def read_data():
"""Reads the states and actions recorded by drive_manually.py"""
print("Reading data")
with gzip.open('./data_from_expert/data_02.pkl.gzip', 'rb') as f:
data = pickle.load(f)
X = utils.vstack(data["state"])
y = utils.vstack(data["action"])
return X, y
def read_data(data_path, use_last=False):
# TODO: Fix the file thing
print("Reading data...")
all_states, _, _, all_actions, _ = utils.read_all_gzip(data_path)
if use_last:
all_states = all_states[-1:]
all_actions = all_actions[-1:]
X = utils.vstack(all_states)
y = utils.vstack(all_actions)
return X, y
output = model(data)
loss = criterion(output, target)
batch_real_data.append(target.numpy())
batch_predicted.append(output.numpy())
batch_size = data.shape[0]
total_loss += loss.item() * batch_size
n_samples = len(data_loader.sampler)
loss = total_loss / n_samples
print('{} Loss: {}'.format(total_loss, loss))
real_data = vstack(batch_real_data)
predicted = vstack(batch_predicted)
print(predicted)
d = 0
for i in range(len(real_data)):
print(predicted[i])
if real_data[i] == np.argmax(predicted[i]):
d += 1
print(d / len(real_data) * 100)
#real_data = dataset.min_max_scaler.inverse_transform(real_data)
#predicted = dataset.min_max_scaler.inverse_transform(predicted)
real_data = np.exp(real_data)
predicted = np.exp(predicted)
gamma = 1.0
print("Loading data...")
train_num = Dataset.load_part('train', 'numeric')
train_cat = Dataset.load_part('train', 'categorical_dummy')
test_num = Dataset.load_part('test', 'numeric')
test_cat = Dataset.load_part('test', 'categorical_dummy')
print("Combining data...")
#vstack 按行拼接
#hstack 按列拼接
#拼接之后kmeans聚类
all_data = hstack((scale(vstack((train_num, test_num)).astype(np.float64)).astype(np.float32), vstack((train_cat, test_cat))))
for n_clusters in [25, 50, 75, 100, 200]:
part_name = 'cluster_rbf_%d' % n_clusters
print("Finding %d clusters..." % n_clusters)
kmeans = MiniBatchKMeans(n_clusters, random_state=17 * n_clusters + 11, n_init=5)
kmeans.fit(all_data)
print("Transforming data...")
cluster_rbf = np.exp(- gamma * kmeans.transform(all_data))
print("Saving...")
al1 = ones(a, 1)
al2 = zeros(a, 1)
bl1 = zeros(b, 1)
bl2 = ones(b, 1)
l1a = al1.T
l1b = bl1.T
l2a = al2.T
l2b = bl2.T
ll = zeros(2, 2)
abB = zeros(a + b, 1)
lB = ones(2, 1)
aA = hstack(aa, ab, al1, al2)
bA = hstack(ba, bb, bl1, bl2)
l1A = hstack(l1a, l1b)
l2A = hstack(l2a, l2b)
lA = hstack(vstack(l1A, l2A), ll)
A = vstack(aA, bA, lA)
B = vstack(abB, lB)
print("A,B =")
print(hstack(A, B))
m: List[bool] = np.zeros((a + b + n), dtype=bool) # mask
m[-2] = 1
m[-1] = 1
# TODO: check that the stationary point is a minimum
v: List[Tuple] = []
while increment(m, 0, a):
while increment(m, a, a + b):