update jdll to improve .npy file to imglib2 array tensor conversion

pom.xml

@@ -5,7 +5,7 @@
 	<parent>
 		<groupId>org.scijava</groupId>
 		<artifactId>pom-scijava</artifactId>
-		<version>36.0.0</version>
+		<version>37.0.0</version>
 		<relativePath />
 	</parent>
 

src/main/java/io/bioimage/modelrunner/numpy/DecodeNumpy.java

@@ -34,11 +34,25 @@
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
+import io.bioimage.modelrunner.tensor.Utils;
 import io.bioimage.modelrunner.utils.IndexingUtils;
 import net.imglib2.Cursor;
 import net.imglib2.RandomAccessibleInterval;
 import net.imglib2.img.Img;
 import net.imglib2.img.array.ArrayImgFactory;
+import net.imglib2.img.array.ArrayImgs;
+import net.imglib2.img.basictypeaccess.ByteAccess;
+import net.imglib2.img.basictypeaccess.DoubleAccess;
+import net.imglib2.img.basictypeaccess.FloatAccess;
+import net.imglib2.img.basictypeaccess.IntAccess;
+import net.imglib2.img.basictypeaccess.LongAccess;
+import net.imglib2.img.basictypeaccess.ShortAccess;
+import net.imglib2.img.basictypeaccess.nio.ByteBufferAccess;
+import net.imglib2.img.basictypeaccess.nio.DoubleBufferAccess;
+import net.imglib2.img.basictypeaccess.nio.FloatBufferAccess;
+import net.imglib2.img.basictypeaccess.nio.IntBufferAccess;
+import net.imglib2.img.basictypeaccess.nio.LongBufferAccess;
+import net.imglib2.img.basictypeaccess.nio.ShortBufferAccess;
 import net.imglib2.type.NativeType;
 import net.imglib2.type.numeric.RealType;
 import net.imglib2.type.numeric.integer.ByteType;
@@ -305,28 +319,49 @@ private static void readData(DataInputStream dis, ByteBuffer data, int len) thro
      *         If the tensor type is not supported.
      */
     @SuppressWarnings("unchecked")
-    public static <T extends NativeType<T>> Img<T> build(ByteBuffer buf, ByteOrder byteOrder, String dtype, long[] shape) throws IllegalArgumentException
+    public static <T extends NativeType<T>> RandomAccessibleInterval<T> build(ByteBuffer buf, ByteOrder byteOrder, String dtype, long[] shape) throws IllegalArgumentException
     {
+    	long[] transposedShape = new long[shape.length];
+    	for (int i = 0; i < shape.length; i ++)
+    		transposedShape[i] = shape[shape.length - i - 1];
     	if (dtype.equals("int8")) {
-    		return (Img<T>) buildInt8(buf, byteOrder, shape);
+    		ByteAccess access = new ByteBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.bytes( access, shape ));
+    		//return (Img<T>) buildInt8(buf, byteOrder, shape);
     	} else if (dtype.equals("uint8")) {
-    		return (Img<T>) buildUInt8(buf, byteOrder, shape);
+    		ByteAccess access = new ByteBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.unsignedBytes( access, shape ));
+    		//return (Img<T>) buildUInt8(buf, byteOrder, shape);
     	} else if (dtype.equals("int16")) {
-    		return (Img<T>) buildInt16(buf, byteOrder, shape);
+    		ShortAccess access = new ShortBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.shorts( access, shape ));
+    		//return (Img<T>) buildInt16(buf, byteOrder, shape);
     	} else if (dtype.equals("uint16")) {
-    		return (Img<T>) buildUInt16(buf, byteOrder, shape);
+    		ShortAccess access = new ShortBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.unsignedShorts( access, shape ));
+    		//return (Img<T>) buildUInt16(buf, byteOrder, shape);
     	} else if (dtype.equals("int32")) {
-    		return (Img<T>) buildInt32(buf, byteOrder, shape);
+    		IntAccess access = new IntBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.ints( access, shape ));
+    		//return (Img<T>) buildInt32(buf, byteOrder, shape);
     	} else if (dtype.equals("uint32")) {
-    		return (Img<T>) buildUInt32(buf, byteOrder, shape);
+    		IntAccess access = new IntBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.unsignedInts( access, shape ));
+    		//return (Img<T>) buildUInt32(buf, byteOrder, shape);
     	} else if (dtype.equals("int64")) {
-    		return (Img<T>) buildInt64(buf, byteOrder, shape);
+    		LongAccess access = new LongBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.longs( access, shape ));
+    		//return (Img<T>) buildInt64(buf, byteOrder, shape);
     	} else if (dtype.equals("float32")) {
-    		return (Img<T>) buildFloat32(buf, byteOrder, shape);
+    		FloatAccess access = new FloatBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.floats( access, shape ));
+    		//return (Img<T>) buildFloat32(buf, byteOrder, shape);
     	} else if (dtype.equals("float64")) {
-    		return (Img<T>) buildFloat64(buf, byteOrder, shape);
+    		DoubleAccess access = new DoubleBufferAccess(buf, true);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(ArrayImgs.doubles( access, shape ));
+    		//return (Img<T>) buildFloat64(buf, byteOrder, shape);
     	} else if (dtype.equals("bool")) {
-    		return (Img<T>) buildBoolean(buf, byteOrder, shape);
+    		return (RandomAccessibleInterval<T>) Utils.transpose(buildBoolean(buf, byteOrder, shape));
     	} else {
             throw new IllegalArgumentException("Unsupported data type of numpy array: " + dtype);
     	}

src/main/java/io/bioimage/modelrunner/tensor/Utils.java

@@ -0,0 +1,59 @@
+/*-
+ * #%L
+ * Use deep learning frameworks from Java in an agnostic and isolated way.
+ * %%
+ * Copyright (C) 2022 - 2023 Institut Pasteur and BioImage.IO developers.
+ * %%
+ * 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.
+ * #L%
+ */
+package io.bioimage.modelrunner.tensor;
+
+import net.imglib2.RandomAccessibleInterval;
+import net.imglib2.transform.integer.MixedTransform;
+import net.imglib2.type.numeric.NumericType;
+import net.imglib2.type.numeric.RealType;
+import net.imglib2.util.Intervals;
+import net.imglib2.view.MixedTransformView;
+import net.imglib2.view.Views;
+
+/**
+ * Provide static methods that can be used to make transformations on tensors
+ *
+ * @author Carlos Garcia Lopez de Haro
+ */
+public final class Utils
+{
+	/**
+	 * Method that transposes every dimension of a {@link RandomAccessibleInterval}.
+	 * If a {@link RandomAccessibleInterval} with {3, 256, 512} dimensions is provided,
+	 * it will be converted into one of {512, 256, 3}
+	 * @param <T>
+	 * 	possible data types of the {@link RandomAccessibleInterval}
+	 * @param rai
+	 * 	{@link RandomAccessibleInterval} to be transposed
+	 * @return the transposed {@link RandomAccessibleInterval}
+	 */
+	public static <T extends NumericType<T> & RealType<T>> 
+	RandomAccessibleInterval<T> transpose(RandomAccessibleInterval<T> rai){
+		long[] tensorShape = rai.dimensionsAsLongArray();
+		MixedTransform t = new MixedTransform( tensorShape.length, tensorShape.length );
+		int[] transposeAxesOrderChange = new int[tensorShape.length];
+		for (int i = 0; i < tensorShape.length; i ++) transposeAxesOrderChange[i] = tensorShape.length - 1 - i;
+		t.setComponentMapping(transposeAxesOrderChange);
+		long[] minMax = new long[tensorShape.length];
+		for (int i = 0; i < tensorShape.length; i ++) minMax[i * 2 + 1] = tensorShape[i];
+		return Views.interval(new MixedTransformView<T>( rai, t ), 
+				Intervals.createMinMax(tensorShape));
+	}
+}