def fte_bte_experiment(
train_x,
train_y,
test_x,
test_y,
ntrees,
shift,
slot,
which_task,
acorn=None,
):
# We initialize lists to store the results
df = pd.DataFrame()
accuracies_across_tasks = []
# Declare the progressive learner model (L2F)
learner = LifelongClassificationForest()
for task_num in range((which_task - 1), 10):
accuracy_per_task = []
# print("Starting Task {} For Shift {} For Slot {}".format(task_num, shift, slot))
if acorn is not None:
np.random.seed(acorn)
# If first task, add task. Else, add a transformer for the task
if task_num == (which_task - 1):
learner.add_task(
X=train_x[(task_num * 900):((task_num + 1) * 900)],
y=train_y[(task_num * 900):((task_num + 1) * 900)],
task_id=0,
)
t_num = 0
# Add transformers for all task up to current task (task t)
while t_num < task_num:
# Make a prediction on task t using the trained learner on test data
llf_task = learner.predict(
test_x[((which_task - 1) * 1000):(which_task * 1000), :],
task_id=0,
)
acc = np.mean(llf_task == test_y[((which_task - 1) *
1000):(which_task * 1000)])
accuracies_across_tasks.append(acc)
learner.add_transformer(
X=train_x[(t_num * 900):((t_num + 1) * 900)],
y=train_y[(t_num * 900):((t_num + 1) * 900)],
)
# Add transformer for next task
t_num = t_num + 1
else:
learner.add_transformer(
X=train_x[(task_num * 900):((task_num + 1) * 900)],
y=train_y[(task_num * 900):((task_num + 1) * 900)],
)
# Make a prediction on task t using the trained learner on test data
llf_task = learner.predict(test_x[((which_task - 1) *
1000):(which_task * 1000), :],
task_id=0)
acc = np.mean(llf_task == test_y[((which_task - 1) *
1000):(which_task * 1000)])
accuracies_across_tasks.append(acc)
# print("Accuracy Across Tasks: {}".format(accuracies_across_tasks))
df["task"] = range(1, 11)
df["task_accuracy"] = accuracies_across_tasks
return df
def label_shuffle_experiment(
train_x,
train_y,
test_x,
test_y,
ntrees,
shift,
slot,
num_points_per_task,
acorn=None,
):
# We initialize lists to store the results
df = pd.DataFrame()
shifts = []
accuracies_across_tasks = []
# Declare the progressive learner model (L2F), with ntrees as a parameter
learner = LifelongClassificationForest(n_estimators=ntrees)
for task_ii in range(10):
print("Starting Task {} For Fold {} For Slot {}".format(
task_ii, shift, slot))
if acorn is not None:
np.random.seed(acorn)
# If task number is 0, add task. Else, add a transformer for the task
if task_ii == 0:
learner.add_task(
X=train_x[task_ii * 5000 +
slot * num_points_per_task:task_ii * 5000 +
(slot + 1) * num_points_per_task],
y=train_y[task_ii * 5000 +
slot * num_points_per_task:task_ii * 5000 +
(slot + 1) * num_points_per_task],
task_id=0,
)
else:
learner.add_transformer(
X=train_x[task_ii * 5000 +
slot * num_points_per_task:task_ii * 5000 +
(slot + 1) * num_points_per_task],
y=train_y[task_ii * 5000 +
slot * num_points_per_task:task_ii * 5000 +
(slot + 1) * num_points_per_task],
)
# Make a prediction on task 0 using the trained learner on test data
llf_task = learner.predict(test_x[:1000], task_id=0)
# Calculate the accuracy of the task 0 predictions
acc = np.mean(llf_task == test_y[:1000])
accuracies_across_tasks.append(acc)
shifts.append(shift)
print("Accuracy Across Tasks: {}".format(accuracies_across_tasks))
df["data_fold"] = shifts
df["task"] = range(1, 11)
df["task_1_accuracy"] = accuracies_across_tasks
return df
def run_experiment(x_data,
y_data,
num_tasks,
num_points_per_task,
ntrees=10,
model='uf',
reps=100):
""" Runs the FTE/BTE experiment.
Referenced Chenyu's code, with modifications to adjust the number of tasks.
"""
# initialize list for storing results
accuracies_across_tasks = []
# format data
if model == 'dnn': # add dnn implementation in the future
x = x_data
y = y_data
elif model == 'uf':
x = x_data.reshape(len(x_data), -1)
y = y_data
# get y values per task
unique_y = np.unique(y_data)
ys_by_task = unique_y.reshape(num_tasks, int(len(unique_y) / num_tasks))
# run experiment over all reps
for rep in range(reps):
#print('Starting rep', rep)
# for each task
for task in range(num_tasks):
# initialize progressive learner
learner = LifelongClassificationForest(
default_n_estimators=ntrees
) #default_max_depth=np.ceil(np.log2(num_points_per_task))
# get train/test data (train = num_points_per_task)
index = np.where(np.in1d(y, ys_by_task[task]))
x_task0 = x[index]
y_task0 = y[index]
train_x_task0, test_x_task0, train_y_task0, test_y_task0 = train_test_split(
x_task0, y_task0, test_size=0.25)
train_x_task0 = train_x_task0[:num_points_per_task]
train_y_task0 = train_y_task0[:num_points_per_task]
# feed to learner and predict on single task
learner.add_task(train_x_task0, train_y_task0)
task_0_predictions = learner.predict(test_x_task0, task_id=0)
accuracies_across_tasks.append(
np.mean(task_0_predictions == test_y_task0))
# evaluate for other tasks
for other_task in range(num_tasks):
if other_task == task:
pass
else:
# get train/test data (train = num_points_per_task)
index = np.random.choice(np.where(
np.in1d(y, ys_by_task[other_task]))[0],
num_points_per_task,
replace=False)
train_x = x[index]
train_y = y[index]
# add transformer from other tasks
learner.add_task(train_x, train_y)
# predict on current task using other tasks
prev_task_predictions = learner.predict(test_x_task0,
task_id=0)
accuracies_across_tasks.append(
np.mean(prev_task_predictions == test_y_task0))
# average results
accuracy_all_task = np.array(accuracies_across_tasks).reshape((reps, -1))
accuracy_all_task = np.mean(accuracy_all_task, axis=0)
return accuracy_all_task
def ftebte_exp(x, y, model, num_tasks, num_trees, reps, shift):
"""
Runs the FTE/BTE experiment given the following parameters:
x - data features
y - data labels
model - "uf" or "nn"
num_tasks - number of tasks
num_trees - number of trees
shift - whether to shift the data
"""
# shift data if indicated
x, y = shift_data(x, y, shift)
# initialize list for storing results
accuracies_across_tasks = []
# get y values per task
ys_by_task = [np.unique(i) for i in y]
# get the count of the least frequent label over all tasks
min_labelct = np.min(
[np.min(np.unique(each_set, return_counts=True)[1]) for each_set in y])
# run experiment over all reps
for rep in range(reps):
train_x_task = []
train_y_task = []
test_x_task = []
test_y_task = []
# sample min points (31) from each dataset
x_sample = []
y_sample = []
for dataset, label in zip(x, y):
sample = []
for unique_label in np.unique(label):
sample += list(
np.random.choice(
np.where(label == unique_label)[0], min_labelct))
x_sample.append(dataset[sample])
y_sample.append(label[sample])
# initialize overall learner
learner = LifelongClassificationForest(default_n_estimators=num_trees,
default_max_depth=30)
# for each task
for task in range(num_tasks):
# get train/test data
tr_x, te_x, tr_y, te_y = train_test_split(x_sample[task],
y_sample[task],
test_size=0.2)
train_x_task.append(tr_x)
train_y_task.append(tr_y)
test_x_task.append(te_x)
test_y_task.append(te_y)
# predict on single task (UF learner) - CHANGE TO UNCERTAINTYFOREST LATER
uf_learner = LifelongClassificationForest(
default_n_estimators=num_trees, default_max_depth=30)
uf_learner.add_task(train_x_task[task], train_y_task[task])
uf_predictions = uf_learner.predict(test_x_task[task], task_id=0)
accuracies_across_tasks.append(
np.mean(uf_predictions == test_y_task[task]))
# feed to overall learner
learner.add_task(train_x_task[task], train_y_task[task])
# evaluate for other tasks
for other_task in range(num_tasks):
if other_task > task:
pass
else:
# predict on current task using other tasks
prev_task_predictions = learner.predict(
test_x_task[other_task], task_id=other_task)
accuracies_across_tasks.append(
np.mean(
prev_task_predictions == test_y_task[other_task]))
# average results
accuracy_all_task = np.array(accuracies_across_tasks).reshape((reps, -1))
accuracy_all_task = np.mean(accuracy_all_task, axis=0)
return accuracy_all_task
def experiment(n_task1, n_task2, n_test=1000,
n_trees=10, max_depth=None, random_state=None):
"""
A function to do progressive experiment between two tasks
where the task data is generated using Gaussian parity.
Parameters
----------
n_task1 : int
Total number of train sample for task 1.
n_task2 : int
Total number of train dsample for task 2
n_test : int, optional (default=1000)
Number of test sample for each task.
n_trees : int, optional (default=10)
Number of total trees to train for each task.
max_depth : int, optional (default=None)
Maximum allowable depth for each tree.
random_state : int, RandomState instance, default=None
Determines random number generation for dataset creation. Pass an int
for reproducible output across multiple function calls.
Returns
-------
errors : array of shape [6]
Elements of the array is organized as single task error task1,
multitask error task1, single task error task2,
multitask error task2, naive UF error task1,
naive UF task2.
"""
if n_task1==0 and n_task2==0:
raise ValueError('Wake up and provide samples to train!!!')
if random_state != None:
np.random.seed(random_state)
errors = np.zeros(6,dtype=float)
progressive_learner = LifelongClassificationForest(default_n_estimators=n_trees)
uf1 = LifelongClassificationForest(default_n_estimators=n_trees)
naive_uf = LifelongClassificationForest(default_n_estimators=n_trees)
uf2 = LifelongClassificationForest(default_n_estimators=n_trees)
#source data
X_task1, y_task1 = generate_spirals(n_task1, 3, noise=0.8)
test_task1, test_label_task1 = generate_spirals(n_test, 3, noise=0.8)
#target data
X_task2, y_task2 = generate_spirals(n_task2, 5, noise=0.4)
test_task2, test_label_task2 = generate_spirals(n_test, 5, noise=0.4)
if n_task1 == 0:
progressive_learner.add_task(X_task2, y_task2, n_estimators=n_trees)
uf2.add_task(X_task2, y_task2, n_estimators=n_trees)
errors[0] = 0.5
errors[1] = 0.5
uf_task2=uf2.predict(test_task2, task_id=0)
l2f_task2=progressive_learner.predict(test_task2, task_id=0)
errors[2] = 1 - np.mean(uf_task2 == test_label_task2)
errors[3] = 1 - np.mean(l2f_task2 == test_label_task2)
errors[4] = 0.5
errors[5] = 1 - np.mean(uf_task2 == test_label_task2)
elif n_task2 == 0:
progressive_learner.add_task(X_task1, y_task1,
n_estimators=n_trees)
uf1.add_task(X_task1, y_task1,
n_estimators=n_trees)
uf_task1=uf1.predict(test_task1, task_id=0)
l2f_task1=progressive_learner.predict(test_task1, task_id=0)
errors[0] = 1 - np.mean(uf_task1 == test_label_task1)
errors[1] = 1 - np.mean(l2f_task1 == test_label_task1)
errors[2] = 0.5
errors[3] = 0.5
errors[4] = 1 - np.mean(uf_task1 == test_label_task1)
errors[5] = 0.5
else:
progressive_learner.add_task(X_task1, y_task1, n_estimators=n_trees)
progressive_learner.add_task(X_task2, y_task2, n_estimators=n_trees)
uf1.add_task(X_task1, y_task1, n_estimators=2*n_trees)
uf2.add_task(X_task2, y_task2, n_estimators=2*n_trees)
naive_uf_train_x = np.concatenate((X_task1,X_task2),axis=0)
naive_uf_train_y = np.concatenate((y_task1,y_task2),axis=0)
naive_uf.add_task(
naive_uf_train_x, naive_uf_train_y, n_estimators=n_trees
)
uf_task1=uf1.predict(test_task1, task_id=0)
l2f_task1=progressive_learner.predict(test_task1, task_id=0)
uf_task2=uf2.predict(test_task2, task_id=0)
l2f_task2=progressive_learner.predict(test_task2, task_id=1)
naive_uf_task1 = naive_uf.predict(
test_task1, task_id=0
)
naive_uf_task2 = naive_uf.predict(
test_task2, task_id=0
)
errors[0] = 1 - np.mean(
uf_task1 == test_label_task1
)
errors[1] = 1 - np.mean(
l2f_task1 == test_label_task1
)
errors[2] = 1 - np.mean(
uf_task2 == test_label_task2
)
errors[3] = 1 - np.mean(
l2f_task2 == test_label_task2
)
errors[4] = 1 - np.mean(
naive_uf_task1 == test_label_task1
)
errors[5] = 1 - np.mean(
naive_uf_task2 == test_label_task2
)
return errors
def experiment(n_xor, n_nxor, n_test, reps, n_trees, max_depth, acorn=None):
"""
Runs the Gaussian XOR N-XOR experiment.
Returns the mean error.
"""
# initialize experiment
if n_xor == 0 and n_nxor == 0:
raise ValueError("Wake up and provide samples to train!!!")
# if acorn is specified, set random seed to it
if acorn != None:
np.random.seed(acorn)
# initialize array for storing errors
errors = np.zeros((reps, 4), dtype=float)
# run the progressive learning algorithm for a number of repetitions
for i in range(reps):
# initialize learners
progressive_learner = LifelongClassificationForest(
n_estimators=n_trees)
uf = UncertaintyForest(n_estimators=2 * n_trees)
# source data
xor, label_xor = generate_gaussian_parity(n_xor, angle_params=0)
test_xor, test_label_xor = generate_gaussian_parity(n_test,
angle_params=0)
# target data
nxor, label_nxor = generate_gaussian_parity(n_nxor,
angle_params=np.pi / 2)
test_nxor, test_label_nxor = generate_gaussian_parity(
n_test, angle_params=np.pi / 2)
if n_xor == 0:
# fit learners and predict
progressive_learner.add_task(nxor, label_nxor)
l2f_task2 = progressive_learner.predict(test_nxor, task_id=0)
uf.fit(nxor, label_nxor)
uf_task2 = uf.predict(test_nxor)
# record errors
errors[
i,
0] = 0.5 # no data, so random chance of guessing correctly (err = 0.5)
errors[
i,
1] = 0.5 # no data, so random chance of guessing correctly (err = 0.5)
errors[i, 2] = 1 - np.sum(uf_task2 == test_label_nxor) / n_test
errors[i, 3] = 1 - np.sum(l2f_task2 == test_label_nxor) / n_test
elif n_nxor == 0:
# fit learners and predict
progressive_learner.add_task(xor, label_xor)
l2f_task1 = progressive_learner.predict(test_xor, task_id=0)
uf.fit(xor, label_xor)
uf_task1 = uf.predict(test_xor)
# record errors
errors[i, 0] = 1 - np.sum(uf_task1 == test_label_xor) / n_test
errors[i, 1] = 1 - np.sum(l2f_task1 == test_label_xor) / n_test
errors[
i,
2] = 0.5 # no data, so random chance of guessing correctly (err = 0.5)
errors[
i,
3] = 0.5 # no data, so random chance of guessing correctly (err = 0.5)
else:
# fit learners and predict
progressive_learner.add_task(xor, label_xor)
progressive_learner.add_task(nxor, label_nxor)
l2f_task1 = progressive_learner.predict(test_xor, task_id=0)
l2f_task2 = progressive_learner.predict(test_nxor, task_id=1)
uf.fit(xor, label_xor)
uf_task1 = uf.predict(test_xor)
uf.fit(nxor, label_nxor)
uf_task2 = uf.predict(test_nxor)
# record errors
errors[i, 0] = 1 - np.sum(uf_task1 == test_label_xor) / n_test
errors[i, 1] = 1 - np.sum(l2f_task1 == test_label_xor) / n_test
errors[i, 2] = 1 - np.sum(uf_task2 == test_label_nxor) / n_test
errors[i, 3] = 1 - np.sum(l2f_task2 == test_label_nxor) / n_test
return np.mean(errors, axis=0)
def LF_experiment(angle,
data_x,
data_y,
granularity,
max_depth,
reps=1,
ntrees=29,
acorn=None):
# Set random seed to acorn if acorn is specified
if acorn is not None:
np.random.seed(acorn)
errors = np.zeros(
2
) # initializes array of errors that will be generated during each rep
for rep in range(reps):
# training and testing subsets are randomly selected by calling the cross_val_data function
train_x1, train_y1, train_x2, train_y2, test_x, test_y = cross_val_data(
data_x, data_y, total_cls=10)
# Change data angle for second task
tmp_data = train_x2.copy()
_tmp_ = np.zeros((32, 32, 3), dtype=int)
total_data = tmp_data.shape[0]
for i in range(total_data):
tmp_ = image_aug(tmp_data[i], angle)
# 2D image is flattened into a 1D array as random forests can only take in flattened images as inputs
tmp_data[i] = tmp_
# .shape gives the dimensions of each numpy array
# .reshape gives a new shape to the numpy array without changing its data
train_x1 = train_x1.reshape((
train_x1.shape[0],
train_x1.shape[1] * train_x1.shape[2] * train_x1.shape[3],
))
tmp_data = tmp_data.reshape((
tmp_data.shape[0],
tmp_data.shape[1] * tmp_data.shape[2] * tmp_data.shape[3],
))
test_x = test_x.reshape(
(test_x.shape[0],
test_x.shape[1] * test_x.shape[2] * test_x.shape[3]))
# number of trees (estimators) to use is passed as an argument because the default is 100 estimators
progressive_learner = LifelongClassificationForest(
n_estimators=ntrees, default_max_depth=max_depth)
# Add the original task
progressive_learner.add_task(X=train_x1, y=train_y1)
# Predict and get errors for original task
llf_single_task = progressive_learner.predict(test_x, task_id=0)
# Add the new transformer
progressive_learner.add_transformer(X=tmp_data, y=train_y2)
# Predict and get errors with the new transformer
llf_task1 = progressive_learner.predict(test_x, task_id=0)
errors[1] = errors[1] + (1 - np.mean(llf_task1 == test_y)
) # errors from transfer learning
errors[0] = errors[0] + (1 - np.mean(llf_single_task == test_y)
) # errors from original task
errors = (
errors / reps
) # errors are averaged across all reps ==> more reps means more accurate errors
# Average errors for original task and transfer learning are returned for the angle tested
return errors
def test(NT, h, names, classifiers, datasets):
i = 1
# iterate over datasets
for ds_cnt, ds in enumerate(datasets):
# preprocess dataset, split into training and test part
X, y = ds
X = StandardScaler().fit_transform(X)
X_train, X_test, y_train, y_test = \
train_test_split(X, y, test_size=.4, random_state=42)
x_min, x_max = X[:, 0].min() - .5, X[:, 0].max() + .5
y_min, y_max = X[:, 1].min() - .5, X[:, 1].max() + .5
xx, yy = np.meshgrid(np.arange(x_min, x_max, h),
np.arange(y_min, y_max, h))
# just plot the dataset first
cm = plt.cm.RdBu
cm_bright = ListedColormap(['#FF0000', '#0000FF'])
ax = plt.subplot(len(datasets), len(classifiers) + 1, i)
if ds_cnt == 0:
ax.set_title("Input data")
# Plot the training points
ax.scatter(X_train[:, 0],
X_train[:, 1],
c=y_train,
cmap=cm_bright,
edgecolors='k')
# Plot the testing points
ax.scatter(X_test[:, 0],
X_test[:, 1],
c=y_test,
cmap=cm_bright,
alpha=0.6,
edgecolors='k')
ax.set_xlim(xx.min(), xx.max())
ax.set_ylim(yy.min(), yy.max())
ax.set_xticks(())
ax.set_yticks(())
i += 1
# iterate over classifiers
for name, clf in zip(names, classifiers):
ax = plt.subplot(len(datasets), len(classifiers) + 1, i)
if "Proglearn" in name:
clf = LifelongClassificationForest(
oblique=True,
default_feature_combinations=1,
default_density=0.5)
clf.add_task(X_train, y_train, n_estimators=NT)
y_hat = clf.predict(X_test, task_id=0)
score = np.sum(y_hat == y_test) / len(y_test)
else:
clf.fit(X_train, y_train)
score = clf.score(X_test, y_test)
# Plot the decision boundary. For that, we will assign a color to each
# point in the mesh [x_min, x_max]x[y_min, y_max].
if hasattr(clf, "decision_function"):
Z = clf.decision_function(np.c_[xx.ravel(), yy.ravel()])
elif "Proglearn" in name:
Z = clf.predict_proba(np.c_[xx.ravel(), yy.ravel()],
task_id=0)[:, 1]
else:
Z = clf.predict_proba(np.c_[xx.ravel(), yy.ravel()])[:, 1]
# Put the result into a color plot
Z = Z.reshape(xx.shape)
ax.contourf(xx, yy, Z, cmap=cm, alpha=.8)
# Plot the training points
ax.scatter(X_train[:, 0],
X_train[:, 1],
c=y_train,
cmap=cm_bright,
edgecolors='k')
# Plot the testing points
ax.scatter(X_test[:, 0],
X_test[:, 1],
c=y_test,
cmap=cm_bright,
edgecolors='k',
alpha=0.6)
ax.set_xlim(xx.min(), xx.max())
ax.set_ylim(yy.min(), yy.max())
ax.set_xticks(())
ax.set_yticks(())
if ds_cnt == 0:
ax.set_title(name)
ax.text(xx.max() - .3,
yy.min() + .3, ('%.2f' % score).lstrip('0'),
size=15,
horizontalalignment='right')
i += 1
def experiment(angle, classifiers, n_xor, n_rxor, n_test):
"""Perform XOR RXOR(XNOR) XOR experiment"""
X_xor, y_xor = generate_gaussian_parity(n_xor)
X_rxor, y_rxor = generate_gaussian_parity(n_rxor, angle_params=angle)
X_xor_2, y_xor_2 = generate_gaussian_parity(n_xor)
test_x_xor, test_y_xor = generate_gaussian_parity(n_test)
test_x_rxor, test_y_rxor = generate_gaussian_parity(n_test,
angle_params=angle)
X_stream = np.concatenate((X_xor, X_rxor, X_xor_2), axis=0)
y_stream = np.concatenate((y_xor, y_rxor, y_xor_2), axis=0)
# Instantiate classifiers
if classifiers[0] == 1:
ht = tree.HoeffdingTreeClassifier(grace_period=2,
split_confidence=1e-01)
if classifiers[1] == 1:
mf = MondrianForestClassifier(n_estimators=10)
if classifiers[2] == 1:
sdt = DecisionTreeClassifier()
if classifiers[3] == 1:
sdf = StreamDecisionForest()
if classifiers[4] == 1:
synf = LifelongClassificationForest(default_n_estimators=10)
errors = np.zeros((10, int(X_stream.shape[0] / 25)))
for i in range(int(X_stream.shape[0] / 25)):
X = X_stream[i * 25:(i + 1) * 25]
y = y_stream[i * 25:(i + 1) * 25]
# Hoeffding Tree Classifier
if classifiers[0] == 1:
ht_partial_fit(ht, X, y)
ht_xor_y_hat, ht_rxor_y_hat = ht_predict(ht, test_x_xor,
test_x_rxor)
errors[0, i] = 1 - np.mean(ht_xor_y_hat == test_y_xor)
errors[1, i] = 1 - np.mean(ht_rxor_y_hat == test_y_rxor)
# Mondrian Forest Classifier
if classifiers[1] == 1:
mf.partial_fit(X, y)
mf_xor_y_hat = mf.predict(test_x_xor)
mf_rxor_y_hat = mf.predict(test_x_rxor)
errors[2, i] = 1 - np.mean(mf_xor_y_hat == test_y_xor)
errors[3, i] = 1 - np.mean(mf_rxor_y_hat == test_y_rxor)
# Stream Decision Tree Classifier
if classifiers[2] == 1:
sdt.partial_fit(X, y, classes=[0, 1])
sdt_xor_y_hat = sdt.predict(test_x_xor)
sdt_rxor_y_hat = sdt.predict(test_x_rxor)
errors[4, i] = 1 - np.mean(sdt_xor_y_hat == test_y_xor)
errors[5, i] = 1 - np.mean(sdt_rxor_y_hat == test_y_rxor)
# Stream Decision Forest Classifier
if classifiers[3] == 1:
sdf.partial_fit(X, y, classes=[0, 1])
sdf_xor_y_hat = sdf.predict(test_x_xor)
sdf_rxor_y_hat = sdf.predict(test_x_rxor)
errors[6, i] = 1 - np.mean(sdf_xor_y_hat == test_y_xor)
errors[7, i] = 1 - np.mean(sdf_rxor_y_hat == test_y_rxor)
# Synergistic Forest Classifier
if classifiers[4] == 1:
if i == 0:
synf.add_task(X, y, n_estimators=10, task_id=0)
synf_xor_y_hat = synf.predict(test_x_xor, task_id=0)
elif i < (n_xor / 25):
synf.update_task(X, y, task_id=0)
synf_xor_y_hat = synf.predict(test_x_xor, task_id=0)
elif i == (n_xor / 25):
synf.add_task(X, y, n_estimators=10, task_id=1)
synf_xor_y_hat = synf.predict(test_x_xor, task_id=0)
synf_rxor_y_hat = synf.predict(test_x_rxor, task_id=1)
elif i < (n_xor + n_rxor) / 25:
synf.update_task(X, y, task_id=1)
synf_xor_y_hat = synf.predict(test_x_xor, task_id=0)
synf_rxor_y_hat = synf.predict(test_x_rxor, task_id=1)
elif i < (2 * n_xor + n_rxor) / 25:
synf.update_task(X, y, task_id=0)
synf_xor_y_hat = synf.predict(test_x_xor, task_id=0)
synf_rxor_y_hat = synf.predict(test_x_rxor, task_id=1)
if i < (n_xor / 25):
errors[8, i] = 1 - np.mean(synf_xor_y_hat == test_y_xor)
if i >= (n_xor / 25):
errors[8, i] = 1 - np.mean(synf_xor_y_hat == test_y_xor)
errors[9, i] = 1 - np.mean(synf_rxor_y_hat == test_y_rxor)
return errors