rapidsegment 1.2.3__tar.gz → 1.2.4__tar.gz

{rapidsegment-1.2.3 → rapidsegment-1.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: rapidsegment
-Version: 1.2.3
+Version: 1.2.4
 Summary: A fast, multi-segment population scoring and combinatorial heuristic rule extraction engine.
 Author: Bishwarup Biswas
 Author-email: Bishwarup Biswas <bishwarup1429@gmail.com>
@@ -38,10 +38,14 @@ Provides-Extra: prettytable
 Provides-Extra: ui
 Description-Content-Type: text/markdown
 
-> [!IMPORTANT]
+> [!IMPORTANT]
 > **Legal Disclaimer**  
 > This open‑source library (`RapidSegment`) is an independent, community‑driven predictive analytics framework. It is **completely unaffiliated** with any commercial products, SaaS platforms, or enterprise solutions of the same or similar name. Any overlap in nomenclature is purely coincidental.
 
+<p align="center">
+  <img width="300" height="500" alt="adfd9cdf-251f-44e4-af79-20802d4a7a01" src="https://github.com/user-attachments/assets/803790d0-7ebe-4e79-a415-65f1daaea40b" />
+</p>
+
 # 🚀 RapidSegment – Strategic Segmentation & Scorecard Engine
 
 [![PyPI version](https://img.shields.io/pypi/v/rapidsegment.svg)](https://pypi.org/project/rapidsegment/)
@@ -56,7 +60,7 @@ Description-Content-Type: text/markdown
 - [✨ Features](#-features)
 - [🌳 Decision Trees vs RapidSegment](#-decision-trees-vs-rapidsegment)
 - [⚡ Quick Start](#-quick-start)
-- [🖥️ Web UI](#-web-ui)
+- [🖥️ Web UI](#️-web-ui)
 - [🧩 Components](#-components)
 - [🏗️ System Architecture](#️-system-architecture)
 - [⚙️ How It Works – Step by Step](#️-how-it-works--step-by-step)
@@ -104,6 +108,7 @@ In short, a decision tree is great when you want a predictive model structure; R
 
 ---
 
+
 ## ⚡ Quick Start
 
 ```python
@@ -555,3 +560,4 @@ Special Thanks to Mr. [Guillermo Navas Palencia](https://github.com/guillermo-na
 
 _Independent, open‑source, and ready for production._
 ```
+

{rapidsegment-1.2.3 → rapidsegment-1.2.4}/README.md

@@ -1,7 +1,11 @@
-> [!IMPORTANT]
+> [!IMPORTANT]
 > **Legal Disclaimer**  
 > This open‑source library (`RapidSegment`) is an independent, community‑driven predictive analytics framework. It is **completely unaffiliated** with any commercial products, SaaS platforms, or enterprise solutions of the same or similar name. Any overlap in nomenclature is purely coincidental.
 
+<p align="center">
+  <img width="300" height="500" alt="adfd9cdf-251f-44e4-af79-20802d4a7a01" src="https://github.com/user-attachments/assets/803790d0-7ebe-4e79-a415-65f1daaea40b" />
+</p>
+
 # 🚀 RapidSegment – Strategic Segmentation & Scorecard Engine
 
 [![PyPI version](https://img.shields.io/pypi/v/rapidsegment.svg)](https://pypi.org/project/rapidsegment/)
@@ -16,7 +20,7 @@
 - [✨ Features](#-features)
 - [🌳 Decision Trees vs RapidSegment](#-decision-trees-vs-rapidsegment)
 - [⚡ Quick Start](#-quick-start)
-- [🖥️ Web UI](#-web-ui)
+- [🖥️ Web UI](#️-web-ui)
 - [🧩 Components](#-components)
 - [🏗️ System Architecture](#️-system-architecture)
 - [⚙️ How It Works – Step by Step](#️-how-it-works--step-by-step)
@@ -64,6 +68,7 @@ In short, a decision tree is great when you want a predictive model structure; R
 
 ---
 
+
 ## ⚡ Quick Start
 
 ```python
@@ -515,3 +520,4 @@ Special Thanks to Mr. [Guillermo Navas Palencia](https://github.com/guillermo-na
 
 _Independent, open‑source, and ready for production._
 ```
+

{rapidsegment-1.2.3 → rapidsegment-1.2.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "rapidsegment"
-version = "1.2.3"
+version = "1.2.4"
 description = "A fast, multi-segment population scoring and combinatorial heuristic rule extraction engine."
 readme = "README.md"
 license = "MIT"

{rapidsegment-1.2.3 → rapidsegment-1.2.4}/pyproject.toml.orig

@@ -1,6 +1,6 @@
 [project]
 name = "rapidsegment"
-version = "1.2.3"
+version = "1.2.4"
 description = "A fast, multi-segment population scoring and combinatorial heuristic rule extraction engine."
 readme = "README.md"
 authors = [

{rapidsegment-1.2.3 → rapidsegment-1.2.4}/src/rapidsegment/ui/_exit.py

@@ -1,124 +1,124 @@
-"""Exit the RapidSegment UI and stop the Streamlit server.
-
-Streamlit has no built-in stop-server button, so when a user clicks Exit we
-terminate the Streamlit server process (and its children). run_ui() records the
-server PID when it launches the app; if the app was launched some other way
-(e.g. `streamlit run app.py` directly) we fall back to walking the process tree
-to find the Streamlit server. psutil is a hard dependency of the package, so the
-process-tree walk is always available.
-"""
-
-import os
-import tempfile
-
-import streamlit as st
-
-# Where run_ui() records the Streamlit server PID (single-user local app).
-_PID_FILE = os.path.join(tempfile.gettempdir(), "rapidsegment_ui_server.pid")
-
-
-def write_server_pid(pid):
-    try:
-        with open(_PID_FILE, "w") as fh:
-            fh.write(str(pid))
-    except Exception:
-        pass
-
-
-def _read_server_pid():
-    try:
-        if os.path.exists(_PID_FILE):
-            with open(_PID_FILE) as fh:
-                return int(fh.read().strip())
-    except Exception:
-        pass
-    return None
-
-
-def _kill_tree(proc):
-    """Kill a psutil process and all of its descendants."""
-    try:
-        for child in proc.children(recursive=True):
-            try:
-                child.kill()
-            except Exception:
-                pass
-        proc.kill()
-        return True
-    except Exception:
-        return False
-
-
-def exit_ui():
-    """Terminate the Streamlit server (and this app run)."""
-    targets = []
-
-    try:
-        import psutil
-
-        pid = _read_server_pid()
-        if pid is not None:
-            try:
-                targets.append(psutil.Process(pid))
-            except Exception:
-                pass
-
-        # Fallback / supplement: walk up the process tree to find the
-        # Streamlit server (its command line contains "streamlit").
-        me = psutil.Process(os.getpid())
-        anc = me
-        for _ in range(8):
-            anc = anc.parent()
-            if anc is None:
-                break
-            cmd = " ".join(anc.cmdline()).lower()
-            if "streamlit" in cmd:
-                targets.append(anc)
-    except Exception:
-        pass
-
-    seen = set()
-    for t in targets:
-        if t.pid in seen:
-            continue
-        seen.add(t.pid)
-        _kill_tree(t)
-
-    if not targets:
-        # Last resort: kill our own process tree.
-        try:
-            import psutil
-            me = psutil.Process(os.getpid())
-            for child in me.children(recursive=True):
-                try:
-                    child.kill()
-                except Exception:
-                    pass
-            me.kill()
-        except Exception:
-            try:
-                os.kill(os.getpid(), 9)
-            except Exception:
-                pass
-
-    # Always halt this script run.
-    try:
-        st.stop()
-    except Exception:
-        pass
-    # If st.stop() did not end it (already torn down), force exit.
-    try:
-        os._exit(0)
-    except Exception:
-        pass
-
-
-def render_exit_button(label="Exit UI", key="exit_ui_global"):
-    if st.sidebar.button(
-        label,
-        key=key,
-        width="content",
-        help="Stop the Streamlit server and close the app.",
-    ):
+"""Exit the RapidSegment UI and stop the Streamlit server.
+
+Streamlit has no built-in stop-server button, so when a user clicks Exit we
+terminate the Streamlit server process (and its children). run_ui() records the
+server PID when it launches the app; if the app was launched some other way
+(e.g. `streamlit run app.py` directly) we fall back to walking the process tree
+to find the Streamlit server. psutil is a hard dependency of the package, so the
+process-tree walk is always available.
+"""
+
+import os
+import tempfile
+
+import streamlit as st
+
+# Where run_ui() records the Streamlit server PID (single-user local app).
+_PID_FILE = os.path.join(tempfile.gettempdir(), "rapidsegment_ui_server.pid")
+
+
+def write_server_pid(pid):
+    try:
+        with open(_PID_FILE, "w") as fh:
+            fh.write(str(pid))
+    except Exception:
+        pass
+
+
+def _read_server_pid():
+    try:
+        if os.path.exists(_PID_FILE):
+            with open(_PID_FILE) as fh:
+                return int(fh.read().strip())
+    except Exception:
+        pass
+    return None
+
+
+def _kill_tree(proc):
+    """Kill a psutil process and all of its descendants."""
+    try:
+        for child in proc.children(recursive=True):
+            try:
+                child.kill()
+            except Exception:
+                pass
+        proc.kill()
+        return True
+    except Exception:
+        return False
+
+
+def exit_ui():
+    """Terminate the Streamlit server (and this app run)."""
+    targets = []
+
+    try:
+        import psutil
+
+        pid = _read_server_pid()
+        if pid is not None:
+            try:
+                targets.append(psutil.Process(pid))
+            except Exception:
+                pass
+
+        # Fallback / supplement: walk up the process tree to find the
+        # Streamlit server (its command line contains "streamlit").
+        me = psutil.Process(os.getpid())
+        anc = me
+        for _ in range(8):
+            anc = anc.parent()
+            if anc is None:
+                break
+            cmd = " ".join(anc.cmdline()).lower()
+            if "streamlit" in cmd:
+                targets.append(anc)
+    except Exception:
+        pass
+
+    seen = set()
+    for t in targets:
+        if t.pid in seen:
+            continue
+        seen.add(t.pid)
+        _kill_tree(t)
+
+    if not targets:
+        # Last resort: kill our own process tree.
+        try:
+            import psutil
+            me = psutil.Process(os.getpid())
+            for child in me.children(recursive=True):
+                try:
+                    child.kill()
+                except Exception:
+                    pass
+            me.kill()
+        except Exception:
+            try:
+                os.kill(os.getpid(), 9)
+            except Exception:
+                pass
+
+    # Always halt this script run.
+    try:
+        st.stop()
+    except Exception:
+        pass
+    # If st.stop() did not end it (already torn down), force exit.
+    try:
+        os._exit(0)
+    except Exception:
+        pass
+
+
+def render_exit_button(label="Exit UI", key="exit_ui_global"):
+    if st.sidebar.button(
+        label,
+        key=key,
+        width="content",
+        help="Stop the Streamlit server and close the app.",
+    ):
         exit_ui()
 

rapidsegment-1.2.3/src/rapidsegment/ui/.rapidsegment_suite/artifacts/exp_20260822_120723_eaf4de/config.json

@@ -1,32 +0,0 @@
-{
-  "experiment_name": "EXP 1 Bank",
-  "description": "",
-  "data_table": "udl_data",
-  "target_col": "Exited__binary",
-  "primary_key": "",
-  "top_n_vars": 15,
-  "max_segments": 10,
-  "max_feature_reuse": 1,
-  "feature_groups": {},
-  "enable_diversity": false,
-  "ignore_features": [
-    "id",
-    "CustomerId",
-    "Surname",
-    "Exited"
-  ],
-  "binning_method": "optimal_cart",
-  "naive_bins": 5,
-  "max_expansion_hops": 0,
-  "enable_1way": true,
-  "enable_2way": true,
-  "enable_3way": true,
-  "selection_metric": "iv",
-  "min_sample_size": 1000,
-  "min_lift": 1.2,
-  "min_events": 100,
-  "param_grid": null,
-  "sort_priority": "rate_lift_count",
-  "n_jobs": -1,
-  "expand_log_mode": "none"
-}

rapidsegment-1.2.3/src/rapidsegment/ui/.rapidsegment_suite/artifacts/exp_20260822_120723_eaf4de/logs.txt

@@ -1,29 +0,0 @@
-2026-08-22 12:07:23 | INFO    | Experiment 'EXP 1 Bank' started — exp_id=exp_20260822_120723_eaf4de
-2026-08-22 12:07:23 | INFO    | Data: 165,034 rows × 15 cols · target=`Exited__binary`
-2026-08-22 12:07:23 | INFO    | 🚀 Starting hierarchical segment extraction...
-2026-08-22 12:07:23 | INFO    | ⚙️ DuckDB Configured for Disk Spilling: Threads=14/16, MemoryLimit=11GB, TempDir=C:\Users\Bishwarup\RapidSegment\rapidsegment\src\rapidsegment\ui\.rapidsegment_suite\artifacts\exp_20260822_120723_eaf4de\tmp
-2026-08-22 12:07:23 | INFO    | 📊 Sort priority: rate_lift_count
-2026-08-22 12:07:23 | INFO    | 📦 Binning method: optimal_cart
-2026-08-22 12:07:24 | INFO    | 🔒 Locking Original Base Rate: 21.16%
-2026-08-22 12:07:24 | INFO    | 🔄 Iteration 1 | Remaining Volume: 165,034 | Base Rate: 21.16%
-2026-08-22 12:07:24 | INFO    | 🔍 Computing IV and bins for 10 features...
-2026-08-22 12:07:28 | INFO    | 📌 Feature Usage Tracker Update -> 'Age' used count = 1
-2026-08-22 12:07:28 | INFO    | 📌 Feature Usage Tracker Update -> 'Geography' used count = 1
-2026-08-22 12:07:28 | INFO    | 📌 Feature Usage Tracker Update -> 'IsActiveMember' used count = 1
-2026-08-22 12:07:28 | INFO    | ✅ Segment 1 Captured (Size Floor: 1000 | Lift Floor: 1.2): rows=3164, events=2773.0, lift=4.14
-  Rule: Age=[48.50, inf) & Geography=[Germany] & IsActiveMember=(-inf, 0.50)
-  SQL: ("Age" >= 48.50) AND ("Geography" IN ('Germany')) AND ("IsActiveMember" < 0.50)
-2026-08-22 12:07:28 | INFO    | 🔄 Iteration 2 | Remaining Volume: 161,870 | Base Rate: 19.86%
-2026-08-22 12:07:28 | INFO    | 🔍 Computing IV and bins for 10 features...
-2026-08-22 12:07:32 | INFO    | 📌 Feature Usage Tracker Update -> 'Balance' used count = 1
-2026-08-22 12:07:32 | INFO    | 📌 Feature Usage Tracker Update -> 'Gender' used count = 1
-2026-08-22 12:07:32 | INFO    | 📌 Feature Usage Tracker Update -> 'NumOfProducts' used count = 1
-2026-08-22 12:07:32 | INFO    | ✅ Segment 2 Captured (Size Floor: 1000 | Lift Floor: 1.2): rows=11930, events=6431.0, lift=2.55
-  Rule: Balance=(-inf, 1893.51) & Gender=[Female] & NumOfProducts=(-inf, 1.50)
-  SQL: ("Balance" < 1893.51) AND ("Gender" IN ('Female')) AND ("NumOfProducts" < 1.50)
-2026-08-22 12:07:32 | INFO    | 🔄 Iteration 3 | Remaining Volume: 149,940 | Base Rate: 17.15%
-2026-08-22 12:07:32 | INFO    | 🔍 Computing IV and bins for 10 features...
-2026-08-22 12:07:34 | INFO    | ⏹️ No candidates cleared the grid. Stopping.
-2026-08-22 12:07:34 | INFO    | 🏁 Extraction complete.
-2026-08-22 12:07:35 | INFO    | Extraction done — computing final coverage (in-memory DuckDB)…
-2026-08-22 12:07:37 | INFO    | Extraction complete — 2 segment(s).