Skip to content

P
pytensor
提交 daacd069 authored 作者: Brandon T. Willard's avatar Brandon T. Willard
浏览文件
操作

Replace theano.tensor alias T with tt in tests.scan_module

上级 b4dc02d6
隐藏空白字符变更
内嵌 并排
正在显示 包含 54 行增加 和 52 行删除
tests/scan_module/test_scan_checkpoints.py
浏览文件 @ daacd069
......@@ -2,7 +2,7 @@ import numpy as np
import pytest
import theano
import theano.tensor as T
import theano.tensor as tt
import theano.gpuarray
try:
......@@ -15,25 +15,25 @@ except ImportError:
class TestScanCheckpoint:
def setup_method(self):
self.k = T.iscalar("k")
self.A = T.vector("A")
self.k = tt.iscalar("k")
self.A = tt.vector("A")
result, _ = theano.scan(
fn=lambda prior_result, A: prior_result * A,
outputs_info=T.ones_like(self.A),
outputs_info=tt.ones_like(self.A),
non_sequences=self.A,
n_steps=self.k,
)
result_check, _ = theano.scan_checkpoints(
fn=lambda prior_result, A: prior_result * A,
outputs_info=T.ones_like(self.A),
outputs_info=tt.ones_like(self.A),
non_sequences=self.A,
n_steps=self.k,
save_every_N=100,
)
self.result = result[-1]
self.result_check = result_check[-1]
self.grad_A = T.grad(self.result.sum(), self.A)
self.grad_A_check = T.grad(self.result_check.sum(), self.A)
self.grad_A = tt.grad(self.result.sum(), self.A)
self.grad_A_check = tt.grad(self.result_check.sum(), self.A)
def test_forward_pass(self):
# Test forward computation of A**k.
......
tests/scan_module/test_scan_opt.py
浏览文件 @ daacd069
import numpy as np
import theano
import theano.tensor as tt
from theano import config
from theano import tensor as T
from theano.scan_module.scan_op import Scan
from tests import unittest_tools as utt
......@@ -40,11 +40,11 @@ class TestGaussNewton:
# create symbolic inputs and targets variables
if batch_size == 1:
x = T.matrix("inputs")
t = T.matrix("targets")
x = tt.matrix("inputs")
t = tt.matrix("targets")
else:
x = T.tensor3("inputs")
t = T.tensor3("inputs")
x = tt.tensor3("inputs")
t = tt.tensor3("inputs")
x.tag.test_value = inputs.get_value(borrow=True)
t.tag.test_value = targets.get_value(borrow=True)
......@@ -66,17 +66,17 @@ class TestGaussNewton:
# recurrent function
def step(x_t, h_tm1):
h = T.tanh(T.dot(h_tm1, W_hh) + T.dot(x_t, W_xh) + b_h)
h = tt.tanh(tt.dot(h_tm1, W_hh) + tt.dot(x_t, W_xh) + b_h)
return h
# build recurrent graph
if batch_size == 1:
h_0 = T.alloc(0.0, 10).astype(config.floatX)
h_0 = tt.alloc(0.0, 10).astype(config.floatX)
else:
h_0 = T.alloc(0.0, batch_size, 10).astype(config.floatX)
h_0 = tt.alloc(0.0, batch_size, 10).astype(config.floatX)
h, updates = theano.scan(step, sequences=[x], outputs_info=[h_0])
# network output
y = T.dot(h, W_hy) + b_y
y = tt.dot(h, W_hy) + b_y
# Create Gauss-Newton-Matrix object. Not really of any use here, but I
# need it for Hessian-Free optimization.
......@@ -95,7 +95,7 @@ class TestGaussNewton:
# Compute Gauss-Newton-Matrix times some vector `v` which is `p` in CG,
# but for simplicity, I just take the parameters vector because it's
# already there.
Gv = gn(v=params, cost=cost, parameters=params, damp=T.constant(1.0))
Gv = gn(v=params, cost=cost, parameters=params, damp=tt.constant(1.0))
# compile Theano function
f = theano.function([], [cost_] + Gv, givens={x: inputs, t: targets}, mode=mode)
......@@ -121,9 +121,11 @@ class GaussNewtonMatrix(object):
def __call__(self, v, cost, parameters, damp):
# compute Gauss-Newton Matrix right-multiplied by `v`
Jv = T.Rop(self._s, parameters, v)
HJv = T.grad(T.sum(T.grad(cost, self._s) * Jv), self._s, consider_constant=[Jv])
JHJv = T.grad(T.sum(HJv * self._s), parameters, consider_constant=[HJv, Jv])
Jv = tt.Rop(self._s, parameters, v)
HJv = tt.grad(
tt.sum(tt.grad(cost, self._s) * Jv), self._s, consider_constant=[Jv]
)
JHJv = tt.grad(tt.sum(HJv * self._s), parameters, consider_constant=[HJv, Jv])
# apply Tikhonov damping
JHJv = [JHJvi + damp * vi for JHJvi, vi in zip(JHJv, v)]
......@@ -142,17 +144,17 @@ class TestPushOutScanOutputDot(object):
# Test the case where the vector input to the dot is not already an
# output of the inner function.
v = T.vector()
m = T.matrix()
output = T.dot(v, m)
v = tt.vector()
m = tt.matrix()
output = tt.dot(v, m)
# Compile the function twice, once with the optimization and once
# without
opt_mode = mode.including("scan")
f_opt = theano.function([v, m], T.jacobian(output, v), mode=opt_mode)
f_opt = theano.function([v, m], tt.jacobian(output, v), mode=opt_mode)
no_opt_mode = mode.excluding("scanOp_pushout_output")
f_no_opt = theano.function([v, m], T.jacobian(output, v), mode=no_opt_mode)
f_no_opt = theano.function([v, m], tt.jacobian(output, v), mode=no_opt_mode)
# Ensure that the optimization was performed correctly in f_opt
# The inner function of scan should have only one output and it should
......@@ -161,7 +163,7 @@ class TestPushOutScanOutputDot(object):
node for node in f_opt.maker.fgraph.toposort() if isinstance(node.op, Scan)
][0]
assert len(scan_node.op.outputs) == 1
assert not isinstance(scan_node.op.outputs[0], T.Dot)
assert not isinstance(scan_node.op.outputs[0], tt.Dot)
# Ensure that the function compiled with the optimization produces
# the same results as the function compiled without
......@@ -177,12 +179,12 @@ class TestPushOutScanOutputDot(object):
# Test the case where the vector input to the dot is already a nitsot
# output of the inner function.
a = T.matrix()
b = T.matrix()
a = tt.matrix()
b = tt.matrix()
def inner_fct(vect, mat):
vect_squared = vect ** 2
return T.dot(vect_squared, mat), vect_squared
return tt.dot(vect_squared, mat), vect_squared
outputs, updates = theano.scan(
fn=inner_fct, outputs_info=[None] * 2, sequences=a, non_sequences=b
......@@ -205,7 +207,7 @@ class TestPushOutScanOutputDot(object):
# NOTE: WHEN INFER_SHAPE IS REENABLED, BELOW THE SCAN MUST
# HAVE ONLY 1 OUTPUT.
assert len(scan_node.op.outputs) == 2
assert not isinstance(scan_node.op.outputs[0], T.Dot)
assert not isinstance(scan_node.op.outputs[0], tt.Dot)
# Ensure that the function compiled with the optimization produces
# the same results as the function compiled without
......@@ -222,12 +224,12 @@ class TestPushOutScanOutputDot(object):
# Test the case where the vector input to the dot is not already a
# non-nitsot (in this case a sitsot) output of the inner function.
a = T.matrix()
b = T.matrix()
a = tt.matrix()
b = tt.matrix()
def inner_fct(seq1, previous_output1, nonseq1):
output1 = previous_output1 + seq1
output2 = T.dot(output1, nonseq1)
output2 = tt.dot(output1, nonseq1)
return output1, output2
outputs, updates = theano.scan(
......@@ -249,7 +251,7 @@ class TestPushOutScanOutputDot(object):
node for node in f_opt.maker.fgraph.toposort() if isinstance(node.op, Scan)
][0]
assert len(scan_node.op.outputs) == 2
assert not isinstance(scan_node.op.outputs[0], T.Dot)
assert not isinstance(scan_node.op.outputs[0], tt.Dot)
# Ensure that the function compiled with the optimization produces
# the same results as the function compiled without
......@@ -293,18 +295,18 @@ class TestPushOutSumOfDot:
W.name = "W"
# Variables and their values
x = T.tensor3("x")
x = tt.tensor3("x")
x_value = np.random.normal(
size=(seq_len, batch_size, dim), scale=0.0001
).astype(config.floatX)
ri = T.tensor3("ri")
ri = tt.tensor3("ri")
ri_value = x_value
zi = T.tensor3("zi")
zi = tt.tensor3("zi")
zi_value = x_value
init = T.alloc(np.cast[config.floatX](0), batch_size, dim)
init = tt.alloc(np.cast[config.floatX](0), batch_size, dim)
def rnn_step1(
# sequences
......@@ -316,12 +318,12 @@ class TestPushOutSumOfDot:
):
pre_r = ri + h.dot(U)
pre_z = zi + h.dot(V)
r = T.nnet.sigmoid(pre_r)
z = T.nnet.sigmoid(pre_z)
r = tt.nnet.sigmoid(pre_r)
z = tt.nnet.sigmoid(pre_z)
after_r = r * h
pre_h = x + after_r.dot(W)
new_h = T.tanh(pre_h)
new_h = tt.tanh(pre_h)
res_h = z * new_h + (1 - z) * h
return res_h
......@@ -338,7 +340,7 @@ class TestPushOutSumOfDot:
mode=opt_mode,
)
cost = h[-1].sum()
grad1 = T.grad(cost, [U, V, W])
grad1 = tt.grad(cost, [U, V, W])
f_opt = theano.function(inputs=[x, ri, zi], outputs=grad1, mode=opt_mode)
no_opt_mode = mode.excluding("scanOp_pushout_output")
......@@ -351,7 +353,7 @@ class TestPushOutSumOfDot:
mode=no_opt_mode,
)
cost = h[-1].sum()
grad1 = T.grad(cost, [U, V, W])
grad1 = tt.grad(cost, [U, V, W])
f_no_opt = theano.function(inputs=[x, ri, zi], outputs=grad1, mode=no_opt_mode)
# Validate that the optimization has been applied
......@@ -361,8 +363,8 @@ class TestPushOutSumOfDot:
for output in scan_node_grad.op.outputs:
assert not (
isinstance(output.owner.op, T.elemwise.Elemwise)
and any([isinstance(i, T.Dot) for i in output.owner.inputs])
isinstance(output.owner.op, tt.elemwise.Elemwise)
and any([isinstance(i, tt.Dot) for i in output.owner.inputs])
)
# Compare the outputs of the two functions on the same input data.
......@@ -373,20 +375,20 @@ class TestPushOutSumOfDot:
def test_non_zero_init(self):
# Test the case where the initial value for the nitsot output is non-zero
input1 = T.tensor3()
input2 = T.tensor3()
input3 = T.tensor3()
input1 = tt.tensor3()
input2 = tt.tensor3()
input3 = tt.tensor3()
W = theano.shared(np.random.normal(size=(4, 5))).astype(config.floatX)
U = theano.shared(np.random.normal(size=(6, 7))).astype(config.floatX)
def inner_fct(seq1, seq2, seq3, previous_output):
temp1 = T.dot(seq1, W) + seq3
temp2 = T.dot(seq2, U)
dot_output = T.dot(temp1, temp2)
temp1 = tt.dot(seq1, W) + seq3
temp2 = tt.dot(seq2, U)
dot_output = tt.dot(temp1, temp2)
return previous_output + dot_output
init = T.as_tensor_variable(np.random.normal(size=(3, 7)))
init = tt.as_tensor_variable(np.random.normal(size=(3, 7)))
# Compile the function twice, once with the optimization and once
# without
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论