Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,131 +1,131 @@
-import streamlit as st
-import pickle
-import numpy as np
-from PIL import Image
-import matplotlib.pyplot as plt
-
-# -----------------------------
-# Load trained Image Classifier
-# -----------------------------
-with open("image_model.pkl", "rb") as f:
-    model = pickle.load(f)
-
-# -----------------------------
-# Streamlit Page Config
-# -----------------------------
-st.set_page_config(page_title="Image Classification App", layout="wide")
-
-# Sidebar Navigation
-st.sidebar.title("Navigation")
-menu = [
-    "Introduction",
-    "Problem Statement",
-    "Data Collection",
-    "Insights",
-    "Solution Development",
-    "Try It Out"
-]
-choice = st.sidebar.radio("Go to", menu)
-
-# -----------------------------
-# Page 1: Introduction
-# -----------------------------
-if choice == "Introduction":
-    st.title("Image Classification Project")
-    st.write("""
-    In many industries today, large volumes of images are generated every day.  
-    Manually checking and sorting them is **time-consuming and error-prone**.  
-
-    I built this project to explore how **automation can speed up image-based tasks**,  
-    reduce dependency on human effort, and improve decision-making.
-    """)
-
-# -----------------------------
-# Page 2: Problem Statement
-# -----------------------------
-elif choice == "Problem Statement":
-    st.title("Problem Statement")
-    st.write("""
-    Organisations face challenges like:
-
-    - Handling **huge sets of unlabelled images**
-    - **Human errors** during manual classification
-    - **Delays in decision-making** due to slow processing
-
-    A system that can **instantly classify images into the correct categories**  
-    can save time, improve accuracy, and reduce operational costs.
-    """)
-
-# -----------------------------
-# Page 3: Data Collection
-# -----------------------------
-elif choice == "Data Collection":
-    st.title("Data Collection")
-    st.write("""
-    - Collected images from multiple categories (labelled data)
-    - Ensured a **balanced set of examples** from each category
-    - Standardised image size and format for consistency
-    - Split into **training and testing sets** to build and evaluate the system
-
-    This helped create a reliable base for the system to learn from.
-    """)
-
-# -----------------------------
-# Page 4: Insights
-# -----------------------------
-elif choice == "Insights":
-    st.title("Key Insights from Data")
-    st.write("""
-    During initial exploration of the dataset:
-
-    - Found clear **visual differences between categories**
-    - Verified **balanced distribution** of images across classes
-    - Confirmed **image sizes and quality were consistent**
-
-    These insights gave confidence that the data was suitable for automation.
-    """)
-
-    st.subheader("Example Images (Randomised)")
-    fig, ax = plt.subplots(1, 3, figsize=(9, 3))
-    for i in range(3):
-        ax[i].imshow(np.random.rand(64, 64, 3))
-        ax[i].set_title(f"Class {i}")
-        ax[i].axis("off")
-    st.pyplot(fig)
-
-# -----------------------------
-# Page 5: Solution Development
-# -----------------------------
-elif choice == "Solution Development":
-    st.title("Solution Development")
-    st.write("""
-    We designed and trained an image classification system that can:
-
-    - Analyse image patterns  
-    - Learn differences between categories  
-    - Predict the category of unseen images  
-
-    The system was tested on new data to validate its performance  
-    before making it available as a user-friendly web app.
-    """)
-
-# -----------------------------
-# Page 6: Try It Out
-# -----------------------------
-elif choice == "Try It Out":
-    st.title("Try the Classifier")
-    st.write("Upload an image below to see how the system classifies it:")
-
-    uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
-
-    if uploaded_file is not None:
-        image = Image.open(uploaded_file).convert("RGB")
-        st.image(image, caption="Uploaded Image", use_column_width=True)
-
-        # Preprocess image (resize -> flatten -> scale handled by the trained pipeline)
-        img_resized = image.resize((64, 64))
-        img_array = np.array(img_resized).flatten().reshape(1, -1)
-
-        # Predict
-        prediction = model.predict(img_array)[0]
-        st.success(f"Predicted Category: **{prediction}**")
+import streamlit as st
+import pickle
+import numpy as np
+from PIL import Image
+import matplotlib.pyplot as plt
+
+# -----------------------------
+# Load trained Image Classifier
+# -----------------------------
+with open("/app/src/image_model.pkl", "rb") as f:
+    model = pickle.load(f)
+
+# -----------------------------
+# Streamlit Page Config
+# -----------------------------
+st.set_page_config(page_title="Image Classification App", layout="wide")
+
+# Sidebar Navigation
+st.sidebar.title("Navigation")
+menu = [
+    "Introduction",
+    "Problem Statement",
+    "Data Collection",
+    "Insights",
+    "Solution Development",
+    "Try It Out"
+]
+choice = st.sidebar.radio("Go to", menu)
+
+# -----------------------------
+# Page 1: Introduction
+# -----------------------------
+if choice == "Introduction":
+    st.title("Image Classification Project")
+    st.write("""
+    In many industries today, large volumes of images are generated every day.  
+    Manually checking and sorting them is **time-consuming and error-prone**.  
+
+    I built this project to explore how **automation can speed up image-based tasks**,  
+    reduce dependency on human effort, and improve decision-making.
+    """)
+
+# -----------------------------
+# Page 2: Problem Statement
+# -----------------------------
+elif choice == "Problem Statement":
+    st.title("Problem Statement")
+    st.write("""
+    Organisations face challenges like:
+
+    - Handling **huge sets of unlabelled images**
+    - **Human errors** during manual classification
+    - **Delays in decision-making** due to slow processing
+
+    A system that can **instantly classify images into the correct categories**  
+    can save time, improve accuracy, and reduce operational costs.
+    """)
+
+# -----------------------------
+# Page 3: Data Collection
+# -----------------------------
+elif choice == "Data Collection":
+    st.title("Data Collection")
+    st.write("""
+    - Collected images from multiple categories (labelled data)
+    - Ensured a **balanced set of examples** from each category
+    - Standardised image size and format for consistency
+    - Split into **training and testing sets** to build and evaluate the system
+
+    This helped create a reliable base for the system to learn from.
+    """)
+
+# -----------------------------
+# Page 4: Insights
+# -----------------------------
+elif choice == "Insights":
+    st.title("Key Insights from Data")
+    st.write("""
+    During initial exploration of the dataset:
+
+    - Found clear **visual differences between categories**
+    - Verified **balanced distribution** of images across classes
+    - Confirmed **image sizes and quality were consistent**
+
+    These insights gave confidence that the data was suitable for automation.
+    """)
+
+    st.subheader("Example Images (Randomised)")
+    fig, ax = plt.subplots(1, 3, figsize=(9, 3))
+    for i in range(3):
+        ax[i].imshow(np.random.rand(64, 64, 3))
+        ax[i].set_title(f"Class {i}")
+        ax[i].axis("off")
+    st.pyplot(fig)
+
+# -----------------------------
+# Page 5: Solution Development
+# -----------------------------
+elif choice == "Solution Development":
+    st.title("Solution Development")
+    st.write("""
+    We designed and trained an image classification system that can:
+
+    - Analyse image patterns  
+    - Learn differences between categories  
+    - Predict the category of unseen images  
+
+    The system was tested on new data to validate its performance  
+    before making it available as a user-friendly web app.
+    """)
+
+# -----------------------------
+# Page 6: Try It Out
+# -----------------------------
+elif choice == "Try It Out":
+    st.title("Try the Classifier")
+    st.write("Upload an image below to see how the system classifies it:")
+
+    uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
+
+    if uploaded_file is not None:
+        image = Image.open(uploaded_file).convert("RGB")
+        st.image(image, caption="Uploaded Image", use_column_width=True)
+
+        # Preprocess image (resize -> flatten -> scale handled by the trained pipeline)
+        img_resized = image.resize((64, 64))
+        img_array = np.array(img_resized).flatten().reshape(1, -1)
+
+        # Predict
+        prediction = model.predict(img_array)[0]
+        st.success(f"Predicted Category: **{prediction}**")