Final working Streamlit app for deployment

app/app.py

@@ -1,45 +1,47 @@
 from __future__ import annotations
+import json
+import sys
+from pathlib import Path
+from typing import Dict
+import pandas as pd
 import streamlit as st
+
+# -------------------------------------------------------------------------
+# Page Configuration
+# -------------------------------------------------------------------------
+st.set_page_config(
+    page_title="Twitter Sentiment Intelligence",
+    page_icon="💼",
+    layout="wide",
+)
+
 st.title("Twitter Sentiment Intelligence")
+st.caption("Streamlit front-end for the Deloitte-ready Twitter Sentiment Intelligence dashboard.")
+
 try:
-    # Main app logic
-    """Streamlit front-end for the Deloitte-ready Twitter Sentiment Intelligence dashboard."""
-    import json
-    import sys
-    from pathlib import Path
-    from typing import Dict
-    import pandas as pd
     # -------------------------------------------------------------------------
-    # Path setup to include the local src/ package for imports
+    # Path setup
     # -------------------------------------------------------------------------
     ROOT = Path(__file__).resolve().parents[1]
     SRC_PATH = ROOT / "src"
     if str(SRC_PATH) not in sys.path:
         sys.path.insert(0, str(SRC_PATH))
+
     from twitter_sentiment.config import load_config
     from twitter_sentiment.predictor import load_artifacts, predict_with_threshold
+
     # -------------------------------------------------------------------------
-    # Streamlit App Configuration
-    # -------------------------------------------------------------------------
-    st.set_page_config(
-        page_title="Twitter Sentiment Intelligence",
-        page_icon="💼",
-        layout="wide",
-    )
-    # -------------------------------------------------------------------------
-    # Cached resource loading (config, pipeline, metrics)
-    # NOTE: artifacts/sentiment_pipeline.joblib is referenced relatively
-    # If artifacts directory or file does not exist, run: python scripts/train.py
+    # Cached dependencies
     # -------------------------------------------------------------------------
     @st.cache_resource(show_spinner=False)
     def _load_dependencies():
         """Load configuration, trained pipeline, and metrics from artifacts."""
         config = load_config()
-        # The load_artifacts function references 'artifacts/sentiment_pipeline.joblib' relatively
         pipeline, metrics = load_artifacts(config)
         return config, pipeline, metrics
+
     # -------------------------------------------------------------------------
-    # Helper function to format prediction probabilities
+    # Format probabilities helper
     # -------------------------------------------------------------------------
     def format_probabilities(probabilities: Dict[str, float]) -> pd.DataFrame:
         """Convert prediction probabilities to a styled DataFrame for display."""
@@ -48,15 +50,69 @@ try:
             .sort_values("confidence", ascending=False)
             .style.format({"confidence": "{:.2%}"})
         )
+
     # -------------------------------------------------------------------------
     # Main Streamlit Application
     # -------------------------------------------------------------------------
     def main() -> None:
         """Render the Deloitte-ready Twitter Sentiment Intelligence Dashboard."""
         config, pipeline, metrics = _load_dependencies()
-        # Main application logic continues here
-        pass
+
+        # ---------------------- Sidebar ----------------------
+        with st.sidebar:
+            st.header("📊 Model Snapshot")
+            st.write("**Classes:**", ", ".join(pipeline.classes_))
+            if metrics:
+                st.metric("Macro F1", f"{metrics.get('f1_macro', 0.0):.2f}")
+                st.metric("Accuracy", f"{metrics.get('accuracy', 0.0):.2f}")
+            else:
+                st.info("Run `python scripts/train.py` to generate metrics.")
+            st.download_button(
+                label="⬇️ Download Metrics JSON",
+                data=json.dumps(metrics or {}, indent=2).encode("utf-8"),
+                file_name="metrics.json",
+                mime="application/json",
+            )
+            st.info(
+                "🚀 Tip: integrate Oracle Autonomous Database by updating `config/settings.yaml`."
+            )
+
+        # ---------------------- Tabs ----------------------
+        tab_predict, tab_metrics = st.tabs(["🔮 Predict", "⚙️ Model Governance"])
+
+        # ---------------------- Prediction Tab ----------------------
+        with tab_predict:
+            st.subheader("Real-Time Sentiment Assessment")
+            user_input = st.text_area("Enter a tweet or customer comment:", height=150)
+            if st.button("Run Analysis", type="primary"):
+                if not user_input.strip():
+                    st.warning("⚠️ Please enter text to analyse.")
+                else:
+                    label, probabilities = predict_with_threshold(user_input, config)
+                    st.success(f"Predicted Sentiment: **{label.title()}**")
+                    st.dataframe(format_probabilities(probabilities), use_container_width=True)
+
+        # ---------------------- Metrics Tab ----------------------
+        with tab_metrics:
+            st.subheader("Operational Metrics")
+            if metrics:
+                metrics_df = (
+                    pd.DataFrame(metrics, index=["score"])
+                    .T.rename(columns={"score": "value"})
+                )
+                st.dataframe(metrics_df, use_container_width=True)
+            else:
+                st.info("Metrics will appear after the first training run (see `scripts/train.py`).")
+
+        # ---------------------- Footer ----------------------
+        st.markdown("---")
+        st.caption("© 2025 Deloitte-aligned Sentiment Analytics Accelerator")
+
+    # -------------------------------------------------------------------------
+    # Entry Point
+    # -------------------------------------------------------------------------
     if __name__ == "__main__":
         main()
+
 except Exception as e:
     st.error(f"Startup failed: {e}")

artifacts/metrics.json

@@ -0,0 +1,11 @@
+{
+  "accuracy": 0.0,
+  "f1_macro": 0.0,
+  "precision_macro": 0.0,
+  "recall_macro": 0.0,
+  "cv_mean": 0.16666666666666666,
+  "cv_std": 0.0,
+  "negative_f1": 0.0,
+  "neutral_f1": 0.0,
+  "positive_f1": 0.0
+}

config/settings.yaml

@@ -1,59 +1,19 @@
-{
-  "project": {
-    "name": "Twitter Sentiment Analysis",
-    "description": "End-to-end sentiment analysis workflow aligned with Deloitte India Oracle Analyst capabilities.",
-    "author": "Your Name"
-  },
-  "live_app": {
-    "streamlit_url": "https://<your-app>.streamlit.app",
-    "vercel_redirect_domain": "https://<your-project>.vercel.app"
-  },
-  "data": {
-    "path": "data/twitter.csv",
-    "text_column": "text",
-    "target_column": "sentiment",
-    "class_order": ["negative", "neutral", "positive"]
-  },
-  "preprocessing": {
-    "lowercase": true,
-    "strip_urls": true,
-    "strip_mentions": true,
-    "strip_hashtags": false,
-    "remove_punctuation": true,
-    "normalize_whitespace": true,
-    "stemming": false,
-    "lemmatize": false
-  },
-  "model": {
-    "artifact_dir": "artifacts",
-    "pipeline_filename": "sentiment_pipeline.joblib",
-    "vectorizer_filename": "tfidf_vectorizer.joblib",
-    "model_filename": "logistic_model.joblib",
-    "test_size": 0.2,
-    "random_state": 42,
-    "max_features": 5000,
-    "ngram_range": [1, 2],
-    "class_weight": "balanced"
-  },
-  "training": {
-    "cv_folds": 5,
-    "scoring": "f1_macro",
-    "n_jobs": -1,
-    "verbose": 1,
-    "probability_thresholds": {
-      "positive": 0.55,
-      "negative": 0.45
-    }
-  },
-  "monitoring": {
-    "enable_model_drift_checks": true,
-    "psi_threshold": 0.2
-  },
-  "oracle_integration": {
-    "enabled": false,
-    "wallet_location": "~/.oci/wallet",
-    "user": "",
-    "dsn": "",
-    "sql_query": "SELECT text, sentiment FROM twitter_training_data"
-  }
-}
+data:
+  train_path: data/twitter.csv
+  text_column: text
+  target_column: sentiment
+  class_order: [negative, neutral, positive]
+
+model:
+  artifact_dir: artifacts
+  pipeline_filename: sentiment_pipeline.joblib
+  test_size: 0.2
+  random_state: 42
+  max_features: 5000
+  ngram_range: [1, 2]
+  class_weight: balanced
+
+training:
+  cv_folds: 5
+  scoring: f1_macro
+

data/datatraining_data.csv

@@ -0,0 +1,11 @@
+text,sentiment
+I love this product,positive
+This is the worst experience ever,negative
+The service was okay,neutral
+Weather is so nice today,positive
+I hate delays,negative
+The meeting went fine,neutral
+Excellent work team!,positive
+I'm feeling bad,negative
+Just a normal day,neutral
+Great job everyone!,positive

data/twitter.csv

@@ -1,7 +1,151 @@
 text,sentiment
-"I love flying with this airline!",positive
-"This is the worst service ever",negative
-"I had a neutral experience",neutral
-"The flight was delayed and staff were rude",negative
-"Amazing in-flight entertainment and service",positive
-"Nothing special, just an average flight",neutral
+I love flying with this airline!,positive
+Amazing service and friendly staff!,positive
+Super smooth check-in process!,positive
+"The new update works perfectly, great job!",positive
+Had an excellent experience today!,positive
+Fast delivery and awesome packaging!,positive
+I’m so happy with my new phone!,positive
+Totally loved the new design!,positive
+Everything was handled professionally.,positive
+This app is getting better every update!,positive
+"The service was okay, nothing special.",neutral
+"I used the app today, it was fine.",neutral
+The product arrived on time.,neutral
+It works as expected.,neutral
+"No complaints, just a normal experience.",neutral
+The new feature is alright.,neutral
+It’s an average performance overall.,neutral
+"Neither good nor bad, just fine.",neutral
+"Nothing remarkable, just standard.",neutral
+"It’s okay, could be improved.",neutral
+This is the worst service ever.,negative
+The staff were rude and unhelpful.,negative
+"The app keeps crashing, very frustrating.",negative
+My package arrived late and damaged.,negative
+Customer support didn’t respond at all.,negative
+"Horrible experience, will never come back.",negative
+This company doesn’t care about customers.,negative
+Terrible product quality.,negative
+The flight was delayed and chaotic.,negative
+Completely disappointed with the outcome.,negative
+I love flying with this airline!,positive
+Amazing service and friendly staff!,positive
+Super smooth check-in process!,positive
+"The new update works perfectly, great job!",positive
+Had an excellent experience today!,positive
+Fast delivery and awesome packaging!,positive
+I’m so happy with my new phone!,positive
+Totally loved the new design!,positive
+Everything was handled professionally.,positive
+This app is getting better every update!,positive
+"The service was okay, nothing special.",neutral
+"I used the app today, it was fine.",neutral
+The product arrived on time.,neutral
+It works as expected.,neutral
+"No complaints, just a normal experience.",neutral
+The new feature is alright.,neutral
+It’s an average performance overall.,neutral
+"Neither good nor bad, just fine.",neutral
+"Nothing remarkable, just standard.",neutral
+"It’s okay, could be improved.",neutral
+This is the worst service ever.,negative
+The staff were rude and unhelpful.,negative
+"The app keeps crashing, very frustrating.",negative
+My package arrived late and damaged.,negative
+Customer support didn’t respond at all.,negative
+"Horrible experience, will never come back.",negative
+This company doesn’t care about customers.,negative
+Terrible product quality.,negative
+The flight was delayed and chaotic.,negative
+Completely disappointed with the outcome.,negative
+I love flying with this airline!,positive
+Amazing service and friendly staff!,positive
+Super smooth check-in process!,positive
+"The new update works perfectly, great job!",positive
+Had an excellent experience today!,positive
+Fast delivery and awesome packaging!,positive
+I’m so happy with my new phone!,positive
+Totally loved the new design!,positive
+Everything was handled professionally.,positive
+This app is getting better every update!,positive
+"The service was okay, nothing special.",neutral
+"I used the app today, it was fine.",neutral
+The product arrived on time.,neutral
+It works as expected.,neutral
+"No complaints, just a normal experience.",neutral
+The new feature is alright.,neutral
+It’s an average performance overall.,neutral
+"Neither good nor bad, just fine.",neutral
+"Nothing remarkable, just standard.",neutral
+"It’s okay, could be improved.",neutral
+This is the worst service ever.,negative
+The staff were rude and unhelpful.,negative
+"The app keeps crashing, very frustrating.",negative
+My package arrived late and damaged.,negative
+Customer support didn’t respond at all.,negative
+"Horrible experience, will never come back.",negative
+This company doesn’t care about customers.,negative
+Terrible product quality.,negative
+The flight was delayed and chaotic.,negative
+Completely disappointed with the outcome.,negative
+I love flying with this airline!,positive
+Amazing service and friendly staff!,positive
+Super smooth check-in process!,positive
+"The new update works perfectly, great job!",positive
+Had an excellent experience today!,positive
+Fast delivery and awesome packaging!,positive
+I’m so happy with my new phone!,positive
+Totally loved the new design!,positive
+Everything was handled professionally.,positive
+This app is getting better every update!,positive
+"The service was okay, nothing special.",neutral
+"I used the app today, it was fine.",neutral
+The product arrived on time.,neutral
+It works as expected.,neutral
+"No complaints, just a normal experience.",neutral
+The new feature is alright.,neutral
+It’s an average performance overall.,neutral
+"Neither good nor bad, just fine.",neutral
+"Nothing remarkable, just standard.",neutral
+"It’s okay, could be improved.",neutral
+This is the worst service ever.,negative
+The staff were rude and unhelpful.,negative
+"The app keeps crashing, very frustrating.",negative
+My package arrived late and damaged.,negative
+Customer support didn’t respond at all.,negative
+"Horrible experience, will never come back.",negative
+This company doesn’t care about customers.,negative
+Terrible product quality.,negative
+The flight was delayed and chaotic.,negative
+Completely disappointed with the outcome.,negative
+I love flying with this airline!,positive
+Amazing service and friendly staff!,positive
+Super smooth check-in process!,positive
+"The new update works perfectly, great job!",positive
+Had an excellent experience today!,positive
+Fast delivery and awesome packaging!,positive
+I’m so happy with my new phone!,positive
+Totally loved the new design!,positive
+Everything was handled professionally.,positive
+This app is getting better every update!,positive
+"The service was okay, nothing special.",neutral
+"I used the app today, it was fine.",neutral
+The product arrived on time.,neutral
+It works as expected.,neutral
+"No complaints, just a normal experience.",neutral
+The new feature is alright.,neutral
+It’s an average performance overall.,neutral
+"Neither good nor bad, just fine.",neutral
+"Nothing remarkable, just standard.",neutral
+"It’s okay, could be improved.",neutral
+This is the worst service ever.,negative
+The staff were rude and unhelpful.,negative
+"The app keeps crashing, very frustrating.",negative
+My package arrived late and damaged.,negative
+Customer support didn’t respond at all.,negative
+"Horrible experience, will never come back.",negative
+This company doesn’t care about customers.,negative
+Terrible product quality.,negative
+The flight was delayed and chaotic.,negative
+Completely disappointed with the outcome.,negative

scripts/create_dataset.py

@@ -0,0 +1,42 @@
+import pandas as pd, os
+
+os.makedirs("data", exist_ok=True)
+
+data = [
+    ("I love flying with this airline!", "positive"),
+    ("Amazing service and friendly staff!", "positive"),
+    ("Super smooth check-in process!", "positive"),
+    ("The new update works perfectly, great job!", "positive"),
+    ("Had an excellent experience today!", "positive"),
+    ("Fast delivery and awesome packaging!", "positive"),
+    ("I’m so happy with my new phone!", "positive"),
+    ("Totally loved the new design!", "positive"),
+    ("Everything was handled professionally.", "positive"),
+    ("This app is getting better every update!", "positive"),
+
+    ("The service was okay, nothing special.", "neutral"),
+    ("I used the app today, it was fine.", "neutral"),
+    ("The product arrived on time.", "neutral"),
+    ("It works as expected.", "neutral"),
+    ("No complaints, just a normal experience.", "neutral"),
+    ("The new feature is alright.", "neutral"),
+    ("It’s an average performance overall.", "neutral"),
+    ("Neither good nor bad, just fine.", "neutral"),
+    ("Nothing remarkable, just standard.", "neutral"),
+    ("It’s okay, could be improved.", "neutral"),
+
+    ("This is the worst service ever.", "negative"),
+    ("The staff were rude and unhelpful.", "negative"),
+    ("The app keeps crashing, very frustrating.", "negative"),
+    ("My package arrived late and damaged.", "negative"),
+    ("Customer support didn’t respond at all.", "negative"),
+    ("Horrible experience, will never come back.", "negative"),
+    ("This company doesn’t care about customers.", "negative"),
+    ("Terrible product quality.", "negative"),
+    ("The flight was delayed and chaotic.", "negative"),
+    ("Completely disappointed with the outcome.", "negative"),
+] * 5  # repeat to reach 150 rows
+
+df = pd.DataFrame(data, columns=["text", "sentiment"])
+df.to_csv("data/twitter.csv", index=False)
+print(f"✅ New dataset created with {len(df)} samples at data/twitter.csv")

scripts/create_dataset.txt

@@ -0,0 +1,42 @@
+import pandas as pd, os
+
+os.makedirs("data", exist_ok=True)
+
+data = [
+    ("I love flying with this airline!", "positive"),
+    ("Amazing service and friendly staff!", "positive"),
+    ("Super smooth check-in process!", "positive"),
+    ("The new update works perfectly, great job!", "positive"),
+    ("Had an excellent experience today!", "positive"),
+    ("Fast delivery and awesome packaging!", "positive"),
+    ("I’m so happy with my new phone!", "positive"),
+    ("Totally loved the new design!", "positive"),
+    ("Everything was handled professionally.", "positive"),
+    ("This app is getting better every update!", "positive"),
+
+    ("The service was okay, nothing special.", "neutral"),
+    ("I used the app today, it was fine.", "neutral"),
+    ("The product arrived on time.", "neutral"),
+    ("It works as expected.", "neutral"),
+    ("No complaints, just a normal experience.", "neutral"),
+    ("The new feature is alright.", "neutral"),
+    ("It’s an average performance overall.", "neutral"),
+    ("Neither good nor bad, just fine.", "neutral"),
+    ("Nothing remarkable, just standard.", "neutral"),
+    ("It’s okay, could be improved.", "neutral"),
+
+    ("This is the worst service ever.", "negative"),
+    ("The staff were rude and unhelpful.", "negative"),
+    ("The app keeps crashing, very frustrating.", "negative"),
+    ("My package arrived late and damaged.", "negative"),
+    ("Customer support didn’t respond at all.", "negative"),
+    ("Horrible experience, will never come back.", "negative"),
+    ("This company doesn’t care about customers.", "negative"),
+    ("Terrible product quality.", "negative"),
+    ("The flight was delayed and chaotic.", "negative"),
+    ("Completely disappointed with the outcome.", "negative"),
+] * 5  # repeat to reach 150 rows
+
+df = pd.DataFrame(data, columns=["text", "sentiment"])
+df.to_csv("data/twitter.csv", index=False)
+print(f"✅ New dataset created with {len(df)} samples at data/twitter.csv")

scripts/train.py

@@ -1,50 +1,40 @@
-"""Command line utility for training the sentiment analysis pipeline."""
+"""Configuration loader that supports both YAML and JSON formats."""
 
 from __future__ import annotations
-
-import argparse
 import json
-import sys
 from pathlib import Path
-
-ROOT = Path(__file__).resolve().parents[1]
-SRC = ROOT / "src"
-if str(SRC) not in sys.path:
-    sys.path.insert(0, str(SRC))
-
-from twitter_sentiment.config import load_config
-from twitter_sentiment.modeling import persist_artifacts, train_and_evaluate
-
-
-def parse_args() -> argparse.Namespace:
-    parser = argparse.ArgumentParser(description="Train the Deloitte-aligned Twitter sentiment model")
-    parser.add_argument(
-        "--config",
-        default="config/settings.yaml",
-        help="Path to the YAML configuration file",
-    )
-    parser.add_argument(
-        "--metrics",
-        default="artifacts/metrics.json",
-        help="Optional path to persist evaluation metrics as JSON",
-    )
-    return parser.parse_args()
-
-
-def main() -> None:
-    args = parse_args()
-    config = load_config(args.config)
-
-    pipeline, metrics_summary = train_and_evaluate(config)
-    artifact_path = persist_artifacts(pipeline, config, metrics_summary)
-
-    metrics_path = Path(args.metrics)
-    metrics_path.parent.mkdir(parents=True, exist_ok=True)
-    metrics_path.write_text(json.dumps(metrics_summary, indent=2), encoding="utf-8")
-
-    print(f"Artifacts saved to {artifact_path}")
-    print(f"Metrics saved to {metrics_path}")
-
-
-if __name__ == "__main__":
-    main()
+from typing import Any, Dict
+import yaml
+
+
+class Config(dict):
+    """Wrapper class to access config like attributes."""
+    def __getattr__(self, item):
+        return self.get(item)
+
+
+def load_config(path: str | Path = "config/settings.yaml") -> Config:
+    """
+    Load configuration from YAML or JSON.
+    Automatically detects file type based on extension.
+    """
+    path = Path(path)
+    if not path.exists():
+        raise FileNotFoundError(f"Config file not found: {path}")
+
+    try:
+        if path.suffix.lower() in [".yaml", ".yml"]:
+            with open(path, "r", encoding="utf-8") as f:
+                data = yaml.safe_load(f)
+        elif path.suffix.lower() == ".json":
+            with open(path, "r", encoding="utf-8") as f:
+                data = json.load(f)
+        else:
+            raise ValueError(f"Unsupported config format: {path.suffix}")
+    except Exception as e:
+        raise ValueError(f"Failed to parse config file {path}: {e}")
+
+    if not isinstance(data, dict):
+        raise ValueError(f"Invalid config structure in {path}")
+
+    return Config(data)