felixm
/
ML4T
1
0
Fork 0
Code Releases Activity

Change best for LinReg to return optimal data

master
Felix Martin 2020-10-05 17:27:09 -04:00
parent a662e302db
commit 381670705b
2 changed files with 192 additions and 152 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

140
defeat_learners/gen_data.py
View File

@ -1,52 +1,88 @@
""" """
template for generating data to fool learners (c) 2016 Tucker Balch template for generating data to fool learners (c) 2016 Tucker Balch
Copyright 2018, Georgia Institute of Technology (Georgia Tech) Copyright 2018, Georgia Institute of Technology (Georgia Tech)
Atlanta, Georgia 30332 Atlanta, Georgia 30332
All Rights Reserved All Rights Reserved
Template code for CS 4646/7646 Template code for CS 4646/7646
Georgia Tech asserts copyright ownership of this template and all derivative Georgia Tech asserts copyright ownership of this template and all derivative
works, including solutions to the projects assigned in this course. Students works, including solutions to the projects assigned in this course. Students
and other users of this template code are advised not to share it with others and other users of this template code are advised not to share it with others
or to make it available on publicly viewable websites including repositories or to make it available on publicly viewable websites including repositories
such as github and gitlab. This copyright statement should not be removed such as github and gitlab. This copyright statement should not be removed
or edited. or edited.
We do grant permission to share solutions privately with non-students such We do grant permission to share solutions privately with non-students such
as potential employers. However, sharing with other current or future as potential employers. However, sharing with other current or future
students of CS 7646 is prohibited and subject to being investigated as a students of CS 7646 is prohibited and subject to being investigated as a
GT honor code violation. GT honor code violation.
-----do not edit anything above this line--- -----do not edit anything above this line---
Student Name: Tucker Balch (replace with your name) Student Name: Tucker Balch (replace with your name)
GT User ID: tb34 (replace with your User ID) GT User ID: tb34 (replace with your User ID)
GT ID: 900897987 (replace with your GT ID) GT ID: 900897987 (replace with your GT ID)
""" """
import numpy as np import numpy as np
import math import pandas as pd
import math
# this function should return a dataset (X and Y) that will work
# better for linear regression than decision trees
def best4LinReg(seed=1489683273): def best4LinReg(seed=1489683273):
np.random.seed(seed) """
X = np.zeros((100,2)) This function should return a dataset (X and Y) that will work better for
Y = np.random.random(size = (100,))*200-100 linear regression than decision trees.
# Here's is an example of creating a Y from randomly generated
# X with multiple columns We make Y a simple linear combination of X. That will give the Linear
# Y = X[:,0] + np.sin(X[:,1]) + X[:,2]**2 + X[:,3]**3 Regression algorithm a very easy time (no RMSE at all) and beat the DT
return X, Y easily.
"""
def best4DT(seed=1489683273): np.random.seed(seed)
np.random.seed(seed) X = np.random.random(size=(100, 2)) * 200 - 100
X = np.zeros((100,2)) Y = X[:, 0] * -2 + X[:, 1] * 3
Y = np.random.random(size = (100,))*200-100 return X, Y
return X, Y
def author(): def best4DT(seed=1489683273):
return 'tb34' #Change this to your user ID """
This function should return a dataset that will work better for decision
if __name__=="__main__": trees than linear regression.
print("they call me Tim.") """
# Z = np.append(X, Y.reshape(Y.shape[0], 1), 1)
# pd.DataFrame(Z).to_csv("Z.csv", header=None, index=None)
# np.random.seed(seed)
# X = np.random.random(size=(100, 10))*1000-100
# Y = np.random.random(size=(100,))*1000-100
np.random.seed(seed)
# X_1 = np.random.random(size=(100, 1))*200-100
# X_2 = np.random.random(size=(100, 1))*200-100
# X_3 = np.random.random(size=(100, 1))*200-100
# X_4 = np.random.random(size=(100, 1))*200-100
# X = np.concatenate([X_1, X_2, X_3, X_4], 1)
# XXX: I honestly don't know how to help the DTLearner, yet.
X_1 = np.asarray([i for i in range(1, 101)]).reshape(100, 1)
X_2 = np.asarray([i for i in range(100, 1100, 10)]).reshape(100, 1)
X_3 = np.asarray([i for i in range(200, 300)]).reshape(100, 1)
X_4 = np.asarray([i for i in range(300, 400)]).reshape(100, 1)
X_5 = np.asarray([i for i in range(1, 101)]).reshape(100, 1)
X_6 = np.asarray([i for i in range(1, 101)]).reshape(100, 1)
X_7 = np.asarray([i for i in range(1, 101)]).reshape(100, 1)
X_8 = np.asarray([i for i in range(1, 101)]).reshape(100, 1)
X = np.concatenate([X_1, X_2, X_3, X_4, X_5, X_6, X_7, X_8], 1)
# Y = X[:, 0] * 2 + X[:, 1] * 3
Y = np.random.random(size=(100,)) * 200 - 100
return X, Y
def author():
return 'felixm' # Change this to your user ID
if __name__ == "__main__":
print("they call me Tim.")

204
defeat_learners/testbest4.py
View File

@ -1,100 +1,104 @@
""" """
Test best4 data generator. (c) 2016 Tucker Balch Test best4 data generator. (c) 2016 Tucker Balch
Copyright 2018, Georgia Institute of Technology (Georgia Tech) Copyright 2018, Georgia Institute of Technology (Georgia Tech)
Atlanta, Georgia 30332 Atlanta, Georgia 30332
All Rights Reserved All Rights Reserved
Template code for CS 4646/7646 Template code for CS 4646/7646
Georgia Tech asserts copyright ownership of this template and all derivative Georgia Tech asserts copyright ownership of this template and all derivative
works, including solutions to the projects assigned in this course. Students works, including solutions to the projects assigned in this course. Students
and other users of this template code are advised not to share it with others and other users of this template code are advised not to share it with others
or to make it available on publicly viewable websites including repositories or to make it available on publicly viewable websites including repositories
such as github and gitlab. This copyright statement should not be removed such as github and gitlab. This copyright statement should not be removed
or edited. or edited.
We do grant permission to share solutions privately with non-students such We do grant permission to share solutions privately with non-students such
as potential employers. However, sharing with other current or future as potential employers. However, sharing with other current or future
students of CS 7646 is prohibited and subject to being investigated as a students of CS 7646 is prohibited and subject to being investigated as a
GT honor code violation. GT honor code violation.
-----do not edit anything above this line--- -----do not edit anything above this line---
""" """
import numpy as np import numpy as np
import math import math
import LinRegLearner as lrl import LinRegLearner as lrl
import DTLearner as dt import DTLearner as dt
from gen_data import best4LinReg, best4DT from gen_data import best4LinReg, best4DT
# compare two learners' rmse out of sample # compare two learners' rmse out of sample
def compare_os_rmse(learner1, learner2, X, Y):
# compute how much of the data is training and testing def compare_os_rmse(learner1, learner2, X, Y):
train_rows = int(math.floor(0.6* X.shape[0]))
test_rows = X.shape[0] - train_rows # compute how much of the data is training and testing
train_rows = int(math.floor(0.6 * X.shape[0]))
# separate out training and testing data test_rows = X.shape[0] - train_rows
train = np.random.choice(X.shape[0], size=train_rows, replace=False)
test = np.setdiff1d(np.array(range(X.shape[0])), train) # separate out training and testing data
trainX = X[train, :] train = np.random.choice(X.shape[0], size=train_rows, replace=False)
trainY = Y[train] test = np.setdiff1d(np.array(range(X.shape[0])), train)
testX = X[test, :] trainX = X[train, :]
testY = Y[test] trainY = Y[train]
testX = X[test, :]
# train the learners testY = Y[test]
learner1.addEvidence(trainX, trainY) # train it
learner2.addEvidence(trainX, trainY) # train it # train the learners
learner1.addEvidence(trainX, trainY) # train it
# evaluate learner1 out of sample learner2.addEvidence(trainX, trainY) # train it
predY = learner1.query(testX) # get the predictions
rmse1 = math.sqrt(((testY - predY) ** 2).sum()/testY.shape[0]) # evaluate learner1 out of sample
predY = learner1.query(testX) # get the predictions
# evaluate learner2 out of sample rmse1 = math.sqrt(((testY - predY) ** 2).sum()/testY.shape[0])
predY = learner2.query(testX) # get the predictions
rmse2 = math.sqrt(((testY - predY) ** 2).sum()/testY.shape[0]) # evaluate learner2 out of sample
predY = learner2.query(testX) # get the predictions
return rmse1, rmse2 rmse2 = math.sqrt(((testY - predY) ** 2).sum()/testY.shape[0])
def test_code(): return rmse1, rmse2
# create two learners and get data
lrlearner = lrl.LinRegLearner(verbose = False) def test_code():
dtlearner = dt.DTLearner(verbose = False, leaf_size = 1)
X, Y = best4LinReg() # create two learners and get data
lrlearner = lrl.LinRegLearner(verbose=False)
# compare the two learners dtlearner = dt.DTLearner(verbose=False, leaf_size=1)
rmseLR, rmseDT = compare_os_rmse(lrlearner, dtlearner, X, Y) X, Y = best4LinReg()
# share results # compare the two learners
print() rmseLR, rmseDT = compare_os_rmse(lrlearner, dtlearner, X, Y)
print("best4LinReg() results")
print(f"RMSE LR : {rmseLR}") # share results
print(f"RMSE DT : {rmseDT}") print()
if rmseLR < 0.9 * rmseDT: print("best4LinReg() results")
print("LR < 0.9 DT: pass") print(f"RMSE LR : {rmseLR}")
else: print(f"RMSE DT : {rmseDT}")
print("LR >= 0.9 DT: fail") if rmseLR < 0.9 * rmseDT:
print print("LR < 0.9 DT: pass")
else:
# get data that is best for a random tree print("LR >= 0.9 DT: fail")
lrlearner = lrl.LinRegLearner(verbose = False) print
dtlearner = dt.DTLearner(verbose = False, leaf_size = 1)
X, Y = best4DT() # get data that is best for a random tree
lrlearner = lrl.LinRegLearner(verbose=False)
# compare the two learners dtlearner = dt.DTLearner(verbose=False, leaf_size=1)
rmseLR, rmseDT = compare_os_rmse(lrlearner, dtlearner, X, Y) X, Y = best4DT()
# share results # compare the two learners
print() rmseLR, rmseDT = compare_os_rmse(lrlearner, dtlearner, X, Y)
print("best4RT() results")
print(f"RMSE LR : {rmseLR}") # share results
print(f"RMSE DT : {rmseDT}") print()
if rmseDT < 0.9 * rmseLR: print("best4RT() results")
print("DT < 0.9 LR: pass") print(f"RMSE LR : {rmseLR}")
else: print(f"RMSE DT : {rmseDT}")
print("DT >= 0.9 LR: fail") if rmseDT < 0.9 * rmseLR:
print print("DT < 0.9 LR: pass")
else:
if __name__=="__main__": print("DT >= 0.9 LR: fail")
test_code() print
if __name__ == "__main__":
test_code()