def setUp(self):
self.op_type = "cross_entropy"
batch_size = 30
class_num = 10
X = randomize_probability(batch_size, class_num, dtype='float64')
label = np.random.randint(0, class_num, (batch_size, 1), dtype="int64")
cross_entropy = np.asmatrix([[-np.log(X[i][label[i][0]])]
for i in range(X.shape[0])],
dtype="float64")
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": False}
def setUp(self):
self.op_type = "cross_entropy"
batch_size = 5
class_num = 37
X = randomize_probability(batch_size, class_num)
label = np.random.uniform(0.1, 1.0,
[batch_size, class_num]).astype("float32")
label /= label.sum(axis=1, keepdims=True)
cross_entropy = (-label * np.log(X)).sum(
axis=1, keepdims=True).astype("float32")
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": True}
def setUp(self):
self.op_type = "cross_entropy"
batch_size = 30
class_num = 10
X = randomize_probability(batch_size, class_num, dtype='float64')
label = np.random.randint(0, class_num, (batch_size, 1), dtype="int64")
cross_entropy = np.asmatrix(
[[-np.log(X[i][label[i][0]])] for i in range(X.shape[0])],
dtype="float64")
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": False}
def setUp(self):
self.op_type = "cross_entropy"
batch_size = 5
class_num = 37
X = randomize_probability(batch_size, class_num)
label = np.random.uniform(0.1, 1.0,
[batch_size, class_num]).astype("float32")
label /= label.sum(axis=1, keepdims=True)
cross_entropy = (-label * np.log(X)).sum(
axis=1, keepdims=True).astype("float32")
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": True}
def setUp(self):
self.op_type = "bpr_loss"
batch_size = 40
class_num = 5
X = randomize_probability(batch_size, class_num, dtype='float64')
label = np.random.randint(0, class_num, (batch_size, 1), dtype="int64")
bpr_loss_result = []
for i in range(batch_size):
sum = 0.0
for j in range(class_num):
if j == label[i][0]:
continue
sum += (-np.log(1.0 + np.exp(X[i][j] - X[i][label[i][0]])))
bpr_loss_result.append(-sum / (class_num - 1))
bpr_loss = np.asmatrix([[x] for x in bpr_loss_result], dtype="float64")
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": bpr_loss}
def setUp(self):
self.op_type = "cross_entropy"
shape = [4, 3]
ins_num = np.prod(np.array(shape))
class_num = 37
X_2d = randomize_probability(ins_num, class_num)
label_2d = np.random.uniform(0.1, 1.0,
[ins_num, class_num]).astype("float32")
label_2d /= label_2d.sum(axis=1, keepdims=True)
cross_entropy_2d = (-label_2d * np.log(X_2d)).sum(
axis=1, keepdims=True).astype("float32")
X = X_2d.reshape(shape + [class_num])
label = label_2d.reshape(shape + [class_num])
cross_entropy = np.array(cross_entropy_2d).reshape(shape + [1])
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": True}
def setUp(self):
self.op_type = "cross_entropy"
shape = [10, 2, 4]
ins_num = np.prod(np.array(shape))
class_num = 10
X_2d = randomize_probability(ins_num, class_num, dtype='float64')
label_2d = np.random.randint(0, class_num, (ins_num, 1), dtype="int64")
cross_entropy_2d = np.asmatrix([[-np.log(X_2d[i][label_2d[i][0]])]
for i in range(X_2d.shape[0])],
dtype="float64")
X = X_2d.reshape(shape + [class_num])
label = label_2d.reshape(shape + [1])
cross_entropy = np.array(cross_entropy_2d).reshape(shape + [1])
self.inputs = {"X": X, "Label": label}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": False}
def setUp(self):
self.op_type = "cross_entropy"
batch_size = 5
class_num = 17
X = randomize_probability(batch_size, class_num)
label_index = np.random.randint(
0, class_num, (batch_size), dtype="int32")
label = np.zeros(X.shape)
label[np.arange(batch_size), label_index] = 1
cross_entropy = np.asmatrix(
[[-np.log(X[i][label_index[i]])] for i in range(X.shape[0])],
dtype="float32")
cross_entropy2 = (-label * np.log(X)).sum(
axis=1, keepdims=True).astype("float32")
self.inputs = {"X": X, "Label": label.astype(np.float32)}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": True}
def setUp(self):
self.op_type = "cross_entropy"
shape = [4, 3, 2]
ins_num = np.prod(np.array(shape))
class_num = 17
X_2d = randomize_probability(ins_num, class_num)
label_index_2d = np.random.randint(0,
class_num, (ins_num),
dtype="int32")
label_2d = np.zeros(X_2d.shape)
label_2d[np.arange(ins_num), label_index_2d] = 1
cross_entropy_2d = np.asmatrix([[-np.log(X_2d[i][label_index_2d[i]])]
for i in range(X_2d.shape[0])],
dtype="float32")
X = X_2d.reshape(shape + [class_num])
label = label_2d.reshape(shape + [class_num])
cross_entropy = np.array(cross_entropy_2d).reshape(shape + [1])
self.inputs = {"X": X, "Label": label.astype(np.float32)}
self.outputs = {"Y": cross_entropy}
self.attrs = {"soft_label": True}
def init_x(self):
self.x = randomize_probability(self.batch_size,
self.class_num,
dtype=self.dtype)
def init_x(self):
self.shape = [4, 3, 2]
self.ins_num = np.prod(np.array(self.shape))
self.X_2d = randomize_probability(self.ins_num,
self.class_num).astype(self.dtype)
self.x = self.X_2d.reshape(self.shape + [self.class_num])
