Convert all the model types without constructor parameters to objects (…

…#492)
Kotlin · Dec 1, 2022 · fc72cce · fc72cce
1 parent ee899bf
commit fc72cce
Show file tree

Hide file tree

Showing 8 changed files with 48 additions and 48 deletions.
diff --git a/examples/src/main/kotlin/examples/onnx/cv/efficicentnet/EfficientNet4Lite.kt b/examples/src/main/kotlin/examples/onnx/cv/efficicentnet/EfficientNet4Lite.kt
@@ -29,7 +29,7 @@ import java.io.File
 fun efficientNet4LitePrediction() {
     val modelHub = ONNXModelHub(cacheDirectory = File("cache/pretrainedModels"))
 
-    val modelType = ONNXModels.CV.EfficientNet4Lite()
+    val modelType = ONNXModels.CV.EfficientNet4Lite
     val model = modelHub.loadModel(modelType)
     model.printSummary()
 

diff --git a/examples/src/main/kotlin/examples/onnx/cv/resnet/ResNet18.kt b/examples/src/main/kotlin/examples/onnx/cv/resnet/ResNet18.kt
@@ -16,7 +16,7 @@ import org.jetbrains.kotlinx.dl.onnx.inference.ONNXModels
  * - Special preprocessing (used in ResNet'18 during training on ImageNet dataset) is applied to each image before prediction.
  */
 fun resnet18prediction() {
-    runImageRecognitionPrediction(ONNXModels.CV.ResNet18())
+    runImageRecognitionPrediction(ONNXModels.CV.ResNet18)
 }
 
 /** */

diff --git a/examples/src/main/kotlin/examples/onnx/cv/resnet/ResNet18LightAPI.kt b/examples/src/main/kotlin/examples/onnx/cv/resnet/ResNet18LightAPI.kt
@@ -20,7 +20,7 @@ fun resnet18LightAPIPrediction() {
     val modelHub =
         ONNXModelHub(cacheDirectory = File("cache/pretrainedModels"))
 
-    val model = ONNXModels.CV.ResNet18().pretrainedModel(modelHub)
+    val model = ONNXModels.CV.ResNet18.pretrainedModel(modelHub)
     model.printSummary()
 
     model.use {

diff --git a/examples/src/test/kotlin/examples/onnx/ExecutionProvidersTestSuite.kt b/examples/src/test/kotlin/examples/onnx/ExecutionProvidersTestSuite.kt
@@ -17,16 +17,16 @@ import java.io.File
 class ExecutionProvidersTestSuite {
     private fun resnetModelsInference(executionProvider: ExecutionProvider) {
         val modelsToTest = listOf(
-            ONNXModels.CV.ResNet101(),
-            ONNXModels.CV.ResNet101v2(),
-            ONNXModels.CV.ResNet152(),
-            ONNXModels.CV.ResNet152v2(),
-            ONNXModels.CV.ResNet18(),
-            ONNXModels.CV.ResNet18v2(),
-            ONNXModels.CV.ResNet34(),
-            ONNXModels.CV.ResNet34v2(),
-            ONNXModels.CV.ResNet50(),
-            ONNXModels.CV.ResNet50v2(),
+            ONNXModels.CV.ResNet101,
+            ONNXModels.CV.ResNet101v2,
+            ONNXModels.CV.ResNet152,
+            ONNXModels.CV.ResNet152v2,
+            ONNXModels.CV.ResNet18,
+            ONNXModels.CV.ResNet18v2,
+            ONNXModels.CV.ResNet34,
+            ONNXModels.CV.ResNet34v2,
+            ONNXModels.CV.ResNet50,
+            ONNXModels.CV.ResNet50v2,
             ONNXModels.CV.ResNet50custom,
         )
 
@@ -78,7 +78,7 @@ class ExecutionProvidersTestSuite {
     @Test
     fun executionProvidersDuplicatesTest() {
         val modelHub = ONNXModelHub(cacheDirectory = File("cache/pretrainedModels"))
-        val model = modelHub.loadModel(ONNXModels.CV.ResNet18())
+        val model = modelHub.loadModel(ONNXModels.CV.ResNet18)
 
         model.use {
             assertDoesNotThrow {
@@ -90,7 +90,7 @@ class ExecutionProvidersTestSuite {
     @Test
     fun twoCpuExecutorsWithDifferentAllocatorsTest() {
         val modelHub = ONNXModelHub(cacheDirectory = File("cache/pretrainedModels"))
-        val model = modelHub.loadModel(ONNXModels.CV.ResNet18())
+        val model = modelHub.loadModel(ONNXModels.CV.ResNet18)
 
         model.use {
             assertThrows<IllegalArgumentException> {

diff --git a/examples/src/test/kotlin/examples/onnx/ModelCopyingTestSuite.kt b/examples/src/test/kotlin/examples/onnx/ModelCopyingTestSuite.kt
@@ -34,7 +34,7 @@ class ModelCopyingTestSuite {
                 ONNXModels.CV.EfficientNetB5(),
                 ONNXModels.CV.EfficientNetB6(),
                 ONNXModels.CV.EfficientNetB7(),
-                ONNXModels.CV.EfficientNet4Lite()
+                ONNXModels.CV.EfficientNet4Lite
             ),
             "datasets/vgg/image0.jpg",
             ImageRecognitionModel::predictObject
@@ -45,17 +45,17 @@ class ModelCopyingTestSuite {
     fun resNetCopyTest() {
         runCopyTest(
             listOf(
-                ONNXModels.CV.ResNet18(),
-                ONNXModels.CV.ResNet18v2(),
-                ONNXModels.CV.ResNet34(),
-                ONNXModels.CV.ResNet34v2(),
-                ONNXModels.CV.ResNet50(),
-                ONNXModels.CV.ResNet50v2(),
+                ONNXModels.CV.ResNet18,
+                ONNXModels.CV.ResNet18v2,
+                ONNXModels.CV.ResNet34,
+                ONNXModels.CV.ResNet34v2,
+                ONNXModels.CV.ResNet50,
+                ONNXModels.CV.ResNet50v2,
                 ONNXModels.CV.ResNet50custom,
-                ONNXModels.CV.ResNet101(),
-                ONNXModels.CV.ResNet101v2(),
-                ONNXModels.CV.ResNet152(),
-                ONNXModels.CV.ResNet152v2()
+                ONNXModels.CV.ResNet101,
+                ONNXModels.CV.ResNet101v2,
+                ONNXModels.CV.ResNet152,
+                ONNXModels.CV.ResNet152v2
             ),
             "datasets/vgg/image0.jpg",
             ImageRecognitionModel::predictObject

diff --git a/examples/src/test/kotlin/examples/onnx/cv/OnnxResNetTestSuite.kt b/examples/src/test/kotlin/examples/onnx/cv/OnnxResNetTestSuite.kt
@@ -26,27 +26,27 @@ class OnnxResNetTestSuite {
 
     @Test
     fun resnet18v2predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet18v2())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet18v2)
     }
 
     @Test
     fun resnet34predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet34())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet34)
     }
 
     @Test
     fun resnet34v2predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet34v2())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet34v2)
     }
 
     @Test
     fun resnet50predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet50())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet50)
     }
 
     @Test
     fun resnet50v2predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet50v2())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet50v2)
     }
 
     @Test
@@ -61,22 +61,22 @@ class OnnxResNetTestSuite {
 
     @Test
     fun resnet101predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet101())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet101)
     }
 
     @Test
     fun resnet101v2predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet101v2())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet101v2)
     }
 
     @Test
     fun resnet152predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet152())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet152)
     }
 
     @Test
     fun resnet152v2predictionTest() {
-        runImageRecognitionPrediction(ONNXModels.CV.ResNet152v2())
+        runImageRecognitionPrediction(ONNXModels.CV.ResNet152v2)
     }
 
     @Test

diff --git a/onnx/src/androidMain/kotlin/org/jetbrains/kotlinx/dl/onnx/inference/ONNXModels.kt b/onnx/src/androidMain/kotlin/org/jetbrains/kotlinx/dl/onnx/inference/ONNXModels.kt
@@ -62,7 +62,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/efficientnet-lite4">
          *    Official EfficientNet4Lite model from ONNX Github.</a>
          */
-        public class EfficientNet4Lite : CV("efficientnet_lite4", channelsFirst = false) {
+        public object EfficientNet4Lite : CV("efficientnet_lite4", channelsFirst = false) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = InputType.TF.preprocessing(channelsLast = !channelsFirst)
         }
@@ -83,7 +83,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/efficientnet-lite4">
          *    Official EfficientNet4Lite model from ONNX Github.</a>
          */
-        public class MobilenetV1 : CV("mobilenet_v1", channelsFirst = false) {
+        public object MobilenetV1 : CV("mobilenet_v1", channelsFirst = false) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = pipeline<Pair<FloatArray, TensorShape>>()
                     .rescale { scalingCoefficient = 255f }

diff --git a/onnx/src/jvmMain/kotlin/org/jetbrains/kotlinx/dl/onnx/inference/ONNXModels.kt b/onnx/src/jvmMain/kotlin/org/jetbrains/kotlinx/dl/onnx/inference/ONNXModels.kt
@@ -71,7 +71,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet18 : CV("models/onnx/cv/resnet/resnet18-v1", channelsFirst = true) {
+        public object ResNet18 : CV("models/onnx/cv/resnet/resnet18-v1", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
         }
@@ -93,7 +93,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet34 : CV("models/onnx/cv/resnet/resnet34-v1", channelsFirst = true) {
+        public object ResNet34 : CV("models/onnx/cv/resnet/resnet34-v1", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
         }
@@ -115,7 +115,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet50 :
+        public object ResNet50 :
             CV("models/onnx/cv/resnet/resnet50-v1", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -138,7 +138,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet101 :
+        public object ResNet101 :
             CV("models/onnx/cv/resnet/resnet101-v1", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -161,7 +161,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet152 :
+        public object ResNet152 :
             CV("models/onnx/cv/resnet/resnet152-v1", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -184,7 +184,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet18v2 :
+        public object ResNet18v2 :
             CV("models/onnx/cv/resnet/resnet18-v2", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -207,7 +207,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet34v2 :
+        public object ResNet34v2 :
             CV("models/onnx/cv/resnet/resnet34-v2", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -230,7 +230,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet50v2 :
+        public object ResNet50v2 :
             CV("models/onnx/cv/resnet/resnet50-v2", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -253,7 +253,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet101v2 :
+        public object ResNet101v2 :
             CV("models/onnx/cv/resnet/resnet101-v2", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -276,7 +276,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/resnet">
          *    Official ResNet model from ONNX Github.</a>
          */
-        public class ResNet152v2 :
+        public object ResNet152v2 :
             CV("models/onnx/cv/resnet/resnet152-v2", channelsFirst = true) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = resNetOnnxPreprocessing()
@@ -299,7 +299,7 @@ public object ONNXModels {
          * @see <a href="https://github.com/onnx/models/tree/main/vision/classification/efficientnet-lite4">
          *    Official EfficientNet4Lite model from ONNX Github.</a>
          */
-        public class EfficientNet4Lite :
+        public object EfficientNet4Lite :
             CV("models/onnx/cv/efficientnet/efficientnet-lite4", channelsFirst = false) {
             override val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
                 get() = InputType.TF.preprocessing(channelsLast = !channelsFirst)