Upload 5 files

app.py

@@ -0,0 +1,16 @@
+import pandas as pd 
+import streamlit as st
+import eda
+import prediction
+
+page = st.sidebar.selectbox("choose page: ", ("Home page","Data exploration","Data Prediction"))
+
+if page == "Home page":
+    st.title("ASL detection")
+    st.write("Name: Dicky Gabriel")
+    st.write("Batch: SBY-002")
+    st.write("Objective : Predict ASL image")
+elif page == "Data exploration":
+    eda.run()
+else:
+    prediction.run()

best_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37a2d5fe1e713c696bb59589651fdc7225cf8da24093378107afc1e692f5ddfe
+size 228635520

eda.py

@@ -0,0 +1,17 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+from matplotlib.ticker import MultipleLocator
+
+def run():
+    st.title("ASL image Prediction")
+
+    st.subheader("Analysis Data for ASL")
+
+    st.write("This page made by Dicky Gabriel")
+    st.markdown("---")
+
+if __name__ == "__main__":
+    run()

prediction.py

@@ -0,0 +1,46 @@
+import streamlit as st
+import tensorflow as tf
+import cv2
+import numpy as np
+
+def run():
+    # Load the saved model
+    model = tf.keras.models.load_model("best_model.h5")
+
+    # Define the label names
+    label_names = ['a', 'b', 'c', 'd', 'e', 'del', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'not', 'o',
+            'p', 'q', 'r', 's', 't', 'space', 'u', 'v', 'w', 'x', 'y', 'z']
+
+    # Define the Streamlit app
+    st.title("ASL image Prediction")
+    st.write("Choose an image to classify.")
+
+    # Allow the user to select an image file
+    uploaded_file = st.file_uploader("Choose an image file", type=["jpg", "jpeg", "png"])
+
+    if uploaded_file is not None:
+        # Load the image using TensorFlow
+        img = tf.keras.utils.load_img(uploaded_file, target_size=(150, 150, 3))
+
+        # Convert the PIL.Image.Image object to a NumPy array
+        x = tf.keras.utils.img_to_array(img)
+
+        # Expand the array to add a batch dimension
+        x = np.expand_dims(x, axis=0)
+
+        # Normalize the image data
+        x = x / 255.0
+
+        # Make the prediction using the loaded model
+        y_pred = model.predict(x)
+
+        # Get the index of the predicted class with the highest probability
+        class_idx = np.argmax(y_pred, axis=1)[0]
+
+        # Display the predicted class label and image to the user
+        st.write(f"Detection for uploaded image: {label_names[class_idx]}")
+        st.image(img, caption=f"{label_names[class_idx]}", use_column_width=True)
+
+
+if __name__=="__main__":
+    run()

requirements.txt

@@ -0,0 +1,4 @@
+streamlit==1.19.0
+numpy==1.23.5
+tensorflow==2.12.0
+opencv-python==4.5.4.58