
splits = [get_type(s)(s) if get_type(s) in (int, float) else s for s in splits]
scores.append(sum(splits[2:]))
data_lines.append(splits)
import pdb
pdb.set_trace()
non_zero_scores = [s for s in scores if s > 0]
std_dev = np.std(non_zero_scores)
grades = [s*1.0/std_dev for s in scores]
max_grade = max(grades)
grades = [g*10/max_grade for g in grades]
averaging_factor = 7 / np.mean(grades) # converts the average to 7
graded_lines = [headers_list]
for key,split_line in enumerate(data_lines):
try:
split_line.append(round(grades[key], 2))
except:
import pdb
pdb.set_trace()
graded_lines.append(split_line)
with open('aaa2.csv', 'w') as fp:
a = csv.writer(fp, delimiter=',')
a.writerows(graded_lines)
def read_multiple_csv(files, key_header, acceptable_keys=None):
"""
takes a list of filesnames and a key_header string to merge the files.
Can merge if there are different columns in the files, different records in
the files.
Assumption - all the files should have the column named key_header
example : files = ['grades.csv', 'previous_grades.csv']
key_header = 'email'
merged_data = read_multiple_csv(files, key_header)
"""
data = {}
# files = ['assignments.csv', 'discuss.csv', 'questions.csv']
# key_header = 'email'
# acceprable keys are list of keys allowed to go into final data
# if not given all will be accepted, this list can come out of
# intersection or other set methods
header_line = []
for filename in files:
with open(filename, 'r+') as fo:
headers = fo.readline().split(',')
headers = [h.strip() for h in headers]
header_line.extend(headers)
for line in fo:
splits = line.split(',')
inner_dict = {}
for k, header in enumerate(headers):
#assumes header is a string. probably it is
if header is not '':
inner_dict[header] = int(splits[k]) if get_type(
splits[k]) is int else splits[k].strip()
data_key = inner_dict.pop(key_header)
if data_key in acceptable_keys:
try:
z = data[data_key].copy()
except KeyError:
z = {}
z.update(inner_dict)
data[data_key] = z
data['headers'] = list(set(header_line))
return data
def read_multiple_csv(files, key_header, acceptable_keys=None):
"""
takes a list of filesnames and a key_header string to merge the files.
Can merge if there are different columns in the files, different records in
the files.
Assumption - all the files should have the column named key_header
example : files = ['grades.csv', 'previous_grades.csv']
key_header = 'email'
merged_data = read_multiple_csv(files, key_header)
"""
data = {}
# files = ['assignments.csv', 'discuss.csv', 'questions.csv']
# key_header = 'email'
# acceprable keys are list of keys allowed to go into final data
# if not given all will be accepted, this list can come out of
# intersection or other set methods
header_line = []
for filename in files:
with open(filename, 'r+') as fo:
headers = fo.readline().split(',')
headers = [h.strip() for h in headers]
header_line.extend(headers)
for line in fo:
splits = line.split(',')
inner_dict = {}
for k,header in enumerate(headers):
#assumes header is a string. probably it is
if header is not '':
inner_dict[header] = int(splits[k]) if get_type(splits[k]) is int else splits[k].strip()
data_key = inner_dict.pop(key_header)
if data_key in acceptable_keys:
try:
z = data[data_key].copy()
except KeyError:
z = {}
z.update(inner_dict)
data[data_key] = z
data['headers'] = list(set(header_line))
return data
def numeric_or_zero(string):
if get_type(string) in [int, float]:
return get_type(string)(string)
return 0
def numeric_or_zero(string):
if get_type(string) in [int, float]:
return get_type(string)(string)
return 0
