#!/usr/bin/env python
# coding: utf-8

# In[90]:


import pandas as pd
import csv


# In[91]:


tsv_data = pd.read_csv('in.tsv', sep='\t',header=None, quoting=csv.QUOTE_NONE)[0]


# In[139]:


expected = pd.read_csv('expected.tsv', sep='\t',header=None)[0]


# In[94]:


print(len(expected))
print(len(tsv_data))


# In[158]:


male={'silnik', 'windows', 'gb', 'mb', 'mecz', 'pc', 'opony', 'apple', 'iphone', 'zwiastuny', 'hd', 'ubuntu', 'system', 'serwer'}
female={'ciąża', 'miesiączki', 'ciasto', 'ciąże', 'zadowolona', 'antykoncepcyjne', 'ginekologia', 'tabletki', 'porodzie', 'mąż', 'krwawienie', 'ciasta'}
male = {x[:6].lower() for x in male}
female = {x[:6].lower() for x in female}


# In[159]:


trimmed_docs=[]
for document in tsv_data:
    new_doc=[]
    for word in str(document).lower().split():
        new_doc.append(word[:6])
    trimmed_docs.append(new_doc)


# In[160]:


male_or_female=[]
for doc in trimmed_docs:
    male_or_female.append((len(male&set(doc)), len(female&set(doc))))


# In[161]:


answers=[]
for i in male_or_female:
    if i[0]>i[1]:
        answers.append(1)
    else:
        answers.append(0)


# In[162]:


result=[]
for i in range(len(answers)):
    if answers[i]==expected[i]:
        result.append(1)
    else:
        result.append(0)


# In[163]:


print(f'Predykcja modelu wynosi {sum(result)/len(result)*100:.6f}%')


# In[167]:


df = pd.DataFrame(result)
df.to_csv('out.tsv', sep = '\t')