[feat][assistant][I3J6V2] add new data operator EMnist

mindspore/ccsrc/minddata/dataset/api/datasets.cc

@@ -96,6 +96,7 @@
 #include "minddata/dataset/engine/ir/datasetops/source/coco_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/csv_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/div2k_node.h"
+#include "minddata/dataset/engine/ir/datasetops/source/emnist_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/flickr_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/image_folder_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/random_node.h"
@@ -1042,6 +1043,33 @@ DIV2KDataset::DIV2KDataset(const std::vector<char> &dataset_dir, const std::vect
   ir_node_ = std::static_pointer_cast<DatasetNode>(ds);
 }
 
+EMnistDataset::EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                             const std::vector<char> &usage, const std::shared_ptr<Sampler> &sampler,
+                             const std::shared_ptr<DatasetCache> &cache) {
+  auto sampler_obj = sampler ? sampler->Parse() : nullptr;
+  auto ds = std::make_shared<EMnistNode>(CharToString(dataset_dir), CharToString(name), CharToString(usage),
+                                         sampler_obj, cache);
+  ir_node_ = std::static_pointer_cast<DatasetNode>(ds);
+}
+
+EMnistDataset::EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                             const std::vector<char> &usage, const Sampler *sampler,
+                             const std::shared_ptr<DatasetCache> &cache) {
+  auto sampler_obj = sampler ? sampler->Parse() : nullptr;
+  auto ds = std::make_shared<EMnistNode>(CharToString(dataset_dir), CharToString(name), CharToString(usage),
+                                         sampler_obj, cache);
+  ir_node_ = std::static_pointer_cast<DatasetNode>(ds);
+}
+
+EMnistDataset::EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                             const std::vector<char> &usage, const std::reference_wrapper<Sampler> sampler,
+                             const std::shared_ptr<DatasetCache> &cache) {
+  auto sampler_obj = sampler.get().Parse();
+  auto ds = std::make_shared<EMnistNode>(CharToString(dataset_dir), CharToString(name), CharToString(usage),
+                                         sampler_obj, cache);
+  ir_node_ = std::static_pointer_cast<DatasetNode>(ds);
+}
+
 FlickrDataset::FlickrDataset(const std::vector<char> &dataset_dir, const std::vector<char> &annotation_file,
                              bool decode, const std::shared_ptr<Sampler> &sampler,
                              const std::shared_ptr<DatasetCache> &cache) {

mindspore/ccsrc/minddata/dataset/api/python/bindings/dataset/engine/ir/datasetops/source/bindings.cc

@@ -33,6 +33,7 @@
 #include "minddata/dataset/engine/ir/datasetops/source/coco_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/csv_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/div2k_node.h"
+#include "minddata/dataset/engine/ir/datasetops/source/emnist_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/flickr_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/generator_node.h"
 #include "minddata/dataset/engine/ir/datasetops/source/image_folder_node.h"
@@ -152,6 +153,17 @@ PYBIND_REGISTER(DIV2KNode, 2, ([](const py::module *m) {
                     }));
                 }));
 
+PYBIND_REGISTER(EMnistNode, 2, ([](const py::module *m) {
+                  (void)py::class_<EMnistNode, DatasetNode, std::shared_ptr<EMnistNode>>(*m, "EMnistNode",
+                                                                                         "to create an EMnistNode")
+                    .def(py::init([](std::string dataset_dir, std::string name, std::string usage, py::handle sampler) {
+                      auto emnist =
+                        std::make_shared<EMnistNode>(dataset_dir, name, usage, toSamplerObj(sampler), nullptr);
+                      THROW_IF_ERROR(emnist->ValidateParams());
+                      return emnist;
+                    }));
+                }));
+
 PYBIND_REGISTER(
   FlickrNode, 2, ([](const py::module *m) {
     (void)py::class_<FlickrNode, DatasetNode, std::shared_ptr<FlickrNode>>(*m, "FlickrNode", "to create a FlickrNode")

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/CMakeLists.txt

@@ -22,6 +22,7 @@ set(DATASET_ENGINE_DATASETOPS_SOURCE_SRC_FILES
     div2k_op.cc
     flickr_op.cc
     qmnist_op.cc
+    emnist_op.cc
     )
 
 set(DATASET_ENGINE_DATASETOPS_SOURCE_SRC_FILES

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/emnist_op.cc

@@ -0,0 +1,146 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://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.
+ */
+#include "minddata/dataset/engine/datasetops/source/emnist_op.h"
+
+#include <algorithm>
+#include <fstream>
+#include <iomanip>
+#include <set>
+#include <utility>
+
+#include "debug/common.h"
+#include "minddata/dataset/core/config_manager.h"
+#include "minddata/dataset/core/tensor_shape.h"
+#include "minddata/dataset/engine/datasetops/source/sampler/sequential_sampler.h"
+#include "minddata/dataset/engine/execution_tree.h"
+#include "utils/file_utils.h"
+#include "utils/ms_utils.h"
+
+namespace mindspore {
+namespace dataset {
+EMnistOp::EMnistOp(const std::string &name, const std::string &usage, int32_t num_workers,
+                   const std::string &folder_path, int32_t queue_size, std::unique_ptr<DataSchema> data_schema,
+                   std::shared_ptr<SamplerRT> sampler)
+    : MnistOp(usage, num_workers, folder_path, queue_size, std::move(data_schema), std::move(sampler)), name_(name) {}
+
+void EMnistOp::Print(std::ostream &out, bool show_all) const {
+  if (!show_all) {
+    // Call the super class for displaying any common 1-liner info.
+    ParallelOp::Print(out, show_all);
+    // Then show any custom derived-internal 1-liner info for this op.
+    out << "\n";
+  } else {
+    // Call the super class for displaying any common detailed info.
+    ParallelOp::Print(out, show_all);
+    // Then show any custom derived-internal stuff.
+    out << "\nNumber of rows:" << num_rows_ << "\n"
+        << DatasetName(true) << " directory: " << folder_path_ << "\nName: " << name_ << "\nUsage: " << usage_
+        << "\n\n";
+  }
+}
+
+Status EMnistOp::WalkAllFiles() {
+  const std::string img_ext = "-images-idx3-ubyte";
+  const std::string lbl_ext = "-labels-idx1-ubyte";
+  const std::string train_prefix = "-train";
+  const std::string test_prefix = "-test";
+  auto realpath = FileUtils::GetRealPath(folder_path_.data());
+  CHECK_FAIL_RETURN_UNEXPECTED(realpath.has_value(), "Get real path failed: " + folder_path_);
+  Path dir(realpath.value());
+  auto dir_it = Path::DirIterator::OpenDirectory(&dir);
+  if (dir_it == nullptr) {
+    RETURN_STATUS_UNEXPECTED("Invalid path, failed to open directory: " + dir.ToString());
+  }
+  std::string prefix;
+  prefix = "emnist-" + name_;  // used to match usage == "all".
+  if (usage_ == "train" || usage_ == "test") {
+    prefix += (usage_ == "test" ? test_prefix : train_prefix);
+  }
+  if (dir_it != nullptr) {
+    while (dir_it->HasNext()) {
+      Path file = dir_it->Next();
+      std::string fname = file.Basename();  // name of the emnist file.
+      if ((fname.find(prefix) != std::string::npos) && (fname.find(img_ext) != std::string::npos)) {
+        image_names_.push_back(file.ToString());
+        MS_LOG(INFO) << DatasetName(true) << " operator found image file at " << fname << ".";
+      } else if ((fname.find(prefix) != std::string::npos) && (fname.find(lbl_ext) != std::string::npos)) {
+        label_names_.push_back(file.ToString());
+        MS_LOG(INFO) << DatasetName(true) << " operator found label file at " << fname << ".";
+      }
+    }
+  } else {
+    MS_LOG(WARNING) << DatasetName(true) << " operator unable to open directory " << dir.ToString() << ".";
+  }
+
+  std::sort(image_names_.begin(), image_names_.end());
+  std::sort(label_names_.begin(), label_names_.end());
+  CHECK_FAIL_RETURN_UNEXPECTED(image_names_.size() == label_names_.size(),
+                               "Invalid data, num of images does not equal to num of labels.");
+
+  return Status::OK();
+}
+
+Status EMnistOp::CountTotalRows(const std::string &dir, const std::string &name, const std::string &usage,
+                                int64_t *count) {
+  // the logic of counting the number of samples is copied from ParseEMnistData() and uses CheckReader().
+  RETURN_UNEXPECTED_IF_NULL(count);
+  *count = 0;
+
+  const int64_t num_samples = 0;
+  const int64_t start_index = 0;
+  auto sampler = std::make_shared<SequentialSamplerRT>(start_index, num_samples);
+  auto schema = std::make_unique<DataSchema>();
+  RETURN_IF_NOT_OK(schema->AddColumn(ColDescriptor("image", DataType(DataType::DE_UINT8), TensorImpl::kCv, 1)));
+  TensorShape scalar = TensorShape::CreateScalar();
+  RETURN_IF_NOT_OK(
+    schema->AddColumn(ColDescriptor("label", DataType(DataType::DE_UINT32), TensorImpl::kFlexible, 0, &scalar)));
+  std::shared_ptr<ConfigManager> cfg = GlobalContext::config_manager();
+  int32_t num_workers = cfg->num_parallel_workers();
+  int32_t op_connect_size = cfg->op_connector_size();
+  auto op =
+    std::make_shared<EMnistOp>(name, usage, num_workers, dir, op_connect_size, std::move(schema), std::move(sampler));
+
+  RETURN_IF_NOT_OK(op->WalkAllFiles());
+
+  for (size_t i = 0; i < op->image_names_.size(); ++i) {
+    std::ifstream image_reader;
+    image_reader.open(op->image_names_[i], std::ios::binary);
+    CHECK_FAIL_RETURN_UNEXPECTED(image_reader.is_open(),
+                                 "Invalid file, failed to open image file: " + op->image_names_[i]);
+    std::ifstream label_reader;
+    label_reader.open(op->label_names_[i], std::ios::binary);
+    CHECK_FAIL_RETURN_UNEXPECTED(label_reader.is_open(),
+                                 "Invalid file, failed to open label file: " + op->label_names_[i]);
+    uint32_t num_images;
+    Status s = op->CheckImage(op->image_names_[i], &image_reader, &num_images);
+    image_reader.close();
+    RETURN_IF_NOT_OK(s);
+
+    uint32_t num_labels;
+    s = op->CheckLabel(op->label_names_[i], &label_reader, &num_labels);
+    label_reader.close();
+    RETURN_IF_NOT_OK(s);
+
+    CHECK_FAIL_RETURN_UNEXPECTED((num_images == num_labels),
+                                 "Invalid data, num of images is not equal to num of labels.");
+    *count = *count + num_images;
+  }
+
+  return Status::OK();
+}
+
+}  // namespace dataset
+}  // namespace mindspore

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/emnist_op.h

@@ -0,0 +1,84 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://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.
+ */
+#ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_EMNIST_OP_H_
+#define MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_EMNIST_OP_H_
+
+#include <algorithm>
+#include <map>
+#include <memory>
+#include <string>
+#include <utility>
+#include <vector>
+
+#include "minddata/dataset/engine/datasetops/source/mnist_op.h"
+
+namespace mindspore {
+namespace dataset {
+// Forward declares
+template <typename T>
+class Queue;
+
+class EMnistOp : public MnistOp {
+ public:
+  // Constructor.
+  // @param const std::string &name - Class of this dataset, can be
+  //     "byclass","bymerge","balanced","letters","digits","mnist".
+  // @param const std::string &usage - Usage of this dataset, can be 'train', 'test' or 'all'.
+  // @param int32_t num_workers - Number of workers reading images in parallel.
+  // @param const std::string &folder_path - Dir directory of emnist.
+  // @param int32_t queue_size - Connector queue size.
+  // @param std::unique_ptr<DataSchema> data_schema - The schema of the Emnist dataset.
+  // @param std::shared_ptr<SamplerRT> sampler - Sampler tells EMnistOp what to read.
+  EMnistOp(const std::string &name, const std::string &usage, int32_t num_workers, const std::string &folder_path,
+           int32_t queue_size, std::unique_ptr<DataSchema> data_schema, std::shared_ptr<SamplerRT> sampler);
+
+  // Destructor.
+  ~EMnistOp() = default;
+
+  // A print method typically used for debugging.
+  // @param std::ostream &out - Out stream.
+  // @param bool show_all - Whether to show all information.
+  void Print(std::ostream &out, bool show_all) const override;
+
+  // Function to count the number of samples in the EMNIST dataset.
+  // @param const std::string &dir - Path to the EMNIST directory.
+  // @param const std::string &name - Class of this dataset, can be
+  //     "byclass","bymerge","balanced","letters","digits","mnist".
+  // @param const std::string &usage - Usage of this dataset, can be 'train', 'test' or 'all'.
+  // @param int64_t *count - Output arg that will hold the minimum of the actual dataset size and numSamples.
+  // @return Status The status code returned.
+  static Status CountTotalRows(const std::string &dir, const std::string &name, const std::string &usage,
+                               int64_t *count);
+
+  // Op name getter.
+  // @return Name of the current Op.
+  std::string Name() const override { return "EMnistOp"; }
+
+  // DatasetName name getter.
+  // \return DatasetName of the current Op.
+  std::string DatasetName(bool upper = false) const override { return upper ? "EMnist" : "emnist"; }
+
+ private:
+  // Read all files in the directory.
+  // @return Status The status code returned.
+  Status WalkAllFiles() override;
+
+  const std::string name_;  // can be "byclass", "bymerge", "balanced", "letters", "digits", "mnist".
+};
+
+}  // namespace dataset
+}  // namespace mindspore
+#endif  // MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_DATASETOPS_SOURCE_EMNIST_OP_H_

mindspore/ccsrc/minddata/dataset/engine/ir/datasetops/dataset_node.h

@@ -83,6 +83,7 @@ constexpr char kCLUENode[] = "CLUEDataset";
 constexpr char kCocoNode[] = "CocoDataset";
 constexpr char kCSVNode[] = "CSVDataset";
 constexpr char kDIV2KNode[] = "DIV2KDataset";
+constexpr char kEMnistNode[] = "EMnistDataset";
 constexpr char kFlickrNode[] = "FlickrDataset";
 constexpr char kGeneratorNode[] = "GeneratorDataset";
 constexpr char kImageFolderNode[] = "ImageFolderDataset";

mindspore/ccsrc/minddata/dataset/engine/ir/datasetops/source/CMakeLists.txt

@@ -12,6 +12,7 @@ set(DATASET_ENGINE_IR_DATASETOPS_SOURCE_SRC_FILES
         coco_node.cc
         csv_node.cc
         div2k_node.cc
+        emnist_node.cc
         flickr_node.cc
         image_folder_node.cc
         manifest_node.cc
@@ -33,4 +34,4 @@ if(ENABLE_PYTHON)
             )
 endif()
 
-add_library(engine-ir-datasetops-source OBJECT ${DATASET_ENGINE_IR_DATASETOPS_SOURCE_SRC_FILES})
+add_library(engine-ir-datasetops-source OBJECT ${DATASET_ENGINE_IR_DATASETOPS_SOURCE_SRC_FILES})

mindspore/ccsrc/minddata/dataset/engine/ir/datasetops/source/emnist_node.cc

@@ -0,0 +1,121 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://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.
+ */
+
+#include "minddata/dataset/engine/ir/datasetops/source/emnist_node.h"
+
+#include <memory>
+#include <string>
+#include <utility>
+#include <vector>
+
+#include "minddata/dataset/engine/datasetops/source/emnist_op.h"
+#include "minddata/dataset/util/status.h"
+
+namespace mindspore {
+namespace dataset {
+EMnistNode::EMnistNode(const std::string &dataset_dir, const std::string &name, const std::string &usage,
+                       std::shared_ptr<SamplerObj> sampler, std::shared_ptr<DatasetCache> cache)
+    : MappableSourceNode(std::move(cache)), dataset_dir_(dataset_dir), name_(name), usage_(usage), sampler_(sampler) {}
+
+std::shared_ptr<DatasetNode> EMnistNode::Copy() {
+  std::shared_ptr<SamplerObj> sampler = (sampler_ == nullptr) ? nullptr : sampler_->SamplerCopy();
+  auto node = std::make_shared<EMnistNode>(dataset_dir_, name_, usage_, sampler, cache_);
+  return node;
+}
+
+void EMnistNode::Print(std::ostream &out) const {
+  out << (Name() + "(cache: " + ((cache_ != nullptr) ? "true" : "false") + ")");
+}
+
+Status EMnistNode::ValidateParams() {
+  RETURN_IF_NOT_OK(DatasetNode::ValidateParams());
+  RETURN_IF_NOT_OK(ValidateDatasetDirParam("EMnistNode", dataset_dir_));
+
+  RETURN_IF_NOT_OK(ValidateDatasetSampler("EMnistNode", sampler_));
+
+  RETURN_IF_NOT_OK(ValidateStringValue("EMnistNode", usage_, {"train", "test", "all"}));
+
+  RETURN_IF_NOT_OK(
+    ValidateStringValue("EMnistNode", name_, {"byclass", "bymerge", "balanced", "letters", "digits", "mnist"}));
+
+  return Status::OK();
+}
+
+Status EMnistNode::Build(std::vector<std::shared_ptr<DatasetOp>> *const node_ops) {
+  // Do internal Schema generation.
+  auto schema = std::make_unique<DataSchema>();
+  RETURN_IF_NOT_OK(schema->AddColumn(ColDescriptor("image", DataType(DataType::DE_UINT8), TensorImpl::kCv, 1)));
+  TensorShape scalar = TensorShape::CreateScalar();
+  RETURN_IF_NOT_OK(
+    schema->AddColumn(ColDescriptor("label", DataType(DataType::DE_UINT32), TensorImpl::kFlexible, 0, &scalar)));
+  std::shared_ptr<SamplerRT> sampler_rt = nullptr;
+  RETURN_IF_NOT_OK(sampler_->SamplerBuild(&sampler_rt));
+
+  auto op = std::make_shared<EMnistOp>(name_, usage_, num_workers_, dataset_dir_, connector_que_size_,
+                                       std::move(schema), std::move(sampler_rt));
+  op->SetTotalRepeats(GetTotalRepeats());
+  op->SetNumRepeatsPerEpoch(GetNumRepeatsPerEpoch());
+  node_ops->push_back(op);
+
+  return Status::OK();
+}
+
+// Get the shard id of node.
+Status EMnistNode::GetShardId(int32_t *shard_id) {
+  *shard_id = sampler_->ShardId();
+
+  return Status::OK();
+}
+
+// Get Dataset size.
+Status EMnistNode::GetDatasetSize(const std::shared_ptr<DatasetSizeGetter> &size_getter, bool estimate,
+                                  int64_t *dataset_size) {
+  if (dataset_size_ > 0) {
+    *dataset_size = dataset_size_;
+    return Status::OK();
+  }
+  int64_t num_rows, sample_size;
+  RETURN_IF_NOT_OK(EMnistOp::CountTotalRows(dataset_dir_, name_, usage_, &num_rows));
+  std::shared_ptr<SamplerRT> sampler_rt = nullptr;
+  RETURN_IF_NOT_OK(sampler_->SamplerBuild(&sampler_rt));
+  sample_size = sampler_rt->CalculateNumSamples(num_rows);
+  if (sample_size == -1) {
+    RETURN_IF_NOT_OK(size_getter->DryRun(shared_from_this(), &sample_size));
+  }
+  *dataset_size = sample_size;
+  dataset_size_ = *dataset_size;
+  return Status::OK();
+}
+
+Status EMnistNode::to_json(nlohmann::json *out_json) {
+  nlohmann::json args, sampler_args;
+  RETURN_IF_NOT_OK(sampler_->to_json(&sampler_args));
+  args["sampler"] = sampler_args;
+  args["num_parallel_workers"] = num_workers_;
+  args["dataset_dir"] = dataset_dir_;
+  args["name"] = name_;
+  args["usage"] = usage_;
+  if (cache_ != nullptr) {
+    nlohmann::json cache_args;
+    RETURN_IF_NOT_OK(cache_->to_json(&cache_args));
+    args["cache"] = cache_args;
+  }
+  *out_json = args;
+  return Status::OK();
+}
+
+}  // namespace dataset
+}  // namespace mindspore

mindspore/ccsrc/minddata/dataset/engine/ir/datasetops/source/emnist_node.h

@@ -0,0 +1,111 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://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.
+ */
+
+#ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_EMNIST_NODE_H_
+#define MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_EMNIST_NODE_H_
+
+#include <memory>
+#include <string>
+#include <vector>
+
+#include "minddata/dataset/engine/ir/datasetops/dataset_node.h"
+
+namespace mindspore {
+namespace dataset {
+class EMnistNode : public MappableSourceNode {
+ public:
+  /// \brief Constructor.
+  /// \param[in] dataset_dir Dataset directory of emnist.
+  /// \param[in] name Class of this dataset, can be "byclass", "bymerge", "balanced", "letters", "digits", "mnist".
+  /// \param[in] usage Usage of this dataset, can be 'train', 'test' or 'all'.
+  /// \param[in] sampler Tells EMnistOp what to read.
+  /// \param[in] cache Tensor cache to use.
+  EMnistNode(const std::string &dataset_dir, const std::string &name, const std::string &usage,
+             std::shared_ptr<SamplerObj> sampler, std::shared_ptr<DatasetCache> cache);
+
+  /// \brief Destructor.
+  ~EMnistNode() = default;
+
+  /// \brief Node name getter.
+  /// \return Name of the current node.
+  std::string Name() const override { return "EMnistNode"; }
+
+  /// \brief Print the description.
+  /// \param[in] out The output stream to write output to.
+  void Print(std::ostream &out) const override;
+
+  /// \brief Copy the node to a new object.
+  /// \return A shared pointer to the new copy.
+  std::shared_ptr<DatasetNode> Copy() override;
+
+  /// \brief A base class override function to create the required runtime dataset op objects for this class.
+  /// \param[in] node_ops A vector containing shared pointer to the Dataset Ops that this object will create.
+  /// \return Status Status::OK() if build successfully.
+  Status Build(std::vector<std::shared_ptr<DatasetOp>> *const node_ops) override;
+
+  /// \brief Parameters validation.
+  /// \return Status Status::OK() if all the parameters are valid.
+  Status ValidateParams() override;
+
+  /// \brief Get the shard id of node.
+  /// \param[in] shard_id The shard id.
+  /// \return Status Status::OK() if get shard id successfully.
+  Status GetShardId(int32_t *shard_id) override;
+
+  /// \brief Base-class override for GetDatasetSize.
+  /// \param[in] size_getter Shared pointer to DatasetSizeGetter.
+  /// \param[in] estimate This is only supported by some of the ops and it's used to speed up the process of getting
+  ///     dataset size at the expense of accuracy.
+  /// \param[out] dataset_size The size of the dataset.
+  /// \return Status of the function.
+  Status GetDatasetSize(const std::shared_ptr<DatasetSizeGetter> &size_getter, bool estimate,
+                        int64_t *dataset_size) override;
+
+  /// \brief Getter functions.
+  /// \return Dataset direction.
+  const std::string &DatasetDir() const { return dataset_dir_; }
+
+  /// \brief Getter functions.
+  /// \return Usage.
+  const std::string &Usage() const { return usage_; }
+
+  /// \brief Getter functions.
+  /// \return Name.
+  const std::string &GetName() const { return name_; }
+
+  /// \brief Get the arguments of node.
+  /// \param[out] out_json JSON string of all attributes.
+  /// \return Status of the function.
+  Status to_json(nlohmann::json *out_json) override;
+
+  /// \brief Sampler getter.
+  /// \return SamplerObj of the current node.
+  std::shared_ptr<SamplerObj> Sampler() override { return sampler_; }
+
+  /// \brief Sampler setter.
+  /// \param[in] sampler Tells EMnistOp what to read.
+  void SetSampler(std::shared_ptr<SamplerObj> sampler) override { sampler_ = sampler; }
+
+ private:
+  std::string dataset_dir_;
+  std::string name_;
+  std::string usage_;
+  std::shared_ptr<SamplerObj> sampler_;
+};
+
+}  // namespace dataset
+}  // namespace mindspore
+#endif  // MINDSPORE_CCSRC_MINDDATA_DATASET_ENGINE_IR_DATASETOPS_SOURCE_EMNIST_NODE_H_

mindspore/ccsrc/minddata/dataset/include/dataset/datasets.h

@@ -1628,6 +1628,93 @@ inline std::shared_ptr<DIV2KDataset> DIV2K(const std::string &dataset_dir, const
                                         decode, sampler, cache);
 }
 
+/// \class EMnistDataset
+/// \brief A source dataset for reading and parsing EMnist dataset.
+class EMnistDataset : public Dataset {
+ public:
+  /// \brief Constructor of EMnistDataset.
+  /// \param[in] dataset_dir Path to the root directory that contains the dataset.
+  /// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits"
+  ///     or "mnist".
+  /// \param[in] usage Part of dataset of EMNIST, can be "train", "test" or "all".
+  /// \param[in] sampler Shared pointer to a sampler object used to choose samples from the dataset. If sampler is not
+  ///     given, a `RandomSampler` will be used to randomly iterate the entire dataset.
+  /// \param[in] cache Tensor cache to use.
+  explicit EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                         const std::vector<char> &usage, const std::shared_ptr<Sampler> &sampler,
+                         const std::shared_ptr<DatasetCache> &cache);
+
+  /// \brief Constructor of EMnistDataset.
+  /// \param[in] dataset_dir Path to the root directory that contains the dataset.
+  /// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits"
+  ///     or "mnist".
+  /// \param[in] usage Part of dataset of EMNIST, can be "train", "test" or "all".
+  /// \param[in] sampler Raw pointer to a sampler object used to choose samples from the dataset.
+  /// \param[in] cache Tensor cache to use.
+  explicit EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                         const std::vector<char> &usage, const Sampler *sampler,
+                         const std::shared_ptr<DatasetCache> &cache);
+
+  /// \brief Constructor of EMnistDataset.
+  /// \param[in] dataset_dir Path to the root directory that contains the dataset.
+  /// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits"
+  ///     or "mnist".
+  /// \param[in] usage Part of dataset of EMNIST, can be "train", "test" or "all".
+  /// \param[in] sampler Sampler object used to choose samples from the dataset.
+  /// \param[in] cache Tensor cache to use.
+  explicit EMnistDataset(const std::vector<char> &dataset_dir, const std::vector<char> &name,
+                         const std::vector<char> &usage, const std::reference_wrapper<Sampler> sampler,
+                         const std::shared_ptr<DatasetCache> &cache);
+  ~EMnistDataset() = default;
+};
+
+/// \brief Function to create a EMnistDataset.
+/// \notes The generated dataset has two columns ["image", "label"].
+/// \param[in] dataset_dir Path to the root directory that contains the dataset.
+/// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits" or "mnist".
+/// \param[in] usage Usage of EMNIST, can be "train", "test" or "all" (default = "all").
+/// \param[in] sampler Shared pointer to a sampler object used to choose samples from the dataset. If sampler is not.
+///     given, a `RandomSampler` will be used to randomly iterate the entire dataset (default = RandomSampler()).
+/// \param[in] cache Tensor cache to use. (default=nullptr which means no cache is used).
+/// \return Shared pointer to the current EMnistDataset.
+inline std::shared_ptr<EMnistDataset> EMnist(
+  const std::string &dataset_dir, const std::string &name, const std::string &usage = "all",
+  const std::shared_ptr<Sampler> &sampler = std::make_shared<RandomSampler>(),
+  const std::shared_ptr<DatasetCache> &cache = nullptr) {
+  return std::make_shared<EMnistDataset>(StringToChar(dataset_dir), StringToChar(name), StringToChar(usage), sampler,
+                                         cache);
+}
+
+/// \brief Function to create a EMnistDataset.
+/// \notes The generated dataset has two columns ["image", "label"].
+/// \param[in] dataset_dir Path to the root directory that contains the dataset
+/// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits" or "mnist".
+/// \param[in] usage Usage of EMNIST, can be "train", "test" or "all".
+/// \param[in] sampler Raw pointer to a sampler object used to choose samples from the dataset.
+/// \param[in] cache Tensor cache to use. (default=nullptr which means no cache is used).
+/// \return Shared pointer to the current EMnistDataset.
+inline std::shared_ptr<EMnistDataset> EMnist(const std::string &dataset_dir, const std::string &usage,
+                                             const std::string &name, const Sampler *sampler,
+                                             const std::shared_ptr<DatasetCache> &cache = nullptr) {
+  return std::make_shared<EMnistDataset>(StringToChar(dataset_dir), StringToChar(name), StringToChar(usage), sampler,
+                                         cache);
+}
+
+/// \brief Function to create a EMnistDataset.
+/// \notes The generated dataset has two columns ["image", "label"].
+/// \param[in] dataset_dir Path to the root directory that contains the dataset.
+/// \param[in] name Name of splits for EMNIST, can be "byclass", "bymerge", "balanced", "letters", "digits" or "mnist".
+/// \param[in] usage Usage of EMNIST, can be "train", "test" or "all".
+/// \param[in] sampler Sampler object used to choose samples from the dataset.
+/// \param[in] cache Tensor cache to use. (default=nullptr which means no cache is used).
+/// \return Shared pointer to the current EMnistDataset.
+inline std::shared_ptr<EMnistDataset> EMnist(const std::string &dataset_dir, const std::string &name,
+                                             const std::string &usage, const std::reference_wrapper<Sampler> sampler,
+                                             const std::shared_ptr<DatasetCache> &cache = nullptr) {
+  return std::make_shared<EMnistDataset>(StringToChar(dataset_dir), StringToChar(name), StringToChar(usage), sampler,
+                                         cache);
+}
+
 /// \class FlickrDataset
 /// \brief A source dataset for reading and parsing Flickr dataset.
 class FlickrDataset : public Dataset {

mindspore/ccsrc/minddata/dataset/include/dataset/samplers.h

@@ -39,6 +39,7 @@ class Sampler : std::enable_shared_from_this<Sampler> {
   friend class CocoDataset;
   friend class CSVDataset;
   friend class DIV2KDataset;
+  friend class EMnistDataset;
   friend class FlickrDataset;
   friend class ImageFolderDataset;
   friend class ManifestDataset;

mindspore/dataset/engine/datasets.py

@@ -66,7 +66,7 @@ from .validators import check_batch, check_shuffle, check_map, check_filter, che
     check_bucket_batch_by_length, check_cluedataset, check_save, check_csvdataset, check_paddeddataset, \
     check_tuple_iterator, check_dict_iterator, check_schema, check_to_device_send, check_flickr_dataset, \
     check_sb_dataset, check_flowers102dataset, check_cityscapes_dataset, check_usps_dataset, check_div2k_dataset, \
-    check_sbu_dataset, check_qmnist_dataset
+    check_sbu_dataset, check_qmnist_dataset, check_emnist_dataset
 from ..core.config import get_callback_timeout, _init_device_info, get_enable_shared_mem, get_num_parallel_workers, \
     get_prefetch_size
 from ..core.datatypes import mstype_to_detype, mstypelist_to_detypelist
@@ -6350,6 +6350,138 @@ class PaddedDataset(GeneratorDataset):
         self.padded_samples = padded_samples
 
 
+class EMnistDataset(MappableDataset):
+    """
+    A source dataset for reading and parsing the EMNIST dataset.
+
+    The generated dataset has two columns :py:obj:`[image, label]`.
+    The tensor of column :py:obj:`image` is of the uint8 type.
+    The tensor of column :py:obj:`label` is a scalar of the uint32 type.
+
+    Args:
+        dataset_dir (str): Path to the root directory that contains the dataset.
+        name (str): Name of splits for this dataset, can be "byclass", "bymerge", "balanced", "letters", "digits"
+            or "mnist".
+        usage (str, optional): Usage of this dataset, can be "train", "test" or "all".
+            (default=None, will read all samples).
+        num_samples (int, optional): The number of images to be included in the dataset
+            (default=None, will read all images).
+        num_parallel_workers (int, optional): Number of workers to read the data
+            (default=None, will use value set in the config).
+        shuffle (bool, optional): Whether or not to perform shuffle on the dataset
+            (default=None, expected order behavior shown in the table).
+        sampler (Sampler, optional): Object used to choose samples from the
+            dataset (default=None, expected order behavior shown in the table).
+        num_shards (int, optional): Number of shards that the dataset will be divided into (default=None).
+            When this argument is specified, `num_samples` reflects the max sample number of per shard.
+        shard_id (int, optional): The shard ID within `num_shards` (default=None). This
+            argument can only be specified when `num_shards` is also specified.
+        cache (DatasetCache, optional): Use tensor caching service to speed up dataset processing.
+            (default=None, which means no cache is used).
+
+    Raises:
+        RuntimeError: If sampler and shuffle are specified at the same time.
+        RuntimeError: If sampler and sharding are specified at the same time.
+        RuntimeError: If num_shards is specified but shard_id is None.
+        RuntimeError: If shard_id is specified but num_shards is None.
+        ValueError: If shard_id is invalid (< 0 or >= num_shards).
+
+    Note:
+        - This dataset can take in a `sampler`. `sampler` and `shuffle` are mutually exclusive.
+          The table below shows what input arguments are allowed and their expected behavior.
+
+    .. list-table:: Expected Order Behavior of Using `sampler` and `shuffle`
+       :widths: 25 25 50
+       :header-rows: 1
+
+       * - Parameter `sampler`
+         - Parameter `shuffle`
+         - Expected Order Behavior
+       * - None
+         - None
+         - random order
+       * - None
+         - True
+         - random order
+       * - None
+         - False
+         - sequential order
+       * - Sampler object
+         - None
+         - order defined by sampler
+       * - Sampler object
+         - True
+         - not allowed
+       * - Sampler object
+         - False
+         - not allowed
+
+    Examples:
+        >>> emnist_dataset_dir = "/path/to/emnist_dataset_directory"
+        >>>
+        >>> # Read 3 samples from EMNIST dataset
+        >>> dataset = ds.EMnistDataset(dataset_dir=emnist_dataset_dir, name="mnist", num_samples=3)
+        >>>
+        >>> # Note: In emnist_dataset dataset, each dictionary has keys "image" and "label"
+
+    About EMNIST dataset:
+
+    The EMNIST dataset is a set of handwritten character digits derived from the NIST Special
+    Database 19 and converted to a 28x28 pixel image format and dataset structure that directly
+    matches the MNIST dataset. Further information on the dataset contents and conversion process
+    can be found in the paper available at https://arxiv.org/abs/1702.05373v1.
+
+    The numbers of characters and classes of each split of EMNIST are as follows:
+
+    By Class: 814,255 characters and 62 unbalanced classes.
+    By Merge: 814,255 characters and 47 unbalanced classes.
+    Balanced: 131,600 characters and 47 balanced classes.
+    Letters: 145,600 characters and 26 balanced classes.
+    Digits: 280,000 characters and 10 balanced classes.
+    MNIST: 70,000 characters and 10 balanced classes.
+
+    Here is the original EMNIST dataset structure.
+    You can unzip the dataset files into this directory structure and read by MindSpore's API.
+
+    .. code-block::
+
+        .
+        └── mnist_dataset_dir
+             ├── emnist-mnist-train-images-idx3-ubyte
+             ├── emnist-mnist-train-labels-idx1-ubyte
+             ├── emnist-mnist-test-images-idx3-ubyte
+             ├── emnist-mnist-test-labels-idx1-ubyte
+             ├── ...
+
+    Citation:
+
+    .. code-block::
+
+        @article{cohen_afshar_tapson_schaik_2017,
+        title        = {EMNIST: Extending MNIST to handwritten letters},
+        DOI          = {10.1109/ijcnn.2017.7966217},
+        journal      = {2017 International Joint Conference on Neural Networks (IJCNN)},
+        author       = {Cohen, Gregory and Afshar, Saeed and Tapson, Jonathan and Schaik, Andre Van},
+        year         = {2017},
+        howpublished = {https://www.westernsydney.edu.au/icns/reproducible_research/
+                        publication_support_materials/emnist}
+        }
+    """
+
+    @check_emnist_dataset
+    def __init__(self, dataset_dir, name, usage=None, num_samples=None, num_parallel_workers=None,
+                 shuffle=None, sampler=None, num_shards=None, shard_id=None, cache=None):
+        super().__init__(num_parallel_workers=num_parallel_workers, sampler=sampler, num_samples=num_samples,
+                         shuffle=shuffle, num_shards=num_shards, shard_id=shard_id, cache=cache)
+
+        self.dataset_dir = dataset_dir
+        self.name = name
+        self.usage = replace_none(usage, "all")
+
+    def parse(self, children=None):
+        return cde.EMnistNode(self.dataset_dir, self.name, self.usage, self.sampler)
+
+
 class FlickrDataset(MappableDataset):
     """
     A source dataset for reading and parsing Flickr8k and Flickr30k dataset.

mindspore/dataset/engine/validators.py

@@ -1,4 +1,4 @@
-# Copyright 2019 Huawei Technologies Co., Ltd
+# Copyright 2019-2021 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -1463,6 +1463,39 @@ def check_to_device_send(method):
     return new_method
 
 
+def check_emnist_dataset(method):
+    """A wrapper that wraps a parameter checker emnist dataset"""
+
+    @wraps(method)
+    def new_method(self, *args, **kwargs):
+        _, param_dict = parse_user_args(method, *args, **kwargs)
+
+        nreq_param_int = ['num_samples', 'num_parallel_workers', 'num_shards', 'shard_id']
+        nreq_param_bool = ['shuffle']
+
+        validate_dataset_param_value(nreq_param_int, param_dict, int)
+        validate_dataset_param_value(nreq_param_bool, param_dict, bool)
+
+        dataset_dir = param_dict.get('dataset_dir')
+        check_dir(dataset_dir)
+
+        name = param_dict.get('name')
+        check_valid_str(name, ["byclass", "bymerge", "balanced", "letters", "digits", "mnist"], "name")
+
+        usage = param_dict.get('usage')
+        if usage is not None:
+            check_valid_str(usage, ["train", "test", "all"], "usage")
+
+        check_sampler_shuffle_shard_options(param_dict)
+
+        cache = param_dict.get('cache')
+        check_cache_option(cache)
+
+        return method(self, *args, **kwargs)
+
+    return new_method
+
+
 def check_flickr_dataset(method):
     """A wrapper that wraps a parameter checker around the original Dataset(Flickr8k, Flickr30k)."""
 

tests/ut/cpp/dataset/CMakeLists.txt

@@ -23,6 +23,7 @@ SET(DE_UT_SRCS
         c_api_dataset_config_test.cc
         c_api_dataset_csv_test.cc
         c_api_dataset_div2k_test.cc
+        c_api_dataset_emnist_test.cc
         c_api_dataset_flickr_test.cc
         c_api_dataset_iterator_test.cc
         c_api_dataset_manifest_test.cc

tests/ut/cpp/dataset/c_api_dataset_emnist_test.cc

@@ -0,0 +1,368 @@
+/**
+ * Copyright 2021 Huawei Technologies Co., Ltd
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://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.
+ */
+#include "common/common.h"
+
+#include "minddata/dataset/include/dataset/datasets.h"
+
+using namespace mindspore::dataset;
+using mindspore::dataset::DataType;
+using mindspore::dataset::Tensor;
+using mindspore::dataset::TensorShape;
+
+class MindDataTestPipeline : public UT::DatasetOpTesting {
+ protected:
+};
+
+TEST_F(MindDataTestPipeline, TestEMnistTrainDataset) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTrainDataset.";
+
+  // Create a EMnist Train Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "train", std::make_shared<RandomSampler>(false, 5));
+
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  // This will trigger the creation of the Execution Tree and launch it.
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  EXPECT_NE(iter, nullptr);
+
+  // Iterate the dataset and get each row
+  std::unordered_map<std::string, mindspore::MSTensor> row;
+  ASSERT_OK(iter->GetNextRow(&row));
+
+  EXPECT_NE(row.find("image"), row.end());
+  EXPECT_NE(row.find("label"), row.end());
+
+  uint64_t i = 0;
+  while (row.size() != 0) {
+    i++;
+    auto image = row["image"];
+    MS_LOG(INFO) << "Tensor image shape: " << image.Shape();
+    ASSERT_OK(iter->GetNextRow(&row));
+  }
+
+  EXPECT_EQ(i, 5);
+  // Manually terminate the pipeline
+  iter->Stop();
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistTestDataset) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTestDataset.";
+
+  // Create a EMNIST Test Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "train", std::make_shared<RandomSampler>(false, 5));
+
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  // This will trigger the creation of the Execution Tree and launch it.
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  EXPECT_NE(iter, nullptr);
+
+  // Iterate the dataset and get each row
+  std::unordered_map<std::string, mindspore::MSTensor> row;
+  ASSERT_OK(iter->GetNextRow(&row));
+
+  EXPECT_NE(row.find("image"), row.end());
+  EXPECT_NE(row.find("label"), row.end());
+
+  uint64_t i = 0;
+  while (row.size() != 0) {
+    i++;
+    auto image = row["image"];
+    MS_LOG(INFO) << "Tensor image shape: " << image.Shape();
+    ASSERT_OK(iter->GetNextRow(&row));
+  }
+
+  EXPECT_EQ(i, 5);
+
+  // Manually terminate the pipeline
+  iter->Stop();
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistTrainDatasetWithPipeline) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTrainDatasetWithPipeline.";
+
+  // Create two Emnist Train Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+
+  std::shared_ptr<Dataset> ds1 = EMnist(folder_path, "mnist", "train", std::make_shared<RandomSampler>(false, 5));
+  std::shared_ptr<Dataset> ds2 = EMnist(folder_path, "byclass", "train", std::make_shared<RandomSampler>(false, 5));
+  EXPECT_NE(ds1, nullptr);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create two Repeat operation on ds
+  int32_t repeat_num = 1;
+  ds1 = ds1->Repeat(repeat_num);
+  EXPECT_NE(ds1, nullptr);
+  repeat_num = 1;
+  ds2 = ds2->Repeat(repeat_num);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create two Project operation on ds
+  std::vector<std::string> column_project = {"image", "label"};
+  ds1 = ds1->Project(column_project);
+  EXPECT_NE(ds1, nullptr);
+  ds2 = ds2->Project(column_project);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create a Concat operation on the ds
+  ds1 = ds1->Concat({ds2});
+  EXPECT_NE(ds1, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  // This will trigger the creation of the Execution Tree and launch it.
+  std::shared_ptr<Iterator> iter = ds1->CreateIterator();
+  EXPECT_NE(iter, nullptr);
+
+  // Iterate the dataset and get each row
+  std::unordered_map<std::string, mindspore::MSTensor> row;
+  ASSERT_OK(iter->GetNextRow(&row));
+
+  EXPECT_NE(row.find("image"), row.end());
+  EXPECT_NE(row.find("label"), row.end());
+
+  uint64_t i = 0;
+  while (row.size() != 0) {
+    i++;
+    auto image = row["image"];
+    MS_LOG(INFO) << "Tensor image shape: " << image.Shape();
+    ASSERT_OK(iter->GetNextRow(&row));
+  }
+
+  EXPECT_EQ(i, 10);
+
+  // Manually terminate the pipeline
+  iter->Stop();
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistTestDatasetWithPipeline) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTestDatasetWithPipeline.";
+
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+
+  // Create two EMnist Test Dataset
+  std::shared_ptr<Dataset> ds1 = EMnist(folder_path, "mnist", "test", std::make_shared<RandomSampler>(false, 5));
+  std::shared_ptr<Dataset> ds2 = EMnist(folder_path, "mnist", "test", std::make_shared<RandomSampler>(false, 5));
+  EXPECT_NE(ds1, nullptr);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create two Repeat operation on ds
+  int32_t repeat_num = 1;
+  ds1 = ds1->Repeat(repeat_num);
+  EXPECT_NE(ds1, nullptr);
+  repeat_num = 1;
+  ds2 = ds2->Repeat(repeat_num);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create two Project operation on ds
+  std::vector<std::string> column_project = {"image", "label"};
+  ds1 = ds1->Project(column_project);
+  EXPECT_NE(ds1, nullptr);
+  ds2 = ds2->Project(column_project);
+  EXPECT_NE(ds2, nullptr);
+
+  // Create a Concat operation on the ds
+  ds1 = ds1->Concat({ds2});
+  EXPECT_NE(ds1, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  // This will trigger the creation of the Execution Tree and launch it.
+  std::shared_ptr<Iterator> iter = ds1->CreateIterator();
+  EXPECT_NE(iter, nullptr);
+
+  // Iterate the dataset and get each row
+  std::unordered_map<std::string, mindspore::MSTensor> row;
+  ASSERT_OK(iter->GetNextRow(&row));
+
+  EXPECT_NE(row.find("image"), row.end());
+  EXPECT_NE(row.find("label"), row.end());
+
+  uint64_t i = 0;
+  while (row.size() != 0) {
+    i++;
+    auto image = row["image"];
+    MS_LOG(INFO) << "Tensor image shape: " << image.Shape();
+    ASSERT_OK(iter->GetNextRow(&row));
+  }
+
+  EXPECT_EQ(i, 10);
+
+  // Manually terminate the pipeline
+  iter->Stop();
+}
+
+TEST_F(MindDataTestPipeline, TestGetEMnistTrainDatasetSize) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestGetEMnistTrainDatasetSize.";
+
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  // Create a EMnist Train Dataset
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "train");
+  EXPECT_NE(ds, nullptr);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+
+  std::shared_ptr<Dataset> ds2 = EMnist(folder_path, "byclass", "train");
+  EXPECT_NE(ds2, nullptr);
+
+  EXPECT_EQ(ds2->GetDatasetSize(), 10);
+}
+
+TEST_F(MindDataTestPipeline, TestGetEMnistTestDatasetSize) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestGetEMnistTestDatasetSize.";
+
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+
+  // Create a EMnist Test Dataset
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "test");
+  EXPECT_NE(ds, nullptr);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistTrainDatasetGetters) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTrainDatasetGetters.";
+
+  // Create a EMnist Train Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "train");
+  EXPECT_NE(ds, nullptr);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  std::vector<DataType> types = ToDETypes(ds->GetOutputTypes());
+  std::vector<TensorShape> shapes = ToTensorShapeVec(ds->GetOutputShapes());
+  std::vector<std::string> column_names = {"image", "label"};
+  int64_t num_classes = ds->GetNumClasses();
+  EXPECT_EQ(types.size(), 2);
+  EXPECT_EQ(types[0].ToString(), "uint8");
+  EXPECT_EQ(types[1].ToString(), "uint32");
+  EXPECT_EQ(shapes.size(), 2);
+  EXPECT_EQ(shapes[0].ToString(), "<28,28,1>");
+  EXPECT_EQ(shapes[1].ToString(), "<>");
+  EXPECT_EQ(num_classes, -1);
+  EXPECT_EQ(ds->GetBatchSize(), 1);
+  EXPECT_EQ(ds->GetRepeatCount(), 1);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  EXPECT_EQ(ToDETypes(ds->GetOutputTypes()), types);
+  EXPECT_EQ(ToTensorShapeVec(ds->GetOutputShapes()), shapes);
+  EXPECT_EQ(ds->GetNumClasses(), -1);
+
+  EXPECT_EQ(ds->GetColumnNames(), column_names);
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  EXPECT_EQ(ToDETypes(ds->GetOutputTypes()), types);
+  EXPECT_EQ(ToTensorShapeVec(ds->GetOutputShapes()), shapes);
+  EXPECT_EQ(ds->GetBatchSize(), 1);
+  EXPECT_EQ(ds->GetRepeatCount(), 1);
+  EXPECT_EQ(ds->GetNumClasses(), -1);
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistTestDatasetGetters) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistTestDatasetGetters.";
+
+  // Create a EMnist Test Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "test");
+  EXPECT_NE(ds, nullptr);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  std::vector<DataType> types = ToDETypes(ds->GetOutputTypes());
+  std::vector<TensorShape> shapes = ToTensorShapeVec(ds->GetOutputShapes());
+  std::vector<std::string> column_names = {"image", "label"};
+  int64_t num_classes = ds->GetNumClasses();
+  EXPECT_EQ(types.size(), 2);
+  EXPECT_EQ(types[0].ToString(), "uint8");
+  EXPECT_EQ(types[1].ToString(), "uint32");
+  EXPECT_EQ(shapes.size(), 2);
+  EXPECT_EQ(shapes[0].ToString(), "<28,28,1>");
+  EXPECT_EQ(shapes[1].ToString(), "<>");
+  EXPECT_EQ(num_classes, -1);
+  EXPECT_EQ(ds->GetBatchSize(), 1);
+  EXPECT_EQ(ds->GetRepeatCount(), 1);
+
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  EXPECT_EQ(ToDETypes(ds->GetOutputTypes()), types);
+  EXPECT_EQ(ToTensorShapeVec(ds->GetOutputShapes()), shapes);
+  EXPECT_EQ(ds->GetNumClasses(), -1);
+
+  EXPECT_EQ(ds->GetColumnNames(), column_names);
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+  EXPECT_EQ(ToDETypes(ds->GetOutputTypes()), types);
+  EXPECT_EQ(ToTensorShapeVec(ds->GetOutputShapes()), shapes);
+  EXPECT_EQ(ds->GetBatchSize(), 1);
+  EXPECT_EQ(ds->GetRepeatCount(), 1);
+  EXPECT_EQ(ds->GetNumClasses(), -1);
+  EXPECT_EQ(ds->GetDatasetSize(), 10);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistDatasetWithInvalidDir) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistDatasetWithInvalidDir.";
+
+  // Create a EMnist Dataset
+  std::shared_ptr<Dataset> ds = EMnist("", "mnist", "train", std::make_shared<RandomSampler>(false, 5));
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  // Expect failure: invalid EMnist input
+  EXPECT_EQ(iter, nullptr);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistDatasetWithInvalidUsage) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistDatasetWithInvalidUsage.";
+
+  // Create a EMnist Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "validation");
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  // Expect failure: invalid EMnist input, validation is not a valid usage
+  EXPECT_EQ(iter, nullptr);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistDatasetWithInvalidName) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistDatasetWithInvalidName.";
+
+  // Create a EMnist Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "validation", "train");
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  // Expect failure: invalid EMnist input, validation is not a valid name
+  EXPECT_EQ(iter, nullptr);
+}
+
+TEST_F(MindDataTestPipeline, TestEMnistDatasetWithNullSampler) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestEMnistDatasetWithNullSampler.";
+
+  // Create a EMnist Dataset
+  std::string folder_path = datasets_root_path_ + "/testEMnistDataset";
+  std::shared_ptr<Dataset> ds = EMnist(folder_path, "mnist", "train", nullptr);
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  // Expect failure: invalid EMnist input, sampler cannot be nullptr
+  EXPECT_EQ(iter, nullptr);
+}

tests/ut/python/dataset/test_datasets_emnist.py

@@ -0,0 +1,481 @@
+# Copyright 2021 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://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.
+# ==============================================================================
+"""
+Test EMnist dataset operators
+"""
+
+import os
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pytest
+
+import mindspore.dataset as ds
+import mindspore.dataset.vision.c_transforms as vision
+from mindspore import log as logger
+
+DATA_DIR = "../data/dataset/testEMnistDataset"
+
+
+def load_emnist(path, usage, name):
+    """
+    load EMnist data
+    """
+    image_path = []
+    label_path = []
+    image_ext = "images-idx3-ubyte"
+    label_ext = "labels-idx1-ubyte"
+    train_prefix = "emnist-" + name + "-train-"
+    test_prefix = "emnist-" + name + "-test-"
+    assert usage in ["train", "test", "all"]
+    if usage == "train":
+        image_path.append(os.path.realpath(os.path.join(path, train_prefix + image_ext)))
+        label_path.append(os.path.realpath(os.path.join(path, train_prefix + label_ext)))
+    elif usage == "test":
+        image_path.append(os.path.realpath(os.path.join(path, test_prefix + image_ext)))
+        label_path.append(os.path.realpath(os.path.join(path, test_prefix + label_ext)))
+    elif usage == "all":
+        image_path.append(os.path.realpath(os.path.join(path, test_prefix + image_ext)))
+        label_path.append(os.path.realpath(os.path.join(path, test_prefix + label_ext)))
+        image_path.append(os.path.realpath(os.path.join(path, train_prefix + image_ext)))
+        label_path.append(os.path.realpath(os.path.join(path, train_prefix + label_ext)))
+    assert len(image_path) == len(label_path)
+    images = []
+    labels = []
+    for i, _ in enumerate(image_path):
+        with open(image_path[i], 'rb') as image_file:
+            image_file.read(16)
+            image = np.fromfile(image_file, dtype=np.uint8)
+            image = image.reshape(-1, 28, 28, 1)
+            image[image > 0] = 255  # Perform binarization to maintain consistency with our API
+            images.append(image)
+        with open(label_path[i], 'rb') as label_file:
+            label_file.read(8)
+            label = np.fromfile(label_file, dtype=np.uint8)
+            labels.append(label)
+
+    images = np.concatenate(images, 0)
+    labels = np.concatenate(labels, 0)
+
+    return images, labels
+
+
+def visualize_dataset(images, labels):
+    """
+    Helper function to visualize the dataset samples
+    """
+    num_samples = len(images)
+    for i in range(num_samples):
+        plt.subplot(1, num_samples, i + 1)
+        plt.imshow(images[i].squeeze(), cmap=plt.cm.gray)
+        plt.title(labels[i])
+    plt.show()
+
+
+def test_emnist_content_check():
+    """
+    Validate EMnistDataset image readings
+    """
+    logger.info("Test EMnistDataset Op with content check")
+    # train mnist
+    train_data = ds.EMnistDataset(DATA_DIR, name="mnist", usage="train", num_samples=10, shuffle=False)
+    images, labels = load_emnist(DATA_DIR, "train", "mnist")
+    num_iter = 0
+    # in this example, each dictionary has keys "image" and "label"
+    image_list, label_list = [], []
+    for i, data in enumerate(train_data.create_dict_iterator(num_epochs=1, output_numpy=True)):
+        image_list.append(data["image"])
+        label_list.append("label {}".format(data["label"]))
+        np.testing.assert_array_equal(data["image"], images[i])
+        np.testing.assert_array_equal(data["label"], labels[i])
+        num_iter += 1
+    assert num_iter == 10
+
+    # train byclass
+    train_data = ds.EMnistDataset(DATA_DIR, name="byclass", usage="train", num_samples=10, shuffle=False)
+    images, labels = load_emnist(DATA_DIR, "train", "byclass")
+    num_iter = 0
+    # in this example, each dictionary has keys "image" and "label"
+    image_list, label_list = [], []
+    for i, data in enumerate(train_data.create_dict_iterator(num_epochs=1, output_numpy=True)):
+        image_list.append(data["image"])
+        label_list.append("label {}".format(data["label"]))
+        np.testing.assert_array_equal(data["image"], images[i])
+        np.testing.assert_array_equal(data["label"], labels[i])
+        num_iter += 1
+    assert num_iter == 10
+
+    # test
+    test_data = ds.EMnistDataset(DATA_DIR, name="mnist", usage="test", num_samples=10, shuffle=False)
+    images, labels = load_emnist(DATA_DIR, "test", "mnist")
+    num_iter = 0
+    # in this example, each dictionary has keys "image" and "label"
+    image_list, label_list = [], []
+    for i, data in enumerate(test_data.create_dict_iterator(num_epochs=1, output_numpy=True)):
+        image_list.append(data["image"])
+        label_list.append("label {}".format(data["label"]))
+        np.testing.assert_array_equal(data["image"], images[i])
+        np.testing.assert_array_equal(data["label"], labels[i])
+        num_iter += 1
+    assert num_iter == 10
+
+
+def test_emnist_basic():
+    """
+    Validate EMnistDataset
+    """
+    logger.info("Test EMnistDataset Op")
+
+    # case 1: test loading whole dataset
+    train_data = ds.EMnistDataset(DATA_DIR, "mnist", "train")
+    num_iter1 = 0
+    for _ in train_data.create_dict_iterator(num_epochs=1):
+        num_iter1 += 1
+    assert num_iter1 == 10
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test")
+    num_iter = 0
+    for _ in test_data.create_dict_iterator(num_epochs=1):
+        num_iter += 1
+    assert num_iter == 10
+
+    # case 2: test num_samples
+    train_data = ds.EMnistDataset(DATA_DIR, "byclass", "train", num_samples=5)
+    num_iter2 = 0
+    for _ in train_data.create_dict_iterator(num_epochs=1):
+        num_iter2 += 1
+    assert num_iter2 == 5
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=5)
+    num_iter2 = 0
+    for _ in test_data.create_dict_iterator(num_epochs=1):
+        num_iter2 += 1
+    assert num_iter2 == 5
+
+    # case 3: test repeat
+    train_data = ds.EMnistDataset(DATA_DIR, "byclass", "train", num_samples=2)
+    train_data = train_data.repeat(5)
+    num_iter3 = 0
+    for _ in train_data.create_dict_iterator(num_epochs=1):
+        num_iter3 += 1
+    assert num_iter3 == 10
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=2)
+    test_data = test_data.repeat(5)
+    num_iter3 = 0
+    for _ in test_data.create_dict_iterator(num_epochs=1):
+        num_iter3 += 1
+    assert num_iter3 == 10
+
+    # case 4: test batch with drop_remainder=False
+    train_data = ds.EMnistDataset(DATA_DIR, "byclass", "train", num_samples=10)
+    assert train_data.get_dataset_size() == 10
+    assert train_data.get_batch_size() == 1
+
+    train_data = train_data.batch(batch_size=7)  # drop_remainder is default to be False
+    assert train_data.get_dataset_size() == 2
+    assert train_data.get_batch_size() == 7
+    num_iter4 = 0
+    for _ in train_data.create_dict_iterator(num_epochs=1):
+        num_iter4 += 1
+    assert num_iter4 == 2
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=10)
+    assert test_data.get_dataset_size() == 10
+    assert test_data.get_batch_size() == 1
+
+    test_data = test_data.batch(
+        batch_size=7)  # drop_remainder is default to be False
+    assert test_data.get_dataset_size() == 2
+    assert test_data.get_batch_size() == 7
+    num_iter4 = 0
+    for _ in test_data.create_dict_iterator(num_epochs=1):
+        num_iter4 += 1
+    assert num_iter4 == 2
+
+    # case 5: test batch with drop_remainder=True
+    train_data = ds.EMnistDataset(DATA_DIR, "byclass", "train", num_samples=10)
+    assert train_data.get_dataset_size() == 10
+    assert train_data.get_batch_size() == 1
+    train_data = train_data.batch(batch_size=7, drop_remainder=True)  # the rest of incomplete batch will be dropped
+    assert train_data.get_dataset_size() == 1
+    assert train_data.get_batch_size() == 7
+    num_iter5 = 0
+    for _ in train_data.create_dict_iterator(num_epochs=1):
+        num_iter5 += 1
+    assert num_iter5 == 1
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=10)
+    assert test_data.get_dataset_size() == 10
+    assert test_data.get_batch_size() == 1
+    test_data = test_data.batch(batch_size=7, drop_remainder=True)  # the rest of incomplete batch will be dropped
+    assert test_data.get_dataset_size() == 1
+    assert test_data.get_batch_size() == 7
+    num_iter5 = 0
+    for _ in test_data.create_dict_iterator(num_epochs=1):
+        num_iter5 += 1
+    assert num_iter5 == 1
+
+    # case 6: test get_col_names
+    dataset = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=10)
+    assert dataset.get_col_names() == ["image", "label"]
+
+
+def test_emnist_pk_sampler():
+    """
+    Test EMnistDataset with PKSampler
+    """
+    logger.info("Test EMnistDataset Op with PKSampler")
+    golden = [0, 0, 0, 1, 1, 1]
+
+    sampler = ds.PKSampler(3)
+    train_data = ds.EMnistDataset(DATA_DIR, "mnist", "train", sampler=sampler)
+    num_iter = 0
+    label_list = []
+    for item in train_data.create_dict_iterator(num_epochs=1, output_numpy=True):
+        label_list.append(item["label"])
+        num_iter += 1
+    np.testing.assert_array_equal(golden, label_list)
+    assert num_iter == 6
+
+    sampler = ds.PKSampler(3)
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "train", sampler=sampler)
+    num_iter = 0
+    label_list = []
+    for item in test_data.create_dict_iterator(num_epochs=1, output_numpy=True):
+        label_list.append(item["label"])
+        num_iter += 1
+    np.testing.assert_array_equal(golden, label_list)
+    assert num_iter == 6
+
+
+def test_emnist_sequential_sampler():
+    """
+    Test EMnistDataset with SequentialSampler
+    """
+    logger.info("Test EMnistDataset Op with SequentialSampler")
+    num_samples = 10
+    sampler = ds.SequentialSampler(num_samples=num_samples)
+    train_data1 = ds.EMnistDataset(DATA_DIR, "mnist", "train", sampler=sampler)
+    train_data2 = ds.EMnistDataset(DATA_DIR, "mnist", "train", shuffle=False, num_samples=num_samples)
+    label_list1, label_list2 = [], []
+    num_iter = 0
+    for item1, item2 in zip(train_data1.create_dict_iterator(num_epochs=1),
+                            train_data2.create_dict_iterator(num_epochs=1)):
+        label_list1.append(item1["label"].asnumpy())
+        label_list2.append(item2["label"].asnumpy())
+        num_iter += 1
+    np.testing.assert_array_equal(label_list1, label_list2)
+    assert num_iter == num_samples
+
+    num_samples = 10
+    sampler = ds.SequentialSampler(num_samples=num_samples)
+    test_data1 = ds.EMnistDataset(DATA_DIR, "mnist", "test", sampler=sampler)
+    test_data2 = ds.EMnistDataset(DATA_DIR, "mnist", "test", shuffle=False, num_samples=num_samples)
+    label_list1, label_list2 = [], []
+    num_iter = 0
+    for item1, item2 in zip(test_data1.create_dict_iterator(num_epochs=1),
+                            test_data2.create_dict_iterator(num_epochs=1)):
+        label_list1.append(item1["label"].asnumpy())
+        label_list2.append(item2["label"].asnumpy())
+        num_iter += 1
+    np.testing.assert_array_equal(label_list1, label_list2)
+    assert num_iter == num_samples
+
+
+def test_emnist_exception():
+    """
+    Test error cases for EMnistDataset
+    """
+    logger.info("Test error cases for EMnistDataset")
+    error_msg_1 = "sampler and shuffle cannot be specified at the same time"
+    with pytest.raises(RuntimeError, match=error_msg_1):
+        ds.EMnistDataset(DATA_DIR, "byclass", "train", shuffle=False, sampler=ds.PKSampler(3))
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", shuffle=False, sampler=ds.PKSampler(3))
+
+    error_msg_2 = "sampler and sharding cannot be specified at the same time"
+    with pytest.raises(RuntimeError, match=error_msg_2):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", sampler=ds.PKSampler(3), num_shards=2, shard_id=0)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", sampler=ds.PKSampler(3), num_shards=2, shard_id=0)
+
+    error_msg_3 = "num_shards is specified and currently requires shard_id as well"
+    with pytest.raises(RuntimeError, match=error_msg_3):
+        ds.EMnistDataset(DATA_DIR, "byclass", "train", num_shards=10)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", num_shards=10)
+
+    error_msg_4 = "shard_id is specified but num_shards is not"
+    with pytest.raises(RuntimeError, match=error_msg_4):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", shard_id=0)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", shard_id=0)
+
+    error_msg_5 = "Input shard_id is not within the required interval"
+    with pytest.raises(ValueError, match=error_msg_5):
+        ds.EMnistDataset(DATA_DIR, "byclass", "train", num_shards=5, shard_id=-1)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", num_shards=5, shard_id=-1)
+    with pytest.raises(ValueError, match=error_msg_5):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", num_shards=5, shard_id=5)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", num_shards=5, shard_id=5)
+    with pytest.raises(ValueError, match=error_msg_5):
+        ds.EMnistDataset(DATA_DIR, "byclass", "train", num_shards=2, shard_id=5)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", num_shards=2, shard_id=5)
+
+    error_msg_6 = "num_parallel_workers exceeds"
+    with pytest.raises(ValueError, match=error_msg_6):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", shuffle=False, num_parallel_workers=0)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", shuffle=False, num_parallel_workers=0)
+    with pytest.raises(ValueError, match=error_msg_6):
+        ds.EMnistDataset(DATA_DIR, "byclass", "train", shuffle=False, num_parallel_workers=256)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", shuffle=False, num_parallel_workers=256)
+    with pytest.raises(ValueError, match=error_msg_6):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", shuffle=False, num_parallel_workers=-2)
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", shuffle=False, num_parallel_workers=-2)
+
+    error_msg_7 = "Argument shard_id"
+    with pytest.raises(TypeError, match=error_msg_7):
+        ds.EMnistDataset(DATA_DIR, "mnist", "train", num_shards=2, shard_id="0")
+        ds.EMnistDataset(DATA_DIR, "mnist", "test", num_shards=2, shard_id="0")
+
+    def exception_func(item):
+        raise Exception("Error occur!")
+
+    error_msg_8 = "The corresponding data files"
+    with pytest.raises(RuntimeError, match=error_msg_8):
+        data = ds.EMnistDataset(DATA_DIR, "mnist", "train")
+        data = data.map(operations=exception_func, input_columns=["image"], num_parallel_workers=1)
+        for _ in data.__iter__():
+            pass
+    with pytest.raises(RuntimeError, match=error_msg_8):
+        data = ds.EMnistDataset(DATA_DIR, "mnist", "train")
+        data = data.map(operations=vision.Decode(), input_columns=["image"], num_parallel_workers=1)
+        data = data.map(operations=exception_func, input_columns=["image"], num_parallel_workers=1)
+        for _ in data.__iter__():
+            pass
+    with pytest.raises(RuntimeError, match=error_msg_8):
+        data = ds.EMnistDataset(DATA_DIR, "mnist", "train")
+        data = data.map(operations=exception_func, input_columns=["label"], num_parallel_workers=1)
+        for _ in data.__iter__():
+            pass
+
+
+def test_emnist_visualize(plot=False):
+    """
+    Visualize EMnistDataset results
+    """
+    logger.info("Test EMnistDataset visualization")
+
+    train_data = ds.EMnistDataset(DATA_DIR, "mnist", "train", num_samples=10, shuffle=False)
+    num_iter = 0
+    image_list, label_list = [], []
+    for item in train_data.create_dict_iterator(num_epochs=1, output_numpy=True):
+        image = item["image"]
+        label = item["label"]
+        image_list.append(image)
+        label_list.append("label {}".format(label))
+        assert isinstance(image, np.ndarray)
+        assert image.shape == (28, 28, 1)
+        assert image.dtype == np.uint8
+        assert label.dtype == np.uint32
+        num_iter += 1
+    assert num_iter == 10
+    if plot:
+        visualize_dataset(image_list, label_list)
+
+    test_data = ds.EMnistDataset(DATA_DIR, "mnist", "test", num_samples=10, shuffle=False)
+    num_iter = 0
+    image_list, label_list = [], []
+    for item in test_data.create_dict_iterator(num_epochs=1, output_numpy=True):
+        image = item["image"]
+        label = item["label"]
+        image_list.append(image)
+        label_list.append("label {}".format(label))
+        assert isinstance(image, np.ndarray)
+        assert image.shape == (28, 28, 1)
+        assert image.dtype == np.uint8
+        assert label.dtype == np.uint32
+        num_iter += 1
+    assert num_iter == 10
+    if plot:
+        visualize_dataset(image_list, label_list)
+
+
+def test_emnist_usage():
+    """
+    Validate EMnistDataset image readings
+    """
+    logger.info("Test EMnistDataset usage flag")
+
+    def test_config(usage, emnist_path=None):
+        emnist_path = DATA_DIR if emnist_path is None else emnist_path
+        try:
+            data = ds.EMnistDataset(emnist_path, "mnist", usage=usage, shuffle=False)
+            num_rows = 0
+            for _ in data.create_dict_iterator(num_epochs=1, output_numpy=True):
+                num_rows += 1
+        except (ValueError, TypeError, RuntimeError) as e:
+            return str(e)
+        return num_rows
+
+    assert test_config("train") == 10
+    assert test_config("test") == 10
+    assert test_config("all") == 20
+
+    assert "usage is not within the valid set of ['train', 'test', 'all']" in test_config("invalid")
+    assert "Argument usage with value ['list'] is not of type [<class 'str'>]" in test_config(["list"])
+
+    # change this directory to the folder that contains all emnist files
+    all_files_path = None
+
+    # the following tests on the entire datasets
+    if all_files_path is not None:
+        assert test_config("train", all_files_path) == 10000
+        assert test_config("test", all_files_path) == 60000
+        assert test_config("all", all_files_path) == 70000
+        assert ds.EMnistDataset(all_files_path, "mnist", usage="test").get_dataset_size() == 10000
+        assert ds.EMnistDataset(all_files_path, "mnist", usage="test").get_dataset_size() == 60000
+        assert ds.EMnistDataset(all_files_path, "mnist", usage="all").get_dataset_size() == 70000
+
+
+def test_emnist_name():
+    """
+    Validate EMnistDataset image readings
+    """
+    def test_config(name, usage, emnist_path=None):
+        emnist_path = DATA_DIR if emnist_path is None else emnist_path
+        try:
+            data = ds.EMnistDataset(emnist_path, name, usage=usage, shuffle=False)
+            num_rows = 0
+            for _ in data.create_dict_iterator(num_epochs=1, output_numpy=True):
+                num_rows += 1
+        except (ValueError, TypeError, RuntimeError) as e:
+            return str(e)
+        return num_rows
+
+    assert test_config("mnist", "train") == 10
+    assert test_config("mnist", "test") == 10
+    assert test_config("byclass", "train") == 10
+    assert "name is not within the valid set of " + \
+            "['byclass', 'bymerge', 'balanced', 'letters', 'digits', 'mnist']" in test_config("invalid", "train")
+    assert "Argument name with value ['list'] is not of type [<class 'str'>]" in test_config(["list"], "train")
+
+
+if __name__ == '__main__':
+    test_emnist_content_check()
+    test_emnist_basic()
+    test_emnist_pk_sampler()
+    test_emnist_sequential_sampler()
+    test_emnist_exception()
+    test_emnist_visualize(plot=True)
+    test_emnist_usage()
+    test_emnist_name()