predict.py

# -*- coding: utf-8 -*-
"""Untitled7.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1dkd1OtzLdrjgfqr_2z83qLapdDEiaz7E
"""

import numpy as np
from sklearn.utils import shuffle
from sklearn.ensemble import RandomForestClassifier
from sklearn import svm
from sklearn.neural_network import MLPClassifier
import sklearn.metrics as metrics

import pandas as pd
def makedataset():
  labels=[]
  X = []
  X_dip=[]
  # X_tri=[]
  X_atom=[]
  train = pd.read_csv('final_amino_acid_result.csv')
  train_dip = pd.read_csv("final_dipeptide_result.csv")
  train_label = pd.read_csv('train.csv')
  # tripip_train = pd.read_csv('Tripeptide.csv')
  atom_train = pd.read_csv('atom.csv')
  for item in train.itertuples():
    X.append(list(item[2:]))
  for item in train_dip.itertuples():
    X_dip.append(list(item[2:]))
  X_dip = np.reshape(np.array(X_dip),(2550,400))
  # X_tri.append([0]*8000)
  # for item in tripip_train.itertuples():
  #   X_tri.append(list(item[1:]))
  # X_tri = np.reshape(np.array(X_tri),(2550,8000))
  for item in atom_train.itertuples():
    X_atom.append(item[2:])
  X_atom = np.reshape(np.array(X_atom),(2550,5))
  for item in train_label.itertuples():
    if(item[2]=='-1'):
      labels.append(-1)
    else:
      labels.append(1)
  labels = np.array(labels) 
  labels = np.reshape(labels,(2550))
  X = np.reshape(np.array(X),(2550,20))
  X = np.concatenate((X,X_dip,X_atom),axis=1)
  return X,labels
# makedataset()

sequence,labels = makedataset()
# sequence, labels = shuffle(sequence, labels)
# X_t,actual = sequence[1785:],labels[1785:]
# sequence,labels = sequence[:1785],labels[:1785]
test = pd.read_csv('test_final.csv')
test_dip = pd.read_csv('dipitest.csv')
test_atom = pd.read_csv('atom_test.csv')
test_data = pd.read_csv('test.csv')
X_t=[]
X_dt =[]
X_at =[]
for item in test.itertuples():
  X_t.append(list(item[2:]))
for item in test_dip.itertuples():
  X_dt.append(list(item[2:]))
for item in test_atom.itertuples():
  X_at.append(list(item[2:]))
X_t = np.reshape(np.array(X_t),(len(X_t),20))
X_dt = np.reshape(np.array(X_dt),(len(X_dt),400))
X_at = np.reshape(np.array(X_at),(len(X_at),5))
X_t = np.concatenate((X_t,X_dt,X_at),axis=1)
print(X_t.shape)

avg=0
for i in range(1):
  clf = RandomForestClassifier(n_estimators=500)
  # clf = svm.SVC()
  # clf = MLPClassifier()
  clf.fit(sequence,labels)
  y_pred = clf.predict(X_t) 
  print(list(y_pred))
#   print("Accuracy:",metrics.accuracy_score(actual, y_pred))
#   avg+=metrics.accuracy_score(actual, y_pred)
# print(avg/10)