Training and testing RandomForest classifier in python

f1.py

import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import f1_score

target = 'e'
features = ['a', 'b', 'c', 'd']
cols = len(features)
rows = 1000

df = pd.DataFrame(np.random.randint(1, 5, (rows, cols)), columns=features)
df[target] = np.random.randint(2, size=rows)

# split the dataset
l = range(rows)
trainLen = int(rows*0.75)
testLen  = int(rows*0.25)
training = df.ix[l[:trainLen]]
test = df.ix[l[trainLen:trainLen+testLen]]

X_train = training[features]
y_train = training[target]
X_test = test[features]
y_test = test[target]

clf = RandomForestClassifier()
clf.fit(X_train, y_train)
preds = clf.predict(X_test)

# show confusion matrix
ct = pd.crosstab(y_test, preds, rownames=['actual'], colnames=['preds'])
print ct

#show f1 score
f1 = f1_score(y_test, preds, average=None)
print f1