cavis/libnd4j/tests_cpu/layers_tests/GraphStateTests.cpp
raver119 3c4e959e21 [WIP] More of CUDA (#95)
* initial commit

Signed-off-by: raver119 <raver119@gmail.com>

* Implementation of hashcode cuda helper. Working edition.

* Fixed parallel test input arangements.

* Fixed tests for hashcode op.

* Fixed shape calculation for image:crop_and_resize op and test.

* NativeOps tests. Initial test suite.

* Added tests for indexReduce methods.

* Added test on execBroadcast with NDArray as dimensions.

* Added test on execBroadcastBool with NDArray as dimensions.

* Added tests on execPairwiseTransform and execPairwiseTransofrmBool.

* Added tests for execReduce with scalar results.

* Added reduce tests for non-empty dims array.

* Added tests for reduce3.

* Added tests for execScalar.

* Added tests for execSummaryStats.

* - provide cpu/cuda code for batch_to_space
- testing it

Signed-off-by: Yurii <yurii@skymind.io>

* - remove old test for batch_to_space (had wrong format and numbers were not checked)

Signed-off-by: Yurii <yurii@skymind.io>

* Fixed complilation errors with test.

* Added test for execTransformFloat.

* Added test for execTransformSame.

* Added test for execTransformBool.

* Added test for execTransformStrict.

* Added tests for execScalar/execScalarBool with TADs.

* Added test for flatten.

* - provide cpu/cuda code for space_to_Batch operaion

Signed-off-by: Yurii <yurii@skymind.io>

* Added test for concat.

* comment unnecessary stuff in s_t_b

Signed-off-by: Yurii <yurii@skymind.io>

* Added test for specialConcat.

* Added tests for memcpy/set routines.

* Fixed pullRow cuda test.

* Added pullRow test.

* Added average test.

* - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...)

Signed-off-by: Yurii <yurii@skymind.io>

* - debugging and fixing cuda tests in JavaInteropTests file

Signed-off-by: Yurii <yurii@skymind.io>

* - correct some tests

Signed-off-by: Yurii <yurii@skymind.io>

* Added test for shuffle.

* Fixed ops declarations.

* Restored omp and added shuffle test.

* Added convertTypes test.

* Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps.

* Added sort tests.

* Added tests for execCustomOp.

* - further debuging and fixing tests terminated with crash

Signed-off-by: Yurii <yurii@skymind.io>

* Added tests for calculateOutputShapes.

* Addded Benchmarks test.

* Commented benchmark tests.

* change assertion

Signed-off-by: raver119 <raver119@gmail.com>

* Added tests for apply_sgd op. Added cpu helper for that op.

* Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps.

* Added test for assign broadcastable.

* Added tests for assign_bp op.

* Added tests for axpy op.

* - assign/execScalar/execTransformAny signature change
- minor test fix

Signed-off-by: raver119 <raver119@gmail.com>

* Fixed axpy op.

* meh

Signed-off-by: raver119 <raver119@gmail.com>

* - fix tests for nativeOps::concat

Signed-off-by: Yurii <yurii@skymind.io>

* sequential transform/scalar

Signed-off-by: raver119 <raver119@gmail.com>

* allow nested parallelism

Signed-off-by: raver119 <raver119@gmail.com>

* assign_bp leak fix

Signed-off-by: raver119 <raver119@gmail.com>

* block setRNG fix

Signed-off-by: raver119 <raver119@gmail.com>

* enable parallelism by default

Signed-off-by: raver119 <raver119@gmail.com>

* enable nested parallelism by default

Signed-off-by: raver119 <raver119@gmail.com>

* Added cuda implementation for row_count helper.

* Added implementation for tnse gains op helper.

* - take into account possible situations when input arrays are empty in reduce_ cuda stuff

Signed-off-by: Yurii <yurii@skymind.io>

* Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces.

* Added kernel for tsne/symmetrized op heleper.

* Implementation of tsne/symmetrized op cuda helper. Working edition.

* Eliminated waste printfs.

* Added test for broadcastgradientargs op.

* host-only fallback for empty reduce float

Signed-off-by: raver119 <raver119@gmail.com>

* - some tests fixes

Signed-off-by: Yurii <yurii@skymind.io>

* - correct the rest of reduce_ stuff

Signed-off-by: Yurii <yurii@skymind.io>

* - further correction of reduce_ stuff

Signed-off-by: Yurii <yurii@skymind.io>

* Added test for Cbow op. Also added cuda implementation for cbow helpers.

* - improve code of stack operation for scalar case

Signed-off-by: Yurii <yurii@skymind.io>

* - provide cuda kernel for gatherND operation

Signed-off-by: Yurii <yurii@skymind.io>

* Implementation of cbow helpers with cuda kernels.

* minor tests tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* minor tests tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* - further correction of cuda stuff

Signed-off-by: Yurii <yurii@skymind.io>

* Implementatation of cbow op helper with cuda kernels. Working edition.

* Skip random testing for cudablas case.

* lstmBlockCell context fix

Signed-off-by: raver119 <raver119@gmail.com>

* Added tests for ELU and ELU_BP ops.

* Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops.

* Added tests for neq_scalar.

* Added test for noop.

* - further work on clipbynorm_bp

Signed-off-by: Yurii <yurii@skymind.io>

* - get rid of concat op call, use instead direct concat helper call

Signed-off-by: Yurii <yurii@skymind.io>

* lstmBlockCell context fix

Signed-off-by: raver119 <raver119@gmail.com>

* Added tests for lrelu and lrelu_bp.

* Added tests for selu and selu_bp.

* Fixed lrelu derivative helpers.

* - some corrections in lstm

Signed-off-by: Yurii <yurii@skymind.io>

* operator * result shape fix

Signed-off-by: raver119 <raver119@gmail.com>

* - correct typo in lstmCell

Signed-off-by: Yurii <yurii@skymind.io>

* few tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA inverse broadcast bool fix

Signed-off-by: raver119 <raver119@gmail.com>

* disable MMAP test for CUDA

Signed-off-by: raver119 <raver119@gmail.com>

* BooleanOp syncToDevice

Signed-off-by: raver119 <raver119@gmail.com>

* meh

Signed-off-by: raver119 <raver119@gmail.com>

* additional data types for im2col/col2im

Signed-off-by: raver119 <raver119@gmail.com>

* Added test for firas_sparse op.

* one more RandomBuffer test excluded

Signed-off-by: raver119 <raver119@gmail.com>

* Added tests for flatten op.

* Added test for Floor op.

* bunch of tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* mmulDot tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* more tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* Implemented floordiv_bp op and tests.

* Fixed scalar case with cuda implementation for bds.

* - work on cuda kernel for clip_by_norm backprop op is completed

Signed-off-by: Yurii <yurii@skymind.io>

* Eliminate cbow crach.

* more tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* more tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* Eliminated abortion with batched nlp test.

* more tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* Fixed shared flag initializing.

* disabled bunch of cpu workspaces tests

Signed-off-by: raver119 <raver119@gmail.com>

* scalar operators fix: missing registerSpecialUse call

Signed-off-by: raver119 <raver119@gmail.com>

* Fixed logdet for cuda and tests.

* - correct clipBynorm_bp

Signed-off-by: Yurii <yurii@skymind.io>

* Fixed crop_and_resize shape datatype.

* - correct some mmul tests

Signed-off-by: Yurii <yurii@skymind.io>
2019-08-05 11:27:05 +10:00

349 lines
11 KiB
C++

/*******************************************************************************
* Copyright (c) 2015-2018 Skymind, Inc.
*
* This program and the accompanying materials are made available under the
* terms of the Apache License, Version 2.0 which is available at
* https://www.apache.org/licenses/LICENSE-2.0.
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
* License for the specific language governing permissions and limitations
* under the License.
*
* SPDX-License-Identifier: Apache-2.0
******************************************************************************/
//
// @author raver119@gmail.com
//
#include "testlayers.h"
#include <graph/GraphState.h>
#include <ops/declarable/CustomOperations.h>
#include <ops/declarable/LegacyTransformOp.h>
#include <ops/declarable/LegacyReduceOp.h>
#include <NativeOps.h>
using namespace nd4j;
using namespace nd4j::graph;
class GraphStateTests : public testing::Test {
public:
GraphStateTests() {
Environment::getInstance()->setDebug(false);
Environment::getInstance()->setVerbose(false);
};
~GraphStateTests() {
Environment::getInstance()->setDebug(false);
Environment::getInstance()->setVerbose(false);
}
};
/*
* PLAN:
* Create GraphState
* Register Scope
* Add few Ops to it
* Call conditional, that refers to scopes
* Check results
*/
TEST_F(GraphStateTests, Basic_Tests_1) {
auto state = (GraphState *) getGraphState(117L);
ASSERT_EQ(117L, state->id());
// this call will create scope internally
state->registerScope(119);
nd4j::ops::add opA;
nd4j::ops::LegacyTransformSameOp opB(transform::Neg); // simdOps::Neg
ArgumentsList argsA;
ArgumentsList argsB;
state->attachOpToScope(119, 1, &opA, argsA);
state->attachOpToScope(119, 2, &opB, argsB);
auto scope = state->getScope(119);
ASSERT_TRUE(scope != nullptr);
ASSERT_EQ(2, scope->size());
deleteGraphState(state);
}
// just separate case for doubles wrapper in NativeOps, nothing else
TEST_F(GraphStateTests, Basic_Tests_2) {
auto state = (GraphState *) getGraphState(117L);
ASSERT_EQ(117L, state->id());
// this call will create scope internally
state->registerScope(119);
nd4j::ops::add opA;
nd4j::ops::LegacyTransformSameOp opB(transform::Neg); // simdOps::Neg
ArgumentsList argsA;
ArgumentsList argsB;
state->attachOpToScope(119, 1, &opA, argsA);
state->attachOpToScope(119, 2, &opB, argsB);
auto scope = state->getScope(119);
ASSERT_TRUE(scope != nullptr);
ASSERT_EQ(2, scope->size());
deleteGraphState(state);
}
/*
TEST_F(GraphStateTests, Stateful_Execution_1) {
auto state = getGraphState(117L);
Nd4jLong scopes[] = {22, 33};
//auto status = execCustomOpWithScope(nullptr, state, 10, scopes, 2, nullptr, nullptr, 0, nullptr, nullptr, 0);
auto status = execCustomOpWithScope(nullptr, state, 10, scopes, 2, nullptr, nullptr, 0, nullptr, nullptr, 0);
ASSERT_EQ(Status::THROW(), status);
deleteGraphState(state);
}
TEST_F(GraphStateTests, Stateful_Execution_2) {
auto state = (GraphState *) getGraphState(117L);
state->registerScope(22);
state->registerScope(33);
Nd4jLong scopes[] = {22, 33};
auto status = execCustomOpWithScope(nullptr, state, 10, scopes, 2, nullptr, nullptr, 0, nullptr, nullptr, 0);
// it's no-op: just LogicScope
ASSERT_EQ(Status::OK(), status);
deleteGraphState(state);
}
// This test checks WHILE loop
TEST_F(GraphStateTests, Stateful_Execution_3) {
auto var0 = NDArrayFactory::create<float>('c', {2, 2}, {1, 2, 3, 4});
auto var1 = NDArrayFactory::create<float>(11.0f);
auto var2 = NDArrayFactory::create<float>(2.0f);
auto res0 = NDArrayFactory::create<float>('c', {2, 2});
auto res1 = NDArrayFactory::create<float>(0.0f);
auto res2 = NDArrayFactory::create<float>(0.0f);
// registering our GraphState holder
auto state = (GraphState *) getGraphState(117L);
// we're prepping pointers to input/output buffers
Nd4jPointer ptrBuffers[] = {(Nd4jPointer) var0.buffer(), (Nd4jPointer) var1.buffer(), (Nd4jPointer)var2.buffer()};
Nd4jPointer ptrShapes[] = {(Nd4jPointer) var0.shapeInfo(), (Nd4jPointer) var1.shapeInfo(), (Nd4jPointer)var2.shapeInfo()};
Nd4jPointer outBuffers[] = {(Nd4jPointer) res0.buffer(), (Nd4jPointer) res1.buffer(), (Nd4jPointer) res2.buffer()};
Nd4jPointer outShapes[] = {(Nd4jPointer) res0.shapeInfo(), (Nd4jPointer) res1.shapeInfo(), (Nd4jPointer) res2.shapeInfo()};
// conditional scope
state->registerScope(22);
nd4j::ops::LegacyReduceSameOp op1(reduce::Sum);
nd4j::ops::lt_scalar op2;
// while sum(var0) < var1
// this op takes sum
ArgumentsList args1({{0, 0}});
// this op compares result of sum to input variable 0:1
ArgumentsList args2({{1, 0}, {0, 1}});
state->attachOpToScope(22, 1, &op1, args1);
state->attachOpToScope(22, 2, &op2, args2);
// body scope
state->registerScope(33);
// var0 + var1 + var1
// this op is var0 + var1
ArgumentsList args3({{0, 0}, {0, 2}});
// this op is result of previous op + 1
ArgumentsList args4({{3, 0}, {0, 2}});
nd4j::ops::add op3;
nd4j::ops::add op4;
state->attachOpToScope(33, 3, &op3, args3);
state->attachOpToScope(33, 4, &op4, args4);
// Now we define RETURN, which returns 1 modified variable, and 2 unmodified variables
ArgumentsList args5({{4, 0}, {0, 1}, {0, 2}});
// so, at the end of body, initial variables will be updated
state->defineReturn(33, 5, args5);
Nd4jLong scopes[] = {22, 33};
// we're executing while loop
auto status = execCustomOpWithScope(nullptr, state, 0, scopes, 2, ptrBuffers, ptrShapes, 3, outBuffers, outShapes, 3);
ASSERT_EQ(Status::OK(), status);
// now we check provided result array
float sum = res0.reduceNumber(reduce::Sum).e<float>(0);
// Expected result is {1, 2, 3, 4} + {2} elementwise + {2} elementwise, which gives { 5, 6, 7, 8}, and sum should be 26
ASSERT_NEAR(26.0f, sum, 1e-5);
// nd4j_printf("0 ------------------\n","");
deleteGraphState(state);
// nd4j_printf("1 ------------------\n","");
}
// This test checks CONDITIONAL execution for FALSE
TEST_F(GraphStateTests, Stateful_Execution_4) {
auto var0 = NDArrayFactory::create<float>('c', {2, 2}, {1, 2, 3, 4});
auto var1 = NDArrayFactory::create<float>(5.0f);
auto res0 = NDArrayFactory::create<float>('c', {2, 2});
auto res1 = NDArrayFactory::create<float>(0.0f);
auto exp = NDArrayFactory::create<float>('c', {2, 2}, {-4, -3, -2, -1});
// registering our GraphState holder
auto state = (GraphState *) getGraphState(117L);
// we're prepping pointers to input/output buffers
Nd4jPointer ptrBuffers[] = {(Nd4jPointer) var0.buffer(), (Nd4jPointer) var1.buffer()};
Nd4jPointer ptrShapes[] = {(Nd4jPointer) var0.shapeInfo(), (Nd4jPointer) var1.shapeInfo()};
Nd4jPointer outBuffers[] = {(Nd4jPointer) res0.buffer(), (Nd4jPointer) res1.buffer()};
Nd4jPointer outShapes[] = {(Nd4jPointer) res0.shapeInfo(), (Nd4jPointer) res1.shapeInfo()};
// conditional scope
state->registerScope(22);
nd4j::ops::LegacyReduceSameOp op1(reduce::Sum);
nd4j::ops::lt_scalar op2;
// if sum(var0) < var1
// this op takes sum
ArgumentsList args1({{0, 0}});
// this op compares result of sum to input variable 0:1
ArgumentsList args2({{1, 0}, {0, 1}});
state->attachOpToScope(22, 1, &op1, args1);
state->attachOpToScope(22, 2, &op2, args2);
// false scope
state->registerScope(33);
ArgumentsList args3({{0, 0}, {0, 1}});
nd4j::ops::subtract op3;
state->attachOpToScope(33, 3, &op3, args3);
// return for false scope
ArgumentsList args10({{3, 0}, {0, 1}});
state->defineReturn(33, 10, args10);
// true scope
state->registerScope(44);
ArgumentsList args4({{0, 0}, {0, 1}});
nd4j::ops::add op4;
state->attachOpToScope(44, 4, &op4, args4);
// return for false scope
ArgumentsList args20({{4, 0}, {0, 1}});
state->defineReturn(44, 20, args20);
Nd4jLong scopes[] = {22, 33, 44};
// we're executing conditional op
auto status = execCustomOpWithScope(nullptr, state, 20, scopes, 3, ptrBuffers, ptrShapes, 2, outBuffers, outShapes, 2);
ASSERT_EQ(Status::OK(), status);
ASSERT_TRUE(exp.isSameShape(&res0));
ASSERT_TRUE(exp.equalsTo(&res0));
deleteGraphState(state);
}
// This test checks CONDITIONAL execution for TRUE
TEST_F(GraphStateTests, Stateful_Execution_5) {
auto var0 = NDArrayFactory::create<float>('c', {2, 2}, {1, 2, 3, 4});
auto var1 = NDArrayFactory::create<float>(5.0f);
auto res0 = NDArrayFactory::create<float>('c', {2, 2});
auto res1 = NDArrayFactory::create<float>(0.0f);
auto exp = NDArrayFactory::create<float>('c', {2, 2}, {6, 7, 8, 9});
// registering our GraphState holder
auto state = (GraphState *) getGraphState(117L);
// we're prepping pointers to input/output buffers
Nd4jPointer ptrBuffers[] = {(Nd4jPointer) var0.buffer(), (Nd4jPointer) var1.buffer()};
Nd4jPointer ptrShapes[] = {(Nd4jPointer) var0.shapeInfo(), (Nd4jPointer) var1.shapeInfo()};
Nd4jPointer outBuffers[] = {(Nd4jPointer) res0.buffer(), (Nd4jPointer) res1.buffer()};
Nd4jPointer outShapes[] = {(Nd4jPointer) res0.shapeInfo(), (Nd4jPointer) res1.shapeInfo()};
// conditional scope
state->registerScope(22);
nd4j::ops::LegacyReduceSameOp op1(reduce::Sum);
nd4j::ops::gt_scalar op2;
// if sum(var0) < var1
// this op takes sum
ArgumentsList args1({{0, 0}});
// this op compares result of sum to input variable 0:1
ArgumentsList args2({{1, 0}, {0, 1}});
state->attachOpToScope(22, 1, &op1, args1);
state->attachOpToScope(22, 2, &op2, args2);
// false scope
state->registerScope(33);
ArgumentsList args3({{0, 0}, {0, 1}});
nd4j::ops::subtract op3;
state->attachOpToScope(33, 3, &op3, args3);
// return for false scope
ArgumentsList args10({{3, 0}, {0, 1}});
state->defineReturn(33, 10, args10);
// true scope
state->registerScope(44);
ArgumentsList args4({{0, 0}, {0, 1}});
nd4j::ops::add op4;
state->attachOpToScope(44, 4, &op4, args4);
// return for false scope
ArgumentsList args20({{4, 0}, {0, 1}});
state->defineReturn(44, 20, args20);
Nd4jLong scopes[] = {22, 33, 44};
// we're executing conditional op
auto status = execCustomOpWithScope(nullptr, state, 20, scopes, 3, ptrBuffers, ptrShapes, 2, outBuffers, outShapes, 2);
ASSERT_EQ(Status::OK(), status);
ASSERT_TRUE(exp.isSameShape(&res0));
ASSERT_TRUE(exp.equalsTo(&res0));
deleteGraphState(state);
}
*/