def model_selection_with_transformation(distance_funcs, scaling_classes,
Xtrain, ytrain, Xval, yval):
# distance_funcs: dictionary of distance funtion
# scaling_classes: diction of scalers
# Xtrain: List[List[int]] train set
# ytrain: List[int] train labels
# Xval: List[List[int]] validation set
# yval: List[int] validation labels
# return best_model: an instance of KNN
# return best_k: best k choosed for best_model
# return best_func: best function choosed for best_model
# return best_scaler: best function choosed for best_model
# ifthere are less than 30 points in dataset, choose n-1 as the upper bound of K.
#'n' is the number of points in dataset.
# You can choose N-1 as best k if N-1 is an odd number.
best_scale = None
best_score = 0.0
best_distance = None
best_func = ""
best_k = 0
max_score = 0.0
#modified due to grading instructions
if len(Xtrain) < 30:
kvals = np.arange(1, len(Xtrain), 2)
kvals = np.arange(1, len(Xtrain), 2)
train_f1_score = 1.0
valid_f1_score = 0.0
model = None
for scaling_name, new_scaler in scaling_classes.items():
print(scaling_name, new_scaler)
scaler = new_scaler()
scaled_Xtrain = scaler(Xtrain)
scaled_Xval = scaler(Xval)
for name, f in distance_funcs.items():
for k in kvals:
if k == 1:
model = KNN(k, f)
model.train(scaled_Xtrain, ytrain)
else:
model.k = k
valid_f1_score = f1_score(yval, model.predict(scaled_Xval))
if valid_f1_score > max_score:
max_score = valid_f1_score
best_distance = f
best_k = k
best_func = name
best_scale = scaling_name
if valid_f1_score == max_score:
if k < best_k:
max_score = valid_f1_score
best_distance = f
best_k = k
best_func = name
best_scale = scaling_name
model.k = best_k
model.distance_function = best_distance
model.f1 = max_score
model.scaler = scaling_classes[best_scale]
best_model = KNN(best_k, best_distance)
best_model.scaler = scaling_classes[best_scale]
best_model.train(Xtrain, ytrain)
#print('best score: ', best_score)
#print('best k: ', best_k)
#print('best scaler: ', scaling_name)
# Dont change any print statement
print('[part 1.1] {name}\tk: {k:d}\t'.format(name=name, k=k) +
'train:{train_f1_score:.5f}\t'.format(
train_f1_score=train_f1_score) +
'valid: {valid_f1_score:.5f}'.format(valid_f1_score=valid_f1_score))
print('[part 1.2] {name}\t{scaling_name}\tk: {k:d}\t'.format(
name=name, scaling_name=scaling_name, k=k) +
'train: {train_f1_score:.5f}\t'.format(
train_f1_score=train_f1_score) +
'valid: {valid_f1_score:.5f}'.format(valid_f1_score=valid_f1_score))
print()
print('[part 1.2] {name}\t{scaling_name}\t'.format(
name=name, scaling_name=scaling_name) +
'best_k: {best_k:d}\t'.format(best_k=best_k))
print()
return best_model, best_k, best_func, best_scale
def model_selection_without_normalization(distance_funcs, Xtrain, ytrain, Xval,
yval):
# distance_funcs: dictionary of distance funtion
# Xtrain: List[List[int]] train set
# ytrain: List[int] train labels
# Xval: List[List[int]] validation set
# yval: List[int] validation labels
# return best_model: an instance of KNN
# return best_k: best k choosed for best_model
# return best_func: best function choosed for best_model
best_distance = None
best_function = ""
best_k = 0
max_score = 0.0
kvals = np.arange(1, len(Xtrain), 2)
train_f1_score = 1.0
valid_f1_score = 0.0
model = None
for name, f in distance_funcs.items():
for k in kvals:
if k == 1:
model = KNN(k, f)
model.train(Xtrain, ytrain)
else:
model.k = k
# train_f1_score =f1_score(ytrain,model.predict(Xtrain))
# predicted=model.predict(Xval)
valid_f1_score = f1_score(yval, model.predict(Xval))
if valid_f1_score > max_score:
max_score = valid_f1_score
print("new valid score: ", valid_f1_score)
best_distance = f
best_function = name
best_k = k
print('**NEW BEST MODEL**')
if valid_f1_score == max_score:
if k < best_k:
max_score = valid_f1_score
print("new valid score: ", valid_f1_score)
best_distance = f
best_function = name
best_k = k
# best_model=KNN(best_k,best_distance)
# best_model.train(Xtrain,ytrain)
model.k = best_k
model.distance_function = best_distance
model.f1 = max_score
best_model = KNN(best_k, best_distance)
best_model.train(Xtrain, ytrain)
# Dont change any print statement
print('[part 1.1] {name}\tk: {k:d}\t'.format(name=name, k=k) +
'train: {train_f1_score:.5f}\t'.format(
train_f1_score=train_f1_score) +
'valid: {valid_f1_score:.5f}'.format(valid_f1_score=valid_f1_score))
print('[part 1.1] {name}\tk: {k:d}\t'.format(name=name, k=k) +
'train: {train_f1_score:.5f}\t'.format(
train_f1_score=train_f1_score) +
'valid: {valid_f1_score:.5f}'.format(valid_f1_score=valid_f1_score))
print()
print('[part 1.1] {name}\tbest_k: {best_k:d}\t'.format(name=name,
best_k=best_k))
return best_model, best_k, best_function