2022-09-20 15:40:53 +02:00

122 lines
4.4 KiB
Java

/*
* ******************************************************************************
* *
* *
* * 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.
* *
* * See the NOTICE file distributed with this work for additional
* * information regarding copyright ownership.
* * 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
* *****************************************************************************
*/
package org.deeplearning4j.datasets.iterator;
import lombok.extern.slf4j.Slf4j;
import org.deeplearning4j.BaseDL4JTest;
import org.deeplearning4j.datasets.iterator.impl.EmnistDataSetIterator;
import org.junit.jupiter.api.Test;
import org.nd4j.linalg.api.buffer.DataType;
import org.nd4j.linalg.api.ndarray.INDArray;
import org.nd4j.linalg.dataset.DataSet;
import org.nd4j.linalg.factory.Nd4j;
import static org.junit.jupiter.api.Assertions.*;
@Slf4j
public class TestEmnistDataSetIterator extends BaseDL4JTest {
@Override
public DataType getDataType(){
return DataType.FLOAT;
}
@Test
public void testEmnistDataSetIterator() throws Exception {
int batchSize = 128;
EmnistDataSetIterator.Set[] sets;
if(isIntegrationTests()){
sets = EmnistDataSetIterator.Set.values();
} else {
sets = new EmnistDataSetIterator.Set[]{EmnistDataSetIterator.Set.MNIST, EmnistDataSetIterator.Set.LETTERS};
}
for (EmnistDataSetIterator.Set s : sets) {
boolean isBalanced = EmnistDataSetIterator.isBalanced(s);
int numLabels = EmnistDataSetIterator.numLabels(s);
INDArray labelCounts = null;
for (boolean train : new boolean[] {true, false}) {
if (isBalanced && train) {
labelCounts = Nd4j.create(numLabels);
} else {
labelCounts = null;
}
log.info("Starting test: {}, {}", s, (train ? "train" : "test"));
EmnistDataSetIterator iter = new EmnistDataSetIterator(s, batchSize, train, 12345);
assertTrue(iter.asyncSupported());
assertTrue(iter.resetSupported());
int expNumExamples;
if (train) {
expNumExamples = EmnistDataSetIterator.numExamplesTrain(s);
} else {
expNumExamples = EmnistDataSetIterator.numExamplesTest(s);
}
assertEquals(numLabels, iter.getLabels().size());
assertEquals(numLabels, iter.getLabelsArrays().length);
char[] labelArr = iter.getLabelsArrays();
for (char c : labelArr) {
boolean isExpected = (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z');
assertTrue(isExpected);
}
int totalCount = 0;
while (iter.hasNext()) {
DataSet ds = iter.next();
assertNotNull(ds.getFeatures());
assertNotNull(ds.getLabels());
assertEquals(ds.getFeatures().size(0), ds.getLabels().size(0));
totalCount += ds.getFeatures().size(0);
assertEquals(784, ds.getFeatures().size(1));
assertEquals(numLabels, ds.getLabels().size(1));
if (isBalanced && train) {
labelCounts.addi(ds.getLabels().sum(0));
}
}
assertEquals(expNumExamples, totalCount);
if (isBalanced && train) {
int min = labelCounts.minNumber().intValue();
int max = labelCounts.maxNumber().intValue();
int exp = expNumExamples / numLabels;
assertTrue(min > 0);
assertEquals(exp, min);
assertEquals(exp, max);
}
}
}
}
}