def run_different_parameter(thread_id, gamma):
figure_name = ('gamma' + str(gamma)).replace('.', 'dot')
bandit_env = StochasticEnvironment(k, sigma_noise)
policies = [EXP3('EXP3', k, gamma, 'black')]
simulator = Simulator(bandit_env, policies, time_horizon)
regret_dict = simulator.run(num_thread, num_mc)
plot_regret(figure_name, policies, regret_dict)
def parameter_multiprocessing_run(thread_id, sigma_noise):
figure_name = ('sigma_noise' + str(sigma_noise)).replace('.', 'dot')
bandit_env = Environment(k, d, sigma_noise)
policies = [LinUCB('LinUCB', k, d, lambda_, arm_norm_bound, theta_norm_bound, delta, sigma_noise, 'black')]
simulator = Simulator(bandit_env, policies, time_horizon)
regret_dict = simulator.multiprocessing_run(num_thread, num_mc)
plot_regret(figure_name, policies, regret_dict, jupyter_notebook)
def solo_processing():
# solo_processing
option = 'solo_processing'
sigma_noise = 0.1
figure_name = ('ucb_convergence_verification_sigma' + str(sigma_noise)).replace('.', 'dot')
policies = [UCB(sigma_noise, k, delta)]
avg_regret = run_experiment(num_mc, option)
plot_regret(figure_name, line_name, avg_regret, jupyter_notebook)
def multi_processing():
global num_thread, sigma_noise, policies
# multi_processing
option = 'multi_processing'
num_thread = 10
sigma_noise = 0.1
figure_name = ('ucb_convergence_verification_sigma' + str(sigma_noise)).replace('.', 'dot')
policies = [UCB(sigma_noise, k, delta)]
avg_regret = run_experiment(num_mc, option)
plot_regret(figure_name, line_name, avg_regret, jupyter_notebook)
def run_model():
data, true_labels = ldl.get_data_linear()
true_buckets = [util.bucket(t) for t in true_labels]
data = np.tile(data, (DATA_MULTIPLIER, 1))
print("DATA SHAPE:", data.shape)
true_buckets = np.tile(true_buckets, DATA_MULTIPLIER)
# tuples of (batch_id, total regret, error while training, eval error, precision, recall)
batch_results = []
for T in range(NUM_BATCHES):
model = Lin_UCB(ALPHA)
#model = LASSO_BANDIT()
if False:
data, true_labels, columns_dict, values_dict = dl.get_data()
true_buckets = [util.bucket(t) for t in true_labels]
#model = Fixed_Dose(columns_dict, values_dict)
#model = Warfarin_Clinical_Dose(columns_dict, values_dict)
#model = Warfarin_Pharmacogenetic_Dose(columns_dict, values_dict)
batch_id = str(random.randint(100000, 999999))
print()
print("Start Batch: ", batch_id)
zipped_data = list(zip(data, true_buckets))
random.shuffle(zipped_data)
data, true_buckets = zip(*zipped_data)
data = np.array(data)
model.train(data, true_buckets)
pred_buckets = model.evaluate(data)
print(batch_id, "Performance on " + str(model))
acc, precision, recall = util.evaluate_performance(
pred_buckets, true_buckets)
print("\tAccuracy:", acc)
print("\tPrecision:", precision)
print("\tRecall:", recall)
plot_regret(model.regret, ALPHA, batch_id)
plot_error_rate(model.error_rate, ALPHA, batch_id)
batch_results.append(
(batch_id, model.get_regret()[-1], model.get_error_rate()[-1],
1 - acc, precision, recall))
with open('batch/regret' + str(model) + batch_id, 'wb') as fp:
pickle.dump(model.regret, fp)
with open('batch/error' + str(model) + batch_id, 'wb') as fp:
pickle.dump(model.error_rate, fp)
return batch_results
tmp = features.T.dot(self.A_inv[i]).dot(features)
p[i] = self.theta[i].dot(features) + self.alpha * np.sqrt(tmp)
choose_action = np.argmax(p)
return choose_action
def test_lin_ucb_full(data, true_buckets, alpha=0.1):
lin_ucb = Lin_UCB(alpha=alpha)
lin_ucb.train(data, true_buckets)
pred_buckets = lin_ucb.evaluate(data)
acc, precision, recall = util.evaluate_performance(pred_buckets,
true_buckets)
#print("accuracy on linear UCB: " + str(acc))
if __name__ == '__main__':
data, true_labels = ldl.get_data_linear()
true_buckets = [util.bucket(t) for t in true_labels]
ALPHA = 0.1
lin_ucb = Lin_UCB(alpha=ALPHA)
lin_ucb.train(data, true_buckets)
pred_buckets = lin_ucb.evaluate(data)
acc, precision, recall = util.evaluate_performance(pred_buckets,
true_buckets)
#print("accuracy on linear UCB: " + str(acc))
plot_regret(lin_ucb.regret, ALPHA)
plot_error_rate(lin_ucb.error_rate, ALPHA)