Upload 5 files

InceptionResNetV2Skripsi.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ce9b335d23bb9811adffe070040def6a0b8feecff69e345c0d1c16364804bfb
+size 56233680

PythonApplication2.pyproj

@@ -0,0 +1,35 @@
+<Project DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003" ToolsVersion="4.0">
+  <PropertyGroup>
+    <Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
+    <SchemaVersion>2.0</SchemaVersion>
+    <ProjectGuid>cb57c127-7b42-46f1-a142-114ef4703e1f</ProjectGuid>
+    <ProjectHome>.</ProjectHome>
+    <StartupFile>PythonApplication2.py</StartupFile>
+    <SearchPath>
+    </SearchPath>
+    <WorkingDirectory>.</WorkingDirectory>
+    <OutputPath>.</OutputPath>
+    <Name>PythonApplication2</Name>
+    <RootNamespace>PythonApplication2</RootNamespace>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)' == 'Debug' ">
+    <DebugSymbols>true</DebugSymbols>
+    <EnableUnmanagedDebugging>false</EnableUnmanagedDebugging>
+  </PropertyGroup>
+  <PropertyGroup Condition=" '$(Configuration)' == 'Release' ">
+    <DebugSymbols>true</DebugSymbols>
+    <EnableUnmanagedDebugging>false</EnableUnmanagedDebugging>
+  </PropertyGroup>
+  <ItemGroup>
+    <Compile Include="app.py" />
+  </ItemGroup>
+  <Import Project="$(MSBuildExtensionsPath32)\Microsoft\VisualStudio\v$(VisualStudioVersion)\Python Tools\Microsoft.PythonTools.targets" />
+  <!-- Uncomment the CoreCompile target to enable the Build command in
+       Visual Studio and specify your pre- and post-build commands in
+       the BeforeBuild and AfterBuild targets below. -->
+  <!--<Target Name="CoreCompile" />-->
+  <Target Name="BeforeBuild">
+  </Target>
+  <Target Name="AfterBuild">
+  </Target>
+</Project>

PythonApplication2.sln

@@ -0,0 +1,23 @@
+
+Microsoft Visual Studio Solution File, Format Version 12.00
+# Visual Studio Version 17
+VisualStudioVersion = 17.2.32526.322
+MinimumVisualStudioVersion = 10.0.40219.1
+Project("{888888A0-9F3D-457C-B088-3A5042F75D52}") = "PythonApplication2", "PythonApplication2.pyproj", "{CB57C127-7B42-46F1-A142-114EF4703E1F}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Any CPU = Debug|Any CPU
+		Release|Any CPU = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{CB57C127-7B42-46F1-A142-114EF4703E1F}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{CB57C127-7B42-46F1-A142-114EF4703E1F}.Release|Any CPU.ActiveCfg = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+	GlobalSection(ExtensibilityGlobals) = postSolution
+		SolutionGuid = {A738FEDC-1393-4DFF-8634-ADAD6A2203CA}
+	EndGlobalSection
+EndGlobal

app1.py

@@ -0,0 +1,84 @@
+import streamlit as st
+import numpy as np
+import tensorflow as tf
+from tcp_latency import measure_latency
+
+# Measure Latency
+latency = measure_latency(host='35.201.127.49', port=443)
+#35.201.127.49:443
+#192.168.18.6:8501
+
+# Load TensorFlow Lite model
+interpreter = tf.lite.Interpreter(model_path="InceptionResNetV2Skripsi.tflite")
+interpreter.allocate_tensors()
+
+# Get input and output tensors
+input_details = interpreter.get_input_details()
+output_details = interpreter.get_output_details()
+
+# Define a function to resize the input image
+def resize_image(image):
+    # Resize the image to 150x150 pixels
+    resized_image = tf.image.resize(image, [150, 150])
+    return resized_image.numpy()
+
+# Define a function to run inference on the TensorFlow Lite model
+def classify_image(image):
+    # Pre-process the input image
+    resized_image = resize_image(image)
+    input_data = np.expand_dims(resized_image, axis=0).astype(np.float32)
+    interpreter.set_tensor(input_details[0]['index'], input_data)
+
+    # Run inference
+    with st.spinner('Classifying...'):
+        interpreter.invoke()
+
+    # Get the output probabilities
+    output_data = interpreter.get_tensor(output_details[0]['index'])
+    return output_data[0]
+
+# Define the labels for the 7 classes
+labels = ['akiec', 'bcc', 'bkl', 'df', 'mel', 'nv', 'vasc']
+
+from PIL import Image
+# Define the main Streamlit app
+def main():
+    st.title("Skin Cancer Classification")
+
+    st.write("The application allows users to quickly and easily check for signs of skin cancer from the comfort of their own homes. It's important to note that the application is not a substitute for medical advice, but rather a tool to help users identify potential skin issues and seek professional help if necessary. Heres the overview:")
+    st.write("1.	The user is prompted to upload an image of a skin lesion.")
+    st.write("2.	Once the user uploads an image, it is displayed on the screen and the application runs the image through a machine learning model.")
+    st.write("3.	The model outputs the top 3 predictions for the type of skin cancer (out of 7 classes) that the lesion may be, along with the confidence level for each prediction.")
+    st.write("4.	The user is shown the top 3 predictions and their confidence levels.")
+    st.write("5.	If the model's confidence level is less than 70%, the application informs the user that the skin is either healthy or the input image is unclear. It is also recommended that the user consults a dermatologist for any skin concerns.")
+    st.write ("6.	The application also measures the latency or response time for the model to classify the image, and displays it to the user.")
+    st.write("Please note that this model still has room for academic revision as it can only classify the following 7 classes")
+    st.write("- ['akiec'](https://en.wikipedia.org/wiki/Squamous-cell_carcinoma) - squamous cell carcinoma (actinic keratoses dan intraepithelial carcinoma),")
+    st.write("- ['bcc'](https://en.wikipedia.org/wiki/Basal-cell_carcinoma) - basal cell carcinoma,")
+    st.write("- ['bkl'](https://en.wikipedia.org/wiki/Seborrheic_keratosis) - benign keratosis (serborrheic keratosis),")
+    st.write("- ['df'](https://en.wikipedia.org/wiki/Dermatofibroma) - dermatofibroma, ")
+    st.write("- ['nv'](https://en.wikipedia.org/wiki/Melanocytic_nevus) - melanocytic nevus, ")
+    st.write("- ['mel'](https://en.wikipedia.org/wiki/Melanoma) - melanoma,")
+    st.write("- ['vasc'](https://en.wikipedia.org/wiki/Vascular_anomaly) - vascular skin (Cherry Angiomas, Angiokeratomas, Pyogenic Granulomas.)")
+    st.write("Due to imperfection of the model and a room of improvement for the future, if the probabilities shown are less than 70%, the skin is either healthy or the input image is unclear. This means that the model can be the first diagnostic of your skin illness. As precautions for your skin illness, it is better to do consultation with dermatologist. ")
+    st.write(latency[0])
+    
+    # Get the input image from the user
+    image = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
+
+    # Show the input image
+    if image is not None:
+        image = np.array(Image.open(image))
+        st.image(image, width=150)
+
+        # Run inference on the input image
+        probs = classify_image(image)
+
+        # Display the top 3 predictions
+        top_3_indices = np.argsort(probs)[::-1][:3]
+        st.write("Top 3 predictions:")
+        for i in range(3):
+            st.write("%d. %s (%.2f%%)" % (i + 1, labels[top_3_indices[i]], probs[top_3_indices[i]] * 100))
+
+if __name__ == '__main__':
+    main()

requirements.txt

@@ -0,0 +1,4 @@
+streamlit
+tensorflow
+numpy
+tcp_latency