!17259 MD CI code warning fixes

From: @cathwong
Reviewed-by: @robingrosman,@pandoublefeng
Signed-off-by: @robingrosman
This commit is contained in:
mindspore-ci-bot 2021-05-30 05:48:04 +08:00 committed by Gitee
commit a0d47b4dea
11 changed files with 28 additions and 22 deletions

View File

@ -241,16 +241,16 @@ PYBIND_REGISTER(TextFileNode, 2, ([](const py::module *m) {
PYBIND_REGISTER(TFRecordNode, 2, ([](const py::module *m) {
(void)py::class_<TFRecordNode, DatasetNode, std::shared_ptr<TFRecordNode>>(*m, "TFRecordNode",
"to create a TFRecordNode")
.def(py::init([](py::list dataset_files, std::shared_ptr<SchemaObj> schema, py::list columns_list,
int64_t num_samples, int32_t shuffle, int32_t num_shards, int32_t shard_id,
bool shard_equal_rows) {
.def(py::init([](const py::list dataset_files, std::shared_ptr<SchemaObj> schema,
const py::list columns_list, int64_t num_samples, int32_t shuffle,
int32_t num_shards, int32_t shard_id, bool shard_equal_rows) {
std::shared_ptr<TFRecordNode> tfrecord = std::make_shared<TFRecordNode>(
toStringVector(dataset_files), schema, toStringVector(columns_list), num_samples,
toShuffleMode(shuffle), num_shards, shard_id, shard_equal_rows, nullptr);
THROW_IF_ERROR(tfrecord->ValidateParams());
return tfrecord;
}))
.def(py::init([](py::list dataset_files, std::string schema, py::list columns_list,
.def(py::init([](const py::list dataset_files, std::string schema, py::list columns_list,
int64_t num_samples, int32_t shuffle, int32_t num_shards, int32_t shard_id,
bool shard_equal_rows) {
std::shared_ptr<TFRecordNode> tfrecord = std::make_shared<TFRecordNode>(

View File

@ -160,7 +160,7 @@ std::vector<std::shared_ptr<DatasetNode>> toDatasetNode(std::shared_ptr<DatasetN
return vector;
}
std::shared_ptr<SamplerObj> toSamplerObj(py::handle py_sampler, bool isMindDataset) {
std::shared_ptr<SamplerObj> toSamplerObj(const py::handle py_sampler, bool isMindDataset) {
if (py_sampler.is_none()) {
return nullptr;
}

View File

@ -71,7 +71,7 @@ std::shared_ptr<TensorOperation> toTensorOperation(py::handle operation);
std::vector<std::shared_ptr<DatasetNode>> toDatasetNode(std::shared_ptr<DatasetNode> self, py::list datasets);
std::shared_ptr<SamplerObj> toSamplerObj(py::handle py_sampler, bool isMindDataset = false);
std::shared_ptr<SamplerObj> toSamplerObj(const py::handle py_sampler, bool isMindDataset = false);
std::shared_ptr<DatasetCache> toDatasetCache(std::shared_ptr<CacheClient> cc);

View File

@ -219,7 +219,7 @@ RandomChoice::RandomChoice(const std::vector<TensorTransform *> &transforms) : d
RandomChoice::RandomChoice(const std::vector<std::shared_ptr<TensorTransform>> &transforms)
: data_(std::make_shared<Data>()) {
(void)std::transform(transforms.begin(), transforms.end(), std::back_inserter(data_->transforms_),
[](std::shared_ptr<TensorTransform> op) -> std::shared_ptr<TensorOperation> {
[](const std::shared_ptr<TensorTransform> op) -> std::shared_ptr<TensorOperation> {
return op != nullptr ? op->Parse() : nullptr;
});
}

View File

@ -658,10 +658,10 @@ struct RandomSelectSubpolicy::Data {
RandomSelectSubpolicy::RandomSelectSubpolicy(
const std::vector<std::vector<std::pair<TensorTransform *, double>>> &policy)
: data_(std::make_shared<Data>()) {
for (int32_t i = 0; i < policy.size(); i++) {
for (uint32_t i = 0; i < policy.size(); i++) {
std::vector<std::pair<std::shared_ptr<TensorOperation>, double>> subpolicy;
for (int32_t j = 0; j < policy[i].size(); j++) {
for (uint32_t j = 0; j < policy[i].size(); j++) {
TensorTransform *op = policy[i][j].first;
std::shared_ptr<TensorOperation> operation = (op ? op->Parse() : nullptr);
double prob = policy[i][j].second;
@ -674,10 +674,10 @@ RandomSelectSubpolicy::RandomSelectSubpolicy(
RandomSelectSubpolicy::RandomSelectSubpolicy(
const std::vector<std::vector<std::pair<std::shared_ptr<TensorTransform>, double>>> &policy)
: data_(std::make_shared<Data>()) {
for (int32_t i = 0; i < policy.size(); i++) {
for (uint32_t i = 0; i < policy.size(); i++) {
std::vector<std::pair<std::shared_ptr<TensorOperation>, double>> subpolicy;
for (int32_t j = 0; j < policy[i].size(); j++) {
for (uint32_t j = 0; j < policy[i].size(); j++) {
std::shared_ptr<TensorTransform> op = policy[i][j].first;
std::shared_ptr<TensorOperation> operation = (op ? op->Parse() : nullptr);
double prob = policy[i][j].second;
@ -900,9 +900,10 @@ UniformAugment::UniformAugment(const std::vector<TensorTransform *> &transforms,
UniformAugment::UniformAugment(const std::vector<std::shared_ptr<TensorTransform>> &transforms, int32_t num_ops)
: data_(std::make_shared<Data>()) {
(void)std::transform(
transforms.begin(), transforms.end(), std::back_inserter(data_->transforms_),
[](std::shared_ptr<TensorTransform> op) -> std::shared_ptr<TensorOperation> { return op ? op->Parse() : nullptr; });
(void)std::transform(transforms.begin(), transforms.end(), std::back_inserter(data_->transforms_),
[](const std::shared_ptr<TensorTransform> op) -> std::shared_ptr<TensorOperation> {
return op ? op->Parse() : nullptr;
});
data_->num_ops_ = num_ops;
}

View File

@ -229,7 +229,7 @@ class ImageFolderOp : public MappableLeafOp {
std::map<std::string, int32_t> class_index_;
std::unique_ptr<DataSchema> data_schema_;
int64_t sampler_ind_;
int64_t dirname_offset_;
uint64_t dirname_offset_;
std::vector<ImageLabelPair> image_label_pairs_;
std::unique_ptr<Queue<std::string>> folder_name_queue_;
std::unique_ptr<Queue<FolderImagesPair>> image_name_queue_;

View File

@ -18,7 +18,7 @@
namespace mindspore {
namespace dataset {
/* ####################################### Validator Functions ############################################ */
Status ValidateProbability(const std::string &op_name, const float probability) {
Status ValidateProbability(const std::string &op_name, const double probability) {
if (probability < 0.0 || probability > 1.0) {
std::string err_msg = op_name + ": probability must be between 0.0 and 1.0, got: " + std::to_string(probability);
MS_LOG(ERROR) << err_msg;

View File

@ -28,7 +28,7 @@
namespace mindspore {
namespace dataset {
// Helper function to validate probability
Status ValidateProbability(const std::string &op_name, const float probability);
Status ValidateProbability(const std::string &op_name, const double probability);
// Helper function to positive int scalar
Status ValidateIntScalarPositive(const std::string &op_name, const std::string &scalar_name, int32_t scalar);

View File

@ -476,4 +476,5 @@ def check_c_tensor_op(param, param_name):
def replace_none(value, default):
"""Replace None value with default"""
return value if value is not None else default

View File

@ -50,6 +50,7 @@ class TensorOperation:
return output_tensor_list[0] if len(output_tensor_list) == 1 else tuple(output_tensor_list)
def parse(self):
"""parse function - not yet implemented"""
raise NotImplementedError("TensorOperation has to implement parse() method.")
@ -307,7 +308,8 @@ class Concatenate(TensorOperation):
Args:
axis (int, optional): Concatenate the tensors along given axis (Default=0).
prepend (numpy.array, optional): NumPy array to be prepended to the already concatenated tensors (Default=None).
prepend (numpy.array, optional): NumPy array to be prepended to the already concatenated tensors
(Default=None).
append (numpy.array, optional): NumPy array to be appended to the already concatenated tensors (Default=None).
Examples:

View File

@ -271,8 +271,9 @@ def check_random_apply(method):
for i, transform in enumerate(transforms):
if str(transform).find("c_transform") >= 0:
raise ValueError("transforms[{}] is not a py transforms. Should not use a c transform in py transform" \
.format(i))
raise ValueError(
"transforms[{}] is not a py transforms. Should not use a c transform in py transform" \
.format(i))
if prob is not None:
type_check(prob, (float, int,), "prob")
@ -293,8 +294,9 @@ def check_transforms_list(method):
type_check(transforms, (list,), "transforms")
for i, transform in enumerate(transforms):
if str(transform).find("c_transform") >= 0:
raise ValueError("transforms[{}] is not a py transforms. Should not use a c transform in py transform" \
.format(i))
raise ValueError(
"transforms[{}] is not a py transforms. Should not use a c transform in py transform" \
.format(i))
return method(self, *args, **kwargs)
return new_method