Tensorflow functions and applications

* initial step
 * Tensorflow models functional model redesign

      -- Based on https://tzolov.github.io/mind-model-services
      -- Resolves #5

 * Add object detection processor README
 * Add image recognition processor README
 * Initial Tensorflow commonn README
 * Initial Tensorflow commonn README
 * Tensorflow common diagram
 * Tensorflow docs code
 * Tensorflow docs code snippets improve
 * Tensorflow docs code snippets improve
 * Tensorflow docs code snippets improve
 * Tensorflow docs code snippets improve
 * Add semantic segmentation function. add object detecteion function readme
 * oo images
 * Furether oo readme improvments
 * Final obj detection readme fixes
 * Add image recognition readme
 * Add image recognition readme 2
 * Semantic segmentation readme
 * Segmentation readme
 * Semantic segmentation readme 3
 * Fix image recognition and object detcion app starter dependecies

 * Add metadata for Tensorflow apps
This commit is contained in:
Christian Tzolov
2020-06-11 17:07:44 +02:00
committed by Soby Chacko
parent 052f86cf91
commit ebe05bc315
100 changed files with 11343 additions and 15 deletions

View File

@@ -0,0 +1,114 @@
/*
* Copyright 2020-2020 the original author or authors.
*
* 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
*
* https://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.
*/
package org.springframework.cloud.stream.app.processor.image.recognition;
import java.io.FileOutputStream;
import java.io.IOException;
import java.util.List;
import java.util.function.Function;
import org.apache.commons.io.IOUtils;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.boot.context.properties.EnableConfigurationProperties;
import org.springframework.cloud.fn.common.tensorflow.deprecated.JsonMapperFunction;
import org.springframework.cloud.fn.image.recognition.ImageRecognition;
import org.springframework.cloud.fn.image.recognition.ImageRecognitionAugmenter;
import org.springframework.cloud.fn.image.recognition.RecognitionResponse;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.integration.support.MessageBuilder;
import org.springframework.messaging.Message;
/**
* @author Christian Tzolov
*/
@Configuration
@EnableConfigurationProperties(ImageRecognitionProcessorProperties.class)
public class ImageRecognitionProcessorConfiguration {
private static final Log logger = LogFactory.getLog(ImageRecognitionProcessorConfiguration.class);
/**
* Name of the Message header containing the JSON encoded recognition response.
*/
public static final String RECOGNIZED_OBJECTS_HEADER = "recognized_objects";
@Bean
public Function<Message<byte[]>, Message<byte[]>> imageRecognitionFunction(ImageRecognitionProcessorProperties properties) {
return input -> {
// You can use file:, http: or classpath: to provide the path to the input image.
byte[] inputImage = input.getPayload();
try (ImageRecognition imageRecognition = createImageRecognitionFunction(properties)) {
List<RecognitionResponse> recognizedObjects =
ImageRecognition.toRecognitionResponse(imageRecognition.recognizeTopK(inputImage));
// Draw the predicted labels on top of the input image.
byte[] augmentedImage = new ImageRecognitionAugmenter().apply(inputImage, recognizedObjects);
String jsonRecognizedObjects = new JsonMapperFunction().apply(recognizedObjects);
Message<byte[]> outMessage = MessageBuilder
.withPayload(augmentedImage)
.setHeader(RECOGNIZED_OBJECTS_HEADER, jsonRecognizedObjects)
.build();
if (properties.isDebugOutput()) {
try {
logger.info("recognized objects = " + jsonRecognizedObjects);
IOUtils.write(augmentedImage, new FileOutputStream(properties.getDebugOutputPath()));
}
catch (IOException e) {
logger.warn("Cloud not produce debug output", e);
}
}
return outMessage;
}
};
}
private static ImageRecognition createImageRecognitionFunction(ImageRecognitionProcessorProperties properties) {
switch (properties.getModelType()) {
case inception:
return ImageRecognition.inception(
properties.getModel(),
properties.getNormalizedImageSize(),
properties.getResponseSize(),
properties.isCacheModel());
case mobilenetv1:
return ImageRecognition.mobileNetV1(
properties.getModel(),
properties.getNormalizedImageSize(),
properties.getResponseSize(),
properties.isCacheModel());
case mobilenetv2:
return ImageRecognition.mobileNetV2(
properties.getModel(),
properties.getNormalizedImageSize(),
properties.getResponseSize(),
properties.isCacheModel());
default:
throw new RuntimeException("Not supported Model Type: " + properties.getModelType());
}
}
}

View File

@@ -0,0 +1,126 @@
/*
* Copyright 2020-2020 the original author or authors.
*
* 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
*
* https://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.
*/
package org.springframework.cloud.stream.app.processor.image.recognition;
import org.springframework.boot.context.properties.ConfigurationProperties;
import org.springframework.validation.annotation.Validated;
/**
* @author Christian Tzolov
*/
@ConfigurationProperties("image.recognition")
@Validated
public class ImageRecognitionProcessorProperties {
enum ModelType {
inception,
mobilenetv1,
mobilenetv2
}
/**
* Supports three different pre-trained tensorflow image recognition models: Inception, MobileNetV1 and MobileNetV2
*
* 1. Inception graph uses "input" as input and "output" as output.
* 2. MobileNetV2 pre-trained models: https://github.com/tensorflow/models/tree/master/research/slim/nets/mobilenet#pretrained-models
* - normalized image size is always square (e.g. H=W)
* - graph uses "input" as input and "MobilenetV2/Predictions/Reshape_1" as output.
* 3. MobileNetV1 pre-trained models: https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet_v1.md#pre-trained-models
* - graph uses "input" as input and "MobilenetV1/Predictions/Reshape_1" as output.
*/
private ModelType modelType = ModelType.mobilenetv2;
/**
* pre-trained tensorflow image recognition model. Note that the model must match the selected model type!
*/
private String model = "https://storage.googleapis.com/mobilenet_v2/checkpoints/mobilenet_v2_1.4_224.tgz#mobilenet_v2_1.4_224_frozen.pb";
/**
* cache the pre-trained tensorflow model.
*/
private boolean cacheModel = true;
/**
* Normalized image size.
*/
private int normalizedImageSize = 224;
/**
* number of recognized images.
*/
private int responseSize = 5;
private boolean debugOutput = false;
private String debugOutputPath = "image-recognition-result.png";
public ModelType getModelType() {
return modelType;
}
public void setModelType(ModelType modelType) {
this.modelType = modelType;
}
public String getModel() {
return model;
}
public void setModel(String model) {
this.model = model;
}
public boolean isCacheModel() {
return cacheModel;
}
public void setCacheModel(boolean cacheModel) {
this.cacheModel = cacheModel;
}
public int getNormalizedImageSize() {
return normalizedImageSize;
}
public void setNormalizedImageSize(int normalizedImageSize) {
this.normalizedImageSize = normalizedImageSize;
}
public int getResponseSize() {
return responseSize;
}
public void setResponseSize(int responseSize) {
this.responseSize = responseSize;
}
public boolean isDebugOutput() {
return debugOutput;
}
public void setDebugOutput(boolean debugOutput) {
this.debugOutput = debugOutput;
}
public String getDebugOutputPath() {
return debugOutputPath;
}
public void setDebugOutputPath(String debugOutputPath) {
this.debugOutputPath = debugOutputPath;
}
}

View File

@@ -0,0 +1,2 @@
configuration-properties.classes=\
org.springframework.cloud.stream.app.processor.image.recognition.ImageRecognitionProcessorProperties

View File

@@ -0,0 +1,11 @@
image:
recognition:
model: https://storage.googleapis.com/mobilenet_v2/checkpoints/mobilenet_v2_1.4_224.tgz#mobilenet_v2_1.4_224_frozen.pb
modelType: mobilenetv2
responseSize: 3
normalizedImageSize: 224
cacheModel: true
spring:
cloud:
function:
definition: imageRecognitionFunction

View File

@@ -0,0 +1,118 @@
/*
* Copyright 2020-2020 the original author or authors.
*
* 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
*
* https://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.
*/
package org.springframework.cloud.stream.app.processor.image.recognition;
import java.io.IOException;
import org.junit.jupiter.api.Test;
import org.springframework.boot.WebApplicationType;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.builder.SpringApplicationBuilder;
import org.springframework.cloud.fn.common.tensorflow.deprecated.GraphicsUtils;
import org.springframework.cloud.stream.binder.test.InputDestination;
import org.springframework.cloud.stream.binder.test.OutputDestination;
import org.springframework.cloud.stream.binder.test.TestChannelBinderConfiguration;
import org.springframework.context.ConfigurableApplicationContext;
import org.springframework.context.annotation.Import;
import org.springframework.messaging.Message;
import org.springframework.messaging.support.GenericMessage;
import static org.assertj.core.api.Assertions.assertThat;
/**
* @author Christian Tzolov
*/
public class ImageRecognitionProcessorTests {
@Test
public void testImageRecognitionProcessorMobileNetV2() throws IOException {
try (ConfigurableApplicationContext context = new SpringApplicationBuilder(
TestChannelBinderConfiguration.getCompleteConfiguration(ImageRecognitionProcessorTestApplication.class))
.web(WebApplicationType.NONE)
.run("--spring.cloud.function.definition=imageRecognitionFunction",
"--image.recognition.modelType=mobilenetv2",
"--image.recognition.responseSize=3",
"--image.recognition.debugOutput=true",
"--image.recognition.debugOutputPath=./target/image-recognition-mobilenetv2.png")) {
InputDestination processorInput = context.getBean(InputDestination.class);
OutputDestination processorOutput = context.getBean(OutputDestination.class);
byte[] inputImage = GraphicsUtils.loadAsByteArray("classpath:/images/giant_panda_in_beijing_zoo_1.jpg");
processorInput.send(new GenericMessage<>(inputImage));
Message<byte[]> sourceMessage = processorOutput.receive(10000);
String jsonRecognizedObjects = (String) sourceMessage.getHeaders().get(ImageRecognitionProcessorConfiguration.RECOGNIZED_OBJECTS_HEADER);
assertThat(jsonRecognizedObjects)
.isEqualTo("[{\"label\":\"giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca\",\"probability\":0.962329626083374}," +
"{\"label\":\"badger\",\"probability\":0.006058811210095882}," +
"{\"label\":\"ram, tup\",\"probability\":0.0010668420000001788}]");
}
}
@Test
public void testImageRecognitionProcessorMobileNetV1() throws IOException {
try (ConfigurableApplicationContext context = new SpringApplicationBuilder(
TestChannelBinderConfiguration.getCompleteConfiguration(ImageRecognitionProcessorTestApplication.class))
.web(WebApplicationType.NONE)
.run("--image.recognition.model=https://download.tensorflow.org/models/mobilenet_v1_2018_08_02/mobilenet_v1_1.0_224.tgz#mobilenet_v1_1.0_224_frozen.pb",
"--image.recognition.modelType=mobilenetv1",
"--image.recognition.responseSize=3",
"--image.recognition.debugOutput=true",
"--image.recognition.debugOutputPath=./target/image-recognition-mobilenetv1.png")) {
InputDestination processorInput = context.getBean(InputDestination.class);
OutputDestination processorOutput = context.getBean(OutputDestination.class);
byte[] inputImage = GraphicsUtils.loadAsByteArray("classpath:/images/giant_panda_in_beijing_zoo_1.jpg");
processorInput.send(new GenericMessage<>(inputImage));
Message<byte[]> sourceMessage = processorOutput.receive(10000);
String jsonRecognizedObjects = (String) sourceMessage.getHeaders().get(ImageRecognitionProcessorConfiguration.RECOGNIZED_OBJECTS_HEADER);
assertThat(jsonRecognizedObjects)
.isEqualTo("[{\"label\":\"giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca\",\"probability\":0.984053909778595},{\"label\":\"ram, tup\",\"probability\":0.0019619385711848736},{\"label\":\"Staffordshire bullterrier, Staffordshire bull terrier\",\"probability\":0.0018697341438382864}]");
}
}
@Test
public void testImageRecognitionProcessorInception() throws IOException {
try (ConfigurableApplicationContext context = new SpringApplicationBuilder(
TestChannelBinderConfiguration.getCompleteConfiguration(ImageRecognitionProcessorTestApplication.class))
.web(WebApplicationType.NONE)
.run("--image.recognition.model=https://storage.googleapis.com/scdf-tensorflow-models/image-recognition/tensorflow_inception_graph.pb",
"--image.recognition.modelType=inception",
"--image.recognition.responseSize=3",
"--image.recognition.debugOutput=true",
"--image.recognition.debugOutputPath=./target/image-recognition-inception.png")) {
InputDestination processorInput = context.getBean(InputDestination.class);
OutputDestination processorOutput = context.getBean(OutputDestination.class);
byte[] inputImage = GraphicsUtils.loadAsByteArray("classpath:/images/giant_panda_in_beijing_zoo_1.jpg");
processorInput.send(new GenericMessage<>(inputImage));
Message<byte[]> sourceMessage = processorOutput.receive(10000);
String jsonRecognizedObjects = (String) sourceMessage.getHeaders().get(ImageRecognitionProcessorConfiguration.RECOGNIZED_OBJECTS_HEADER);
assertThat(jsonRecognizedObjects)
.isEqualTo("[{\"label\":\"giant panda\",\"probability\":0.9946685433387756},{\"label\":\"Arctic fox\",\"probability\":0.0036631159018725157},{\"label\":\"ice bear\",\"probability\":3.378273395355791E-4}]");
}
}
@SpringBootApplication
@Import({ ImageRecognitionProcessorConfiguration.class })
public static class ImageRecognitionProcessorTestApplication {
}
}

Binary file not shown.

After

Width:  |  Height:  |  Size: 113 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 11 KiB