!3039 [Dataset Python UT] adding coverage to Cifar Op and update test dataset

Merge pull request !3039 from TinaMengtingZhang/dataset-op-ut-cifar

mindspore/dataset/core/validator_helpers.py

@@ -271,6 +271,8 @@ def check_sampler_shuffle_shard_options(param_dict):
     if sampler is not None:
         if shuffle is not None:
             raise RuntimeError("sampler and shuffle cannot be specified at the same time.")
+        if num_shards is not None:
+            raise RuntimeError("sampler and sharding cannot be specified at the same time.")
 
     if num_shards is not None:
         check_pos_int32(num_shards)

tests/ut/data/dataset/testCifar100Data/datasetSchema.json

@@ -1,21 +0,0 @@
-{
-  "datasetType": "CIFAR100",
-  "numRows": 100,
-  "columns": {
-    "image": {
-      "type": "uint8",
-      "rank": 1,
-	  "t_impl": "cvmat"
-    },
-    "coarse_label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    },
-    "fine_label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    }
-  }
-}

tests/ut/data/dataset/testCifar100Data/datasetSchemaTestRepeat.json

@@ -1,21 +0,0 @@
-{
-  "datasetType": "CIFAR100",
-  "numRows": 33,
-  "columns": {
-    "image": {
-      "type": "uint8",
-      "rank": 1,
-	  "t_impl": "cvmat"
-    },
-    "coarse_label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    },
-    "fine_label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    }
-  }
-}

tests/ut/data/dataset/testCifar10Data/datasetDistributionAll.json

@@ -1,9 +0,0 @@
-{ 
-  "deviceNum" : 3,
-  "deviceId" : 1,
-  "shardConfig" : "ALL",
-  "shuffle" : "ON",
-  "seed" : 0,
-  "epoch" : 2
-}
-

tests/ut/data/dataset/testCifar10Data/datasetDistributionRandom.json

@@ -1,9 +0,0 @@
-{ 
-  "deviceNum" : 3,
-  "deviceId" : 1,
-  "shardConfig" : "RANDOM",
-  "shuffle" : "ON",
-  "seed" : 0,
-  "epoch" : 1
-}
-

tests/ut/data/dataset/testCifar10Data/datasetDistributionUnique.json

@@ -1,9 +0,0 @@
-{ 
-  "deviceNum" : 3,
-  "deviceId" : 1,
-  "shardConfig" : "UNIQUE",
-  "shuffle" : "ON",
-  "seed" : 0,
-  "epoch" : 3
-}
-

tests/ut/data/dataset/testCifar10Data/datasetSchema.json

@@ -1,16 +0,0 @@
-{
-  "datasetType": "CIFAR10",
-  "numRows": 60000,
-  "columns": {
-    "image": {
-      "type": "uint8",
-      "rank": 1,
-	  "t_impl": "cvmat"
-    },
-    "label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    }
-  }
-}

tests/ut/data/dataset/testCifar10Data/datasetSchemaTestRepeat.json

@@ -1,16 +0,0 @@
-{
-  "datasetType": "CIFAR10",
-  "numRows": 33,
-  "columns": {
-    "image": {
-      "type": "uint8",
-      "rank": 1,
-	  "t_impl": "cvmat"
-    },
-    "label" : {
-      "type": "uint32",
-      "rank": 1,
-      "t_impl": "flex"
-    }
-  }
-}

tests/ut/python/dataset/test_cifarop.py

@@ -1,91 +0,0 @@
-# Copyright 2019 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.
-# ==============================================================================
-import os
-
-import numpy as np
-
-import mindspore.dataset as ds
-from mindspore import log as logger
-
-# Data for CIFAR and MNIST are not part of build tree
-# They need to be downloaded directly
-# prep_data.py can be executed or code below
-# import sys
-# sys.path.insert(0,"../../data")
-# import prep_data
-# prep_data.download_all_for_test("../../data")
-DATA_DIR_10 = "../data/dataset/testCifar10Data"
-DATA_DIR_100 = "../data/dataset/testCifar100Data"
-
-
-def load_cifar(path):
-    raw = np.empty(0, dtype=np.uint8)
-    for file_name in os.listdir(path):
-        if file_name.endswith(".bin"):
-            with open(os.path.join(path, file_name), mode='rb') as file:
-                raw = np.append(raw, np.fromfile(file, dtype=np.uint8), axis=0)
-    raw = raw.reshape(-1, 3073)
-    labels = raw[:, 0]
-    images = raw[:, 1:]
-    images = images.reshape(-1, 3, 32, 32)
-    images = images.transpose(0, 2, 3, 1)
-    return images, labels
-
-
-def test_case_dataset_cifar10():
-    """
-    dataset parameter
-    """
-    logger.info("Test dataset parameter")
-    # apply dataset operations
-    data1 = ds.Cifar10Dataset(DATA_DIR_10, 100)
-
-    num_iter = 0
-    for _ in data1.create_dict_iterator():
-        # in this example, each dictionary has keys "image" and "label"
-        num_iter += 1
-    assert num_iter == 100
-
-
-def test_case_dataset_cifar100():
-    """
-    dataset parameter
-    """
-    logger.info("Test dataset parameter")
-    # apply dataset operations
-    data1 = ds.Cifar100Dataset(DATA_DIR_100, 100)
-
-    num_iter = 0
-    for _ in data1.create_dict_iterator():
-        # in this example, each dictionary has keys "image" and "label"
-        num_iter += 1
-    assert num_iter == 100
-
-
-def test_reading_cifar10():
-    """
-    Validate CIFAR10 image readings
-    """
-    data1 = ds.Cifar10Dataset(DATA_DIR_10, 100, shuffle=False)
-    images, labels = load_cifar(DATA_DIR_10)
-    for i, d in enumerate(data1.create_dict_iterator()):
-        np.testing.assert_array_equal(d["image"], images[i])
-        np.testing.assert_array_equal(d["label"], labels[i])
-
-
-if __name__ == '__main__':
-    test_case_dataset_cifar10()
-    test_case_dataset_cifar100()
-    test_reading_cifar10()

tests/ut/python/dataset/test_config.py

@@ -245,17 +245,17 @@ def test_deterministic_run_distribution():
 
     # First dataset
     data1 = ds.TFRecordDataset(DATA_DIR, SCHEMA_DIR, columns_list=["image"], shuffle=False)
-    random_crop_op = c_vision.RandomHorizontalFlip(0.1)
+    random_horizontal_flip_op = c_vision.RandomHorizontalFlip(0.1)
     decode_op = c_vision.Decode()
     data1 = data1.map(input_columns=["image"], operations=decode_op)
-    data1 = data1.map(input_columns=["image"], operations=random_crop_op)
+    data1 = data1.map(input_columns=["image"], operations=random_horizontal_flip_op)
 
     # Second dataset
     data2 = ds.TFRecordDataset(DATA_DIR, SCHEMA_DIR, columns_list=["image"], shuffle=False)
     data2 = data2.map(input_columns=["image"], operations=decode_op)
     # If seed is set up on constructor, so the two ops output deterministic sequence
-    random_crop_op2 = c_vision.RandomHorizontalFlip(0.1)
+    random_horizontal_flip_op2 = c_vision.RandomHorizontalFlip(0.1)
-    data2 = data2.map(input_columns=["image"], operations=random_crop_op2)
+    data2 = data2.map(input_columns=["image"], operations=random_horizontal_flip_op2)
 
     for item1, item2 in zip(data1.create_dict_iterator(), data2.create_dict_iterator()):
         np.testing.assert_equal(item1["image"], item2["image"])