def writeData(data,fname):
"""Write a data into csv and pickle objec.
:param data: The data to be wrote
:type data: A list of tuple
:param fname: A filename
:type fname: String
:returns: None
"""
dr.tupleToCSV(data, "result/"+fname,None)
def statistic(f,t):
"""Compute the numbers of somethings for a given years range.
Read the file with name imdb_year.csv in data folder.
:param f: From year
:type f: Integer
:param t: To year
:type t: Integer
:returns: A list of (year:number)
"""
something = {u'actors':11,
u'writers':6,
u'directors':9,
u'genres':13,
u'countries':5,
u'languages':3,
}
data = readDataInRange(f,t,True,True)
datalist = readDataInRange(f,t,False, False)
bigres = {u'actors':[],
u'writers':[],
u'directors':[],
u'genres':[],
u'countries':[],
u'languages':[]
}
for y in sorted(data.keys()):
details = data[y]
for key, value in something.items():
l = listToList([x[value] for x in details])
bigres[key].extend(l)
# do top-n analyses
top_n_produktiv_something_in_range(f,t,bigres)
#
for key, value in bigres.items():
v = nltk.FreqDist(value)
bigres[key] = len(v.keys())
bigres["movies"] = len(datalist)
rating = [float(x[1][2]) for x in datalist if float(x[1][2]) != 0]
maxrating = numpy.max(rating)
minrating = numpy.min(rating)
bigres["average_rating"] = numpy.average(rating)
bigres["max_rating"] = maxrating
bigres["min_rating"] = minrating
runtime = [x[1][3] for x in datalist]
p = reg.compile("\d+")
runtime = [float(reg.match(p,x[1][3]).group(0)) for x in datalist if x[1][3] != "Unknown"]
runtime_sum = numpy.sum(runtime)
runtime_average = numpy.average(runtime)
runtime_min = numpy.min(runtime)
runtime_max = numpy.max(runtime)
bigres["runtime_sum"] = runtime_sum
bigres["runtime_average"] = runtime_average
bigres["runtime_min"] = runtime_min
bigres["runtime_max"] = runtime_max
if t == f:
fn = "statistic_in_{f}".format(f=f)
else:
fn = "statistic_in_{f}_{t}".format(t=t, f=f)
c = [(k, v) for (k,v) in bigres.items()]
dr.tupleToCSV(c,"result/"+fn)
