from get_data import GetData
getData = GetData()
fields = ['Open', 'High', 'Low', 'Close', 'Adj_Close']
accuracy = {}
features = getData.getAllFeatures()
symbols = getData.getAllSymbols()
# get_data_block_end
for symbol in symbols:
accuracy[symbol] = []
for field in range(1, 5):
labels = getData.getSymbolCLFLabels(symbol, field)
########################
# now the real MA work #
########################
# create train and test data set
X_test, X_train, y_test, y_train = train_test_split(features,
labels,
test_size=.5)
# create classifier
my_classifier = tree.DecisionTreeClassifier()
# train the classifier
my_classifier.fit(X_train, y_train)
# do prediction
import time
from datetime import datetime, date, time, timedelta
# get_data_block_start
from get_data import GetData
from save_data import SaveData
getData = GetData()
saveData = SaveData()
symbols = getData.getAllSymbols()
for symbol in symbols:
# we just predict up/down of close price #
result = getData.getSymbolCLFLabels(symbol, 4)
features = getData.getSymbolFeaturesWithoutDate(symbol)
allFeatures = getData.getSymbolFeatures(symbol)
dates = []
for feature in allFeatures:
dates.append(feature[0])
# create train and test data set #
high = len(features)
mid = high - 100
low = 0
X_train = features[low:mid]
y_train = result[low:mid]
X_test = features[mid+1:]
y_test = result[mid+1:]
