Upload 6 files

README.md

@@ -1,19 +1,16 @@
 ---
-title: Penguins Predictor Inclass
-emoji: 🚀
-colorFrom: red
-colorTo: red
-sdk: docker
-app_port: 8501
-tags:
-- streamlit
+title: Palmer Penguin Species Predictor
+emoji: 🐧
+colorFrom: blue
+colorTo: green
+sdk: streamlit
+app_file: penguin_streamlit_app.py
 pinned: false
-short_description: Streamlit template space
 ---
 
-# Welcome to Streamlit!
+# Palmer Penguin Species Predictor
 
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
+This Streamlit application predicts the species of Palmer Penguins (Adelie, Chinstrap, or Gentoo)
+based on their culmen length, culmen depth, flipper length, and body mass.
 
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+The model was trained on the Palmer Penguins dataset.

label_encoder.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2de28ef9c07a06b12496cf1c58ced82f158c1d246fe8cf755ee73d48cd9b8cae
+size 561

logistic_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6e071b1e6a70faf8f77ca05493b7505d619f1a7974267ecb848e5f0e063d49d
+size 1007

penguin_streamlit_app.py

@@ -0,0 +1,169 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import joblib
+import os
+import requests
+
+# Ensure these classes are available for joblib to unpickle
+from sklearn.preprocessing import StandardScaler, LabelEncoder
+from sklearn.linear_model import LogisticRegression
+
+# --- Configuration ---
+# Define paths to the artifacts
+MODEL_PATH = 'logistic_model.joblib'
+ENCODER_PATH = 'label_encoder.joblib'
+SCALER_PATH = 'scaler.joblib'
+
+# --- Load Artifacts ---
+@st.cache_resource  # Cache loading for performance
+def load_artifacts():
+    try:
+        model = joblib.load(MODEL_PATH)
+        label_encoder = joblib.load(ENCODER_PATH)
+        scaler = joblib.load(SCALER_PATH)
+        return model, label_encoder, scaler
+    except FileNotFoundError as e:
+        st.error(f"Error: One or more artifact files not found. {e}")
+        st.error(f"Expected files: {MODEL_PATH}, {ENCODER_PATH}, {SCALER_PATH}")
+        return None, None, None
+    except Exception as e:
+        st.error(f"An error occurred while loading artifacts: {e}")
+        return None, None, None
+
+# --- Feature Statistics for Input Guidance ---
+# These values are derived from the training data
+feature_stats = {
+    'culmen_length_mm': {'min': 32.1, 'max': 59.6, 'mean': 43.92, 'step': 0.1},
+    'culmen_depth_mm': {'min': 13.1, 'max': 21.5, 'mean': 17.15, 'step': 0.1},
+    'flipper_length_mm': {'min': 172.0, 'max': 231.0, 'mean': 200.92, 'step': 1.0},
+    'body_mass_g': {'min': 2700.0, 'max': 6300.0, 'mean': 4207.06, 'step': 100.0}
+}
+
+# --- Penguin Species Images ---
+species_image_map = {
+    "Adelie": "https://upload.wikimedia.org/wikipedia/commons/e/e3/Hope_Bay-2016-Trinity_Peninsula–Adélie_penguin_%28Pygoscelis_adeliae%29_04.jpg",
+    "Gentoo": "https://upload.wikimedia.org/wikipedia/commons/0/00/Brown_Bluff-2016-Tabarin_Peninsula–Gentoo_penguin_%28Pygoscelis_papua%29_03.jpg",
+    "Chinstrap": "https://upload.wikimedia.org/wikipedia/commons/0/08/South_Shetland-2016-Deception_Island–Chinstrap_penguin_%28Pygoscelis_antarctica%29_04.jpg"
+}
+
+# --- App UI ---
+st.set_page_config(
+    page_title="Palmer Penguin Predictor",
+    page_icon="🐧",
+    layout="wide"
+)
+
+st.title("🐧 Palmer Penguin Species Predictor")
+st.markdown("""
+This app predicts the species of a Palmer Penguin based on its physical measurements.
+Enter the measurements in the sidebar and click 'Predict' to see the results!
+""")
+
+# Load the model and preprocessors
+model, label_encoder, scaler = load_artifacts()
+
+if model is not None and label_encoder is not None and scaler is not None:
+    # Move input controls to sidebar
+    st.sidebar.header("Input Penguin Measurements")
+    
+    # Create input fields for each feature in the sidebar
+    inputs = {}
+    for feature, stats in feature_stats.items():
+        # Create a more user-friendly label
+        label = feature.replace('_', ' ').title()
+        unit = "mm" if "mm" in feature else "g"
+        
+        inputs[feature] = st.sidebar.slider(
+            f"{label} ({unit})",
+            min_value=float(stats['min']),
+            max_value=float(stats['max']),
+            value=float(stats['mean']),
+            step=stats['step'],
+            help=f"Typical range: {stats['min']} - {stats['max']} (Average: {stats['mean']})"
+        )
+    
+    # Create a button to trigger prediction in the sidebar
+    predict_button = st.sidebar.button("🔍 Predict Penguin Species", type="primary")
+    
+    # Main content area
+    if predict_button:
+        # Create a DataFrame from inputs
+        input_df = pd.DataFrame([inputs])
+        
+        # Display the input values
+        st.subheader("Your Input Values:")
+        st.dataframe(input_df.style.format("{:.1f}"))
+        
+        # Scale the input features
+        input_scaled = scaler.transform(input_df)
+        
+        # Make prediction
+        prediction_encoded = model.predict(input_scaled)
+        prediction_proba = model.predict_proba(input_scaled)
+        
+        # Decode the prediction
+        predicted_species = label_encoder.inverse_transform(prediction_encoded)[0]
+        
+        # Display the prediction result
+        st.subheader("Prediction Result:")
+        st.markdown(f"### This penguin is a **{predicted_species}**!")
+        
+        # Display the probabilities
+        st.subheader("Prediction Probabilities:")
+        proba_df = pd.DataFrame(
+            prediction_proba, 
+            columns=label_encoder.classes_
+        )
+        st.dataframe(proba_df.style.format("{:.2%}"))
+        
+        # Display the penguin image using streamlit's image component directly
+        st.subheader(f"{predicted_species} Penguin:")
+        st.image(species_image_map[predicted_species], width=400, caption=f"{predicted_species} Penguin")
+    
+    # Add information about the features
+    with st.expander("About the Measurements"):
+        st.markdown("""
+        ### Penguin Measurements Explained
+        
+        - **Culmen Length**: The length of the penguin's bill (in mm)
+        - **Culmen Depth**: The depth (height) of the penguin's bill (in mm)
+        - **Flipper Length**: The length of the penguin's flipper (in mm)
+        - **Body Mass**: The weight of the penguin (in grams)
+        
+        These measurements are used by researchers to study penguin populations and can also help identify different species.
+        """)
+        
+        # Show a table of the feature statistics
+        st.subheader("Feature Statistics from Training Data:")
+        stats_df = pd.DataFrame(feature_stats).T
+        st.dataframe(stats_df.style.format("{:.1f}"))
+    
+    # Add information about the penguin species
+    with st.expander("About the Penguin Species"):
+        st.markdown("""
+        ### Palmer Penguin Species
+        
+        The Palmer Archipelago in Antarctica is home to three penguin species:
+        
+        - **Adelie**: Smaller penguins with a white ring around the eye
+        - **Chinstrap**: Named for the narrow black band under their head
+        - **Gentoo**: Larger penguins with bright orange-red bills and feet
+        
+        This model was trained on the Palmer Penguins dataset, which contains measurements of these three species.
+        """)
+        
+        # Display all three penguin species images using streamlit's image component
+        species_cols = st.columns(3)
+        for i, (species, url) in enumerate(species_image_map.items()):
+            with species_cols[i]:
+                st.markdown(f"**{species}**")
+                st.image(url, width=200, caption=species)
+    
+    # Footer
+    st.markdown("---")
+    st.markdown("Created with Streamlit • Data from [Palmer Penguins Dataset](https://github.com/allisonhorst/palmerpenguins)")
+    
+else:
+    st.error("Could not load the model or preprocessors. Please check that the model files exist in the correct location.")
+    st.info("Make sure you've run the training script first to generate the model files.")

requirements.txt

@@ -1,3 +1,6 @@
-altair
+streamlit
 pandas
-streamlit
+numpy
+scikit-learn
+requests
+joblib

scaler.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b68dfae5697d6847f2b92c0e0e1a59193e6193a1072dad34dee6a3a63ff84d6d
+size 1095