Spaces:
Running
Running
| import gradio as gr | |
| import pandas as pd | |
| import folium | |
| import numpy as np | |
| import os | |
| import re | |
| import json | |
| BASE = os.path.dirname(os.path.abspath(__file__)) if "__file__" in dir() else os.getcwd() | |
| STAY_POINTS = os.path.join(BASE, "data", "stay_points_inference_sample.csv") | |
| POI_PATH = os.path.join(BASE, "data", "poi_inference_sample.csv") | |
| DEMO_PATH = os.path.join(BASE, "data", "demographics_inference_sample.csv") | |
| COT_PATH = os.path.join(BASE, "data", "inference_results_sample.json") | |
| SEX_MAP = {1:"Male", 2:"Female", -8:"Unknown", -7:"Prefer not to answer"} | |
| EDU_MAP = {1:"Less than HS", 2:"HS Graduate/GED", 3:"Some College/Associate", | |
| 4:"Bachelor's Degree", 5:"Graduate/Professional Degree", | |
| -1:"N/A", -7:"Prefer not to answer", -8:"Unknown"} | |
| INC_MAP = {1:"<$10,000", 2:"$10,000–$14,999", 3:"$15,000–$24,999", | |
| 4:"$25,000–$34,999", 5:"$35,000–$49,999", 6:"$50,000–$74,999", | |
| 7:"$75,000–$99,999", 8:"$100,000–$124,999", 9:"$125,000–$149,999", | |
| 10:"$150,000–$199,999", 11:"$200,000+", | |
| -7:"Prefer not to answer", -8:"Unknown", -9:"Not ascertained"} | |
| RACE_MAP = {1:"White", 2:"Black or African American", 3:"Asian", | |
| 4:"American Indian or Alaska Native", | |
| 5:"Native Hawaiian or Other Pacific Islander", | |
| 6:"Multiple races", 97:"Other", | |
| -7:"Prefer not to answer", -8:"Unknown"} | |
| ACT_MAP = {0:"Transportation", 1:"Home", 2:"Work", 3:"School", 4:"ChildCare", | |
| 5:"BuyGoods", 6:"Services", 7:"EatOut", 8:"Errands", 9:"Recreation", | |
| 10:"Exercise", 11:"Visit", 12:"HealthCare", 13:"Religious", | |
| 14:"SomethingElse", 15:"DropOff"} | |
| print("Loading data...") | |
| sp = pd.read_csv(STAY_POINTS) | |
| poi = pd.read_csv(POI_PATH) | |
| demo = pd.read_csv(DEMO_PATH) | |
| sp = sp.merge(poi, on="poi_id", how="left") | |
| sp["start_datetime"] = pd.to_datetime(sp["start_datetime"], utc=True) | |
| sp["end_datetime"] = pd.to_datetime(sp["end_datetime"], utc=True) | |
| sp["duration_min"] = ((sp["end_datetime"] - sp["start_datetime"]).dt.total_seconds() / 60).round(1) | |
| def parse_act_types(x): | |
| try: | |
| codes = list(map(int, str(x).strip("[]").split())) | |
| return ", ".join(ACT_MAP.get(c, str(c)) for c in codes) | |
| except: | |
| return str(x) | |
| sp["act_label"] = sp["act_types"].apply(parse_act_types) | |
| # Load CoT JSON (optional) | |
| cot_by_agent = {} | |
| if os.path.exists(COT_PATH): | |
| with open(COT_PATH, "r") as f: | |
| cot_raw = json.load(f) | |
| # Support both list and {"inference_results": [...]} formats | |
| records = cot_raw if isinstance(cot_raw, list) else cot_raw.get("inference_results", []) | |
| for result in records: | |
| cot_by_agent[int(result["agent_id"])] = result | |
| print(f"Loaded CoT for {len(cot_by_agent)} agents.") | |
| sample_agents = sorted(sp["agent_id"].unique().tolist()) | |
| print(f"Ready. {len(sample_agents)} agents loaded.") | |
| def get_cot(agent_id): | |
| result = cot_by_agent.get(int(agent_id), {}) | |
| s1 = result.get("step1_response", "") | |
| s2 = result.get("step2_response", "") | |
| s3 = result.get("step3_response", "") | |
| return s1, s2, s3 | |
| # ── Mobility text builders ──────────────────────────────────────────────────── | |
| def build_mobility_summary(agent_sp): | |
| top5 = (agent_sp.groupby("name")["duration_min"] | |
| .agg(visits="count", avg_dur="mean") | |
| .sort_values("visits", ascending=False) | |
| .head(5)) | |
| obs_start = agent_sp["start_datetime"].min().strftime("%Y-%m-%d") | |
| obs_end = agent_sp["end_datetime"].max().strftime("%Y-%m-%d") | |
| days = (agent_sp["end_datetime"].max() - agent_sp["start_datetime"].min()).days | |
| # Top activity types | |
| act_counts = agent_sp["act_label"].value_counts().head(3) | |
| top_acts = ", ".join(f"{a} ({n})" for a, n in act_counts.items()) | |
| # Time of day | |
| agent_sp2 = agent_sp.copy() | |
| agent_sp2["hour"] = agent_sp2["start_datetime"].dt.hour | |
| def tod(h): | |
| if 5 <= h < 12: return "Morning" | |
| if 12 <= h < 17: return "Afternoon" | |
| if 17 <= h < 21: return "Evening" | |
| return "Night" | |
| agent_sp2["tod"] = agent_sp2["hour"].apply(tod) | |
| peak_tod = agent_sp2["tod"].value_counts().idxmax() | |
| agent_sp2["is_weekend"] = agent_sp2["start_datetime"].dt.dayofweek >= 5 | |
| wd_pct = int((~agent_sp2["is_weekend"]).mean() * 100) | |
| lines = [ | |
| f"Period: {obs_start} ~ {obs_end} ({days} days)", | |
| f"Stay points: {len(agent_sp)} | Unique locations: {agent_sp['name'].nunique()}", | |
| f"Weekday/Weekend: {wd_pct}% / {100-wd_pct}% | Peak time: {peak_tod}", | |
| f"Top activities: {top_acts}", | |
| "", | |
| "Top Locations:", | |
| ] | |
| for i, (name, row) in enumerate(top5.iterrows(), 1): | |
| lines.append(f" {i}. {name} — {int(row['visits'])} visits, avg {int(row['avg_dur'])} min") | |
| return "\n".join(lines) | |
| def build_weekly_checkin(agent_sp, max_days=None): | |
| agent_sp2 = agent_sp.copy() | |
| agent_sp2["date"] = agent_sp2["start_datetime"].dt.date | |
| all_dates = sorted(agent_sp2["date"].unique()) | |
| dates_to_show = all_dates[:max_days] if max_days else all_dates | |
| total_days = len(all_dates) | |
| lines = ["WEEKLY CHECK-IN SUMMARY", "======================="] | |
| for date in dates_to_show: | |
| grp = agent_sp2[agent_sp2["date"] == date] | |
| dow = grp["start_datetime"].iloc[0].strftime("%A") | |
| label = "Weekend" if grp["start_datetime"].iloc[0].dayofweek >= 5 else "Weekday" | |
| lines.append(f"\n--- {dow}, {date} ({label}) ---") | |
| lines.append(f"Total activities: {len(grp)}") | |
| for _, row in grp.iterrows(): | |
| lines.append( | |
| f"- {row['start_datetime'].strftime('%H:%M')}-" | |
| f"{row['end_datetime'].strftime('%H:%M')} " | |
| f"({int(row['duration_min'])} mins): " | |
| f"{row['name']} - {row['act_label']}" | |
| ) | |
| if max_days and total_days > max_days: | |
| lines.append(f"\n... ({total_days - max_days} more days)") | |
| return "\n".join(lines) | |
| # ── HTML reasoning chain ────────────────────────────────────────────────────── | |
| # ── Paste this entire block into app.py, replacing the existing CHAIN_CSS, render_chain, and helper functions ── | |
| import re | |
| CHAIN_CSS = """ | |
| <style> | |
| @import url('https://fonts.googleapis.com/css2?family=DM+Mono:wght@400;500&family=DM+Sans:wght@300;400;500;600&display=swap'); | |
| .hct-root { | |
| font-family: 'DM Sans', sans-serif; | |
| display: flex; | |
| flex-direction: column; | |
| gap: 0; | |
| padding: 4px 0 8px; | |
| } | |
| /* ── Stage shell ── */ | |
| .hct-stage { | |
| border-radius: 12px; | |
| overflow: hidden; | |
| transition: opacity 0.3s, filter 0.3s; | |
| } | |
| .hct-stage.dim { opacity: 0.28; filter: grayscale(0.6); pointer-events: none; } | |
| .hct-stage.active { opacity: 1; } | |
| /* ── Stage header strip ── */ | |
| .hct-head { | |
| display: flex; | |
| align-items: center; | |
| gap: 10px; | |
| padding: 9px 14px; | |
| } | |
| .hct-num { | |
| font-family: 'DM Mono', monospace; | |
| font-size: 10px; | |
| font-weight: 500; | |
| letter-spacing: 0.12em; | |
| padding: 2px 7px; | |
| border-radius: 4px; | |
| } | |
| .hct-title { | |
| font-size: 11px; | |
| font-weight: 600; | |
| letter-spacing: 0.04em; | |
| text-transform: uppercase; | |
| } | |
| /* Stage 1 colors */ | |
| .hct-s1 { background: #f4f6fb; border: 1.5px solid #d4daf0; } | |
| .hct-s1 .hct-head { background: #eaecf7; border-bottom: 1px solid #d4daf0; } | |
| .hct-s1 .hct-num { background: #dde2f3; color: #3a4a80; } | |
| .hct-s1 .hct-title { color: #3a4a80; } | |
| /* Stage 2 colors */ | |
| .hct-s2 { background: #fdf8f2; border: 1.5px solid #e8d5b8; } | |
| .hct-s2 .hct-head { background: #f7ede0; border-bottom: 1px solid #e8d5b8; } | |
| .hct-s2 .hct-num { background: #f0dcbf; color: #7a4a10; } | |
| .hct-s2 .hct-title { color: #7a4a10; } | |
| /* Stage 3 colors */ | |
| .hct-s3 { background: #fff6f5; border: 2px solid #d4453a; } | |
| .hct-s3 .hct-head { background: #fce8e7; border-bottom: 1px solid #d4453a; } | |
| .hct-s3 .hct-num { background: #d4453a; color: #fff; } | |
| .hct-s3 .hct-title { color: #b0302a; } | |
| /* ── Body ── */ | |
| .hct-body { padding: 12px 14px; } | |
| /* ── Arrow connector ── */ | |
| .hct-arrow { | |
| display: flex; | |
| align-items: center; | |
| gap: 8px; | |
| padding: 5px 18px; | |
| transition: opacity 0.3s; | |
| } | |
| .hct-arrow-line { flex: 1; height: 1px; background: #d8d4ce; } | |
| .hct-arrow-label { | |
| font-family: 'DM Mono', monospace; | |
| font-size: 9px; | |
| color: #b0a898; | |
| letter-spacing: 0.08em; | |
| text-transform: uppercase; | |
| white-space: nowrap; | |
| background: white; | |
| padding: 2px 8px; | |
| border: 1px solid #e0dbd4; | |
| border-radius: 20px; | |
| } | |
| /* ── Stage 1: Location table ── */ | |
| .hct-loc-table { | |
| width: 100%; | |
| border-collapse: collapse; | |
| font-size: 11.5px; | |
| margin-bottom: 10px; | |
| } | |
| .hct-loc-table th { | |
| font-family: 'DM Mono', monospace; | |
| font-size: 9px; | |
| font-weight: 500; | |
| letter-spacing: 0.1em; | |
| text-transform: uppercase; | |
| color: #8090b0; | |
| border-bottom: 1px solid #d4daf0; | |
| padding: 3px 6px 5px; | |
| text-align: left; | |
| } | |
| .hct-loc-table th:not(:first-child) { text-align: right; } | |
| .hct-loc-table td { | |
| padding: 5px 6px; | |
| color: #2a3050; | |
| border-bottom: 1px solid #eaecf5; | |
| line-height: 1.3; | |
| } | |
| .hct-loc-table td:not(:first-child) { text-align: right; font-family: 'DM Mono', monospace; font-size: 11px; color: #5060a0; } | |
| .hct-loc-table tr:last-child td { border-bottom: none; } | |
| .hct-loc-name { font-weight: 500; max-width: 160px; overflow: hidden; text-overflow: ellipsis; white-space: nowrap; display: block; } | |
| .hct-visit-bar-wrap { display: flex; align-items: center; gap: 6px; justify-content: flex-end; } | |
| .hct-visit-bar { height: 4px; border-radius: 2px; background: #6878c8; opacity: 0.55; } | |
| /* ── Stage 1: Temporal panel ── */ | |
| .hct-temporal { | |
| display: grid; | |
| grid-template-columns: 1fr 1fr; | |
| gap: 8px; | |
| } | |
| .hct-temp-block { | |
| background: #eef0fa; | |
| border-radius: 8px; | |
| padding: 8px 10px; | |
| } | |
| .hct-temp-label { | |
| font-family: 'DM Mono', monospace; | |
| font-size: 9px; | |
| font-weight: 500; | |
| letter-spacing: 0.1em; | |
| text-transform: uppercase; | |
| color: #7080b0; | |
| margin-bottom: 6px; | |
| } | |
| .hct-seg-row { display: flex; height: 10px; border-radius: 5px; overflow: hidden; margin-bottom: 5px; } | |
| .hct-seg { display: flex; align-items: center; justify-content: center; font-size: 0; transition: width 0.5s; } | |
| .seg-morning { background: #fbbf24; } | |
| .seg-afternoon{ background: #f97316; } | |
| .seg-evening { background: #8b5cf6; } | |
| .seg-night { background: #1e3a5f; } | |
| .seg-weekday { background: #6878c8; } | |
| .seg-weekend { background: #e8c080; } | |
| .hct-legend { display: flex; flex-wrap: wrap; gap: 4px 10px; } | |
| .hct-leg-item { display: flex; align-items: center; gap: 4px; font-size: 10px; color: #5a6080; } | |
| .hct-leg-dot { width: 8px; height: 8px; border-radius: 2px; flex-shrink: 0; } | |
| .hct-dist-line { | |
| margin-top: 8px; | |
| font-size: 11px; | |
| color: #6070a0; | |
| font-family: 'DM Mono', monospace; | |
| padding: 5px 8px; | |
| background: #eef0fa; | |
| border-radius: 6px; | |
| display: flex; | |
| align-items: center; | |
| gap: 6px; | |
| } | |
| /* ── Stage 2: 2×2 grid ── */ | |
| .hct-dim-grid { | |
| display: grid; | |
| grid-template-columns: 1fr 1fr; | |
| gap: 8px; | |
| } | |
| .hct-dim-card { | |
| background: #fff; | |
| border: 1px solid #e8d5b8; | |
| border-radius: 8px; | |
| padding: 9px 11px; | |
| } | |
| .hct-dim-head { | |
| display: flex; | |
| align-items: center; | |
| gap: 6px; | |
| margin-bottom: 5px; | |
| } | |
| .hct-dim-icon { font-size: 13px; line-height: 1; } | |
| .hct-dim-name { | |
| font-family: 'DM Mono', monospace; | |
| font-size: 9px; | |
| font-weight: 500; | |
| letter-spacing: 0.1em; | |
| text-transform: uppercase; | |
| color: #a07040; | |
| } | |
| .hct-dim-text { | |
| font-size: 11px; | |
| color: #3a2a10; | |
| line-height: 1.55; | |
| } | |
| .hct-dim-empty { color: #ccc; font-style: italic; } | |
| /* ── Stage 3: prediction ── */ | |
| .hct-pred-row { | |
| display: flex; | |
| align-items: flex-start; | |
| gap: 16px; | |
| margin-bottom: 10px; | |
| } | |
| .hct-pred-badge { | |
| background: #d4453a; | |
| color: white; | |
| border-radius: 8px; | |
| padding: 8px 14px; | |
| text-align: center; | |
| flex-shrink: 0; | |
| } | |
| .hct-pred-val { font-size: 18px; font-weight: 600; line-height: 1.2; white-space: nowrap; } | |
| .hct-pred-sub { font-family: 'DM Mono', monospace; font-size: 9px; opacity: 0.8; letter-spacing: 0.08em; text-transform: uppercase; margin-top: 2px; } | |
| .hct-conf-col { flex: 1; padding-top: 4px; } | |
| .hct-conf-label { font-family: 'DM Mono', monospace; font-size: 9px; color: #a04040; letter-spacing: 0.08em; text-transform: uppercase; margin-bottom: 4px; } | |
| .hct-conf-track { height: 6px; background: #f0d0cf; border-radius: 3px; overflow: hidden; margin-bottom: 6px; } | |
| .hct-conf-fill { height: 100%; background: linear-gradient(90deg, #e74c3c, #8b0000); border-radius: 3px; } | |
| .hct-reasoning { | |
| font-size: 11.5px; | |
| color: #4a2020; | |
| line-height: 1.6; | |
| border-left: 3px solid #e8b0ae; | |
| padding-left: 10px; | |
| } | |
| /* ── Idle / loading states ── */ | |
| .hct-idle { font-size: 12px; color: #b0bac8; padding: 6px 0; font-style: italic; } | |
| .hct-loading { | |
| font-size: 12px; padding: 6px 0; | |
| display: flex; align-items: center; gap: 8px; | |
| } | |
| .hct-dot { width: 6px; height: 6px; border-radius: 50%; display: inline-block; animation: hct-pulse 1.2s ease-in-out infinite; } | |
| .hct-dot:nth-child(2) { animation-delay: 0.2s; } | |
| .hct-dot:nth-child(3) { animation-delay: 0.4s; } | |
| @keyframes hct-pulse { | |
| 0%,100% { opacity: 0.2; transform: scale(0.8); } | |
| 50% { opacity: 1; transform: scale(1.1); } | |
| } | |
| .hct-s1 .hct-dot { background: #6878c8; } | |
| .hct-s2 .hct-dot { background: #c08040; } | |
| .hct-s3 .hct-dot { background: #d4453a; } | |
| </style> | |
| """ | |
| def _loading(msg): | |
| return f'<div class="hct-loading"><span class="hct-dot"></span><span class="hct-dot"></span><span class="hct-dot"></span><span style="color:#8090a0;font-size:12px">{msg}</span></div>' | |
| def _parse_s1(text): | |
| """Returns (locations, tod, wk, dist)""" | |
| locations = [] | |
| dur_map = {} | |
| tod = {} | |
| wk = {} | |
| dist = None | |
| for line in text.splitlines(): | |
| s = line.strip() | |
| # Location inventory: "- Name: N visits, ..." | |
| m = re.match(r'-\s+(.+?):\s+(\d+)\s+visit', s, re.IGNORECASE) | |
| if m: | |
| locations.append((m.group(1).strip(), int(m.group(2)))) | |
| # Duration: "- LocationName: Average duration of X minutes" | |
| m2 = re.match(r'-?\s*(.+?):\s+Average duration of ([\d.]+)\s+min', s, re.IGNORECASE) | |
| if m2: | |
| dur_map[m2.group(1).strip()] = float(m2.group(2)) | |
| # TOD: "65% morning, 23% afternoon, 6% evening, 5% night" | |
| if not tod: | |
| m3 = re.search(r'(\d+)%\s*morning.*?(\d+)%\s*afternoon.*?(\d+)%\s*evening.*?(\d+)%\s*night', s, re.IGNORECASE) | |
| if m3: | |
| tod = {'Morning': int(m3.group(1)), 'Afternoon': int(m3.group(2)), | |
| 'Evening': int(m3.group(3)), 'Night': int(m3.group(4))} | |
| # Weekday/weekend | |
| if not wk: | |
| m4 = re.search(r'(\d+)%\s*weekday.*?(\d+)%\s*weekend', s, re.IGNORECASE) | |
| if m4: | |
| wk = {'Weekday': int(m4.group(1)), 'Weekend': int(m4.group(2))} | |
| # Distance | |
| if not dist: | |
| m5 = re.search(r'average distance of approximately ([\d.]+)\s*miles', s, re.IGNORECASE) | |
| if m5: | |
| dist = float(m5.group(1)) | |
| result_locs = [(n, v, dur_map.get(n)) for n, v in locations[:7]] | |
| return result_locs, tod, wk, dist | |
| def _parse_s2(text): | |
| """Returns dict: ROUTINE, ECONOMIC, SOCIAL, URBAN, STABILITY → short summary string""" | |
| DIMS = { | |
| 'ROUTINE': ['ROUTINE', 'SCHEDULE'], | |
| 'ECONOMIC': ['ECONOMIC', 'SPENDING'], | |
| 'SOCIAL': ['SOCIAL', 'LIFESTYLE'], | |
| 'URBAN': ['URBAN', 'COMMUNITY'], | |
| 'STABILITY': ['STABILITY', 'REGULARITY', 'CONSISTENCY'], | |
| } | |
| sections = {} | |
| current_key = None | |
| current_lines = [] | |
| for line in text.splitlines(): | |
| s = line.strip() | |
| # Format A: "1. TITLE ANALYSIS:" or "2. ECONOMIC BEHAVIOR PATTERNS:" | |
| mA = re.match(r'^\d+\.\s+([A-Z][A-Z\s&]+?)(?:\s+ANALYSIS|\s+PATTERNS|\s+INDICATORS|\s+CHARACTERISTICS|\s+STABILITY)?:\s*$', s, re.IGNORECASE) | |
| # Format B: "STEP 1: ROUTINE & SCHEDULE ANALYSIS" | |
| mB = re.match(r'^STEP\s+\d+:\s+([A-Z][A-Z\s&]+?)(?:\s+ANALYSIS|\s+PATTERNS|\s+INDICATORS|\s+CHARACTERISTICS|\s+STABILITY)?\s*$', s, re.IGNORECASE) | |
| mm = mA or mB | |
| if mm: | |
| if current_key and current_lines: | |
| sections[current_key] = ' '.join(current_lines) | |
| current_key = mm.group(1).upper().strip() | |
| current_lines = [] | |
| elif current_key and s: | |
| if re.match(r'^\d+\.\d+', s): | |
| sub = re.sub(r'^\d+\.\d+[^:]*:\s*', '', s) | |
| if sub: | |
| current_lines.append(sub) | |
| elif s.startswith('-'): | |
| current_lines.append(s.lstrip('-').strip()) | |
| elif not re.match(r'^\d+\.', s): | |
| current_lines.append(s) | |
| if current_key and current_lines: | |
| sections[current_key] = ' '.join(current_lines) | |
| result = {} | |
| for dim, keywords in DIMS.items(): | |
| for k, txt in sections.items(): | |
| if any(kw in k for kw in keywords) and txt: | |
| sents = re.split(r'(?<=[.!?])\s+', txt.strip()) | |
| summary = ' '.join(sents[:2]) | |
| if len(summary) > 160: | |
| summary = summary[:157] + '…' | |
| result[dim] = summary | |
| break | |
| return result | |
| def _parse_s3(text): | |
| pred, conf, reasoning = '', 0, '' | |
| in_r = False | |
| r_lines = [] | |
| for line in text.splitlines(): | |
| s = line.strip() | |
| if s.startswith('INCOME_PREDICTION:'): | |
| pred = s.replace('INCOME_PREDICTION:', '').strip() | |
| elif s.startswith('INCOME_CONFIDENCE:'): | |
| try: | |
| conf = int(re.search(r'\d+', s).group()) | |
| except: | |
| conf = 0 | |
| elif s.startswith('INCOME_REASONING:'): | |
| in_r = True | |
| r_lines.append(s.replace('INCOME_REASONING:', '').strip()) | |
| elif in_r: | |
| if re.match(r'^2\.', s) or s.startswith('INCOME_'): | |
| break | |
| if s: | |
| r_lines.append(s) | |
| reasoning = ' '.join(r_lines).strip() | |
| sents = re.split(r'(?<=[.!?])\s+', reasoning) | |
| reasoning = ' '.join(sents[:3]) | |
| if len(reasoning) > 280: | |
| reasoning = reasoning[:277] + '…' | |
| return pred, conf, reasoning | |
| def _s1_body(text, active): | |
| if not active: | |
| return '<div class="hct-idle">Press ▶ to start</div>' | |
| if not text: | |
| return _loading('Extracting features') | |
| locs, tod, wk, dist = _parse_s1(text) | |
| # Location table | |
| max_v = max((v for _, v, _ in locs), default=1) | |
| rows = '' | |
| for name, visits, dur in locs: | |
| bar_w = int(60 * visits / max_v) | |
| dur_str = f'{int(dur)}m' if dur else '—' | |
| rows += ( | |
| f'<tr>' | |
| f'<td><span class="hct-loc-name" title="{name}">{name}</span></td>' | |
| f'<td><div class="hct-visit-bar-wrap">' | |
| f'<div class="hct-visit-bar" style="width:{bar_w}px"></div>' | |
| f'{visits}</div></td>' | |
| f'<td>{dur_str}</td>' | |
| f'</tr>' | |
| ) | |
| table = ( | |
| f'<table class="hct-loc-table">' | |
| f'<thead><tr><th>Location</th><th>Visits</th><th>Avg Stay</th></tr></thead>' | |
| f'<tbody>{rows}</tbody>' | |
| f'</table>' | |
| ) if rows else '' | |
| # Temporal panels | |
| def seg_bar(data, seg_classes): | |
| total = sum(data.values()) or 1 | |
| segs = ''.join( | |
| f'<div class="hct-seg {cls}" style="width:{int(100*v/total)}%"></div>' | |
| for (label, v), cls in zip(data.items(), seg_classes) | |
| ) | |
| legend = ''.join( | |
| f'<div class="hct-leg-item"><div class="hct-leg-dot {cls}"></div>{label} {v}%</div>' | |
| for (label, v), cls in zip(data.items(), seg_classes) | |
| ) | |
| return f'<div class="hct-seg-row">{segs}</div><div class="hct-legend">{legend}</div>' | |
| tod_panel = '' | |
| if tod: | |
| tod_panel = ( | |
| f'<div class="hct-temp-block">' | |
| f'<div class="hct-temp-label">Time of Day</div>' | |
| f'{seg_bar(tod, ["seg-morning","seg-afternoon","seg-evening","seg-night"])}' | |
| f'</div>' | |
| ) | |
| wk_panel = '' | |
| if wk: | |
| wk_panel = ( | |
| f'<div class="hct-temp-block">' | |
| f'<div class="hct-temp-label">Weekday / Weekend</div>' | |
| f'{seg_bar(wk, ["seg-weekday","seg-weekend"])}' | |
| f'</div>' | |
| ) | |
| temporal = f'<div class="hct-temporal">{tod_panel}{wk_panel}</div>' if (tod_panel or wk_panel) else '' | |
| dist_line = '' | |
| if dist: | |
| dist_line = f'<div class="hct-dist-line">📍 Avg trip distance {dist} mi</div>' | |
| return table + temporal + dist_line | |
| def _s2_body(text, active): | |
| if not active: | |
| return '<div class="hct-idle">Waiting…</div>' | |
| if not text: | |
| return _loading('Analyzing behavior') | |
| dims = _parse_s2(text) | |
| DIM_META = [ | |
| ('ROUTINE', '🕐', 'Schedule'), | |
| ('ECONOMIC', '💰', 'Economic'), | |
| ('SOCIAL', '👥', 'Social'), | |
| ('STABILITY', '🔄', 'Stability'), | |
| ] | |
| # fallback to URBAN if STABILITY missing | |
| if 'STABILITY' not in dims and 'URBAN' in dims: | |
| dims['STABILITY'] = dims['URBAN'] | |
| cards = '' | |
| for key, icon, label in DIM_META: | |
| txt = dims.get(key, '') | |
| content = f'<div class="hct-dim-text">{txt}</div>' if txt else '<div class="hct-dim-text hct-dim-empty">—</div>' | |
| cards += ( | |
| f'<div class="hct-dim-card">' | |
| f'<div class="hct-dim-head">' | |
| f'<span class="hct-dim-icon">{icon}</span>' | |
| f'<span class="hct-dim-name">{label}</span>' | |
| f'</div>' | |
| f'{content}' | |
| f'</div>' | |
| ) | |
| return f'<div class="hct-dim-grid">{cards}</div>' | |
| def _s3_body(text, active): | |
| if not active: | |
| return '<div class="hct-idle">Waiting…</div>' | |
| if not text: | |
| return _loading('Inferring demographics') | |
| pred, conf, reasoning = _parse_s3(text) | |
| conf_pct = int(conf / 5 * 100) | |
| return ( | |
| f'<div class="hct-pred-row">' | |
| f'<div class="hct-pred-badge">' | |
| f'<div class="hct-pred-val">{pred or "—"}</div>' | |
| f'<div class="hct-pred-sub">Income</div>' | |
| f'</div>' | |
| f'<div class="hct-conf-col">' | |
| f'<div class="hct-conf-label">Confidence {conf}/5</div>' | |
| f'<div class="hct-conf-track"><div class="hct-conf-fill" style="width:{conf_pct}%"></div></div>' | |
| f'</div>' | |
| f'</div>' | |
| f'<div class="hct-reasoning">{reasoning}</div>' | |
| ) | |
| def render_chain(s1_text, s2_text, s3_text, status="idle"): | |
| s1_on = status in ("running1", "running2", "running3", "done") | |
| s2_on = status in ("running2", "running3", "done") | |
| s3_on = status in ("running3", "done") | |
| # For "running" states the text may be empty → show loading dots | |
| s1_body = _s1_body(s1_text if s1_on else '', s1_on) | |
| s2_body = _s2_body(s2_text if s2_on else '', s2_on) | |
| s3_body = _s3_body(s3_text if s3_on else '', s3_on) | |
| def stage(cls, num, title, body, on): | |
| dim_cls = 'active' if on else 'dim' | |
| return ( | |
| f'<div class="hct-stage hct-{cls} {dim_cls}">' | |
| f'<div class="hct-head">' | |
| f'<span class="hct-num">{num}</span>' | |
| f'<span class="hct-title">{title}</span>' | |
| f'</div>' | |
| f'<div class="hct-body">{body}</div>' | |
| f'</div>' | |
| ) | |
| def arrow(label, on): | |
| op = '1' if on else '0.2' | |
| return ( | |
| f'<div class="hct-arrow" style="opacity:{op}">' | |
| f'<div class="hct-arrow-line"></div>' | |
| f'<div class="hct-arrow-label">{label}</div>' | |
| f'<div class="hct-arrow-line"></div>' | |
| f'</div>' | |
| ) | |
| html = CHAIN_CSS + '<div class="hct-root">' | |
| html += stage('s1', 'Stage 01', 'Feature Extraction', s1_body, s1_on) | |
| html += arrow('behavioral abstraction', s2_on) | |
| html += stage('s2', 'Stage 02', 'Behavioral Analysis', s2_body, s2_on) | |
| html += arrow('demographic inference', s3_on) | |
| html += stage('s3', 'Stage 03', 'Demographic Inference', s3_body, s3_on) | |
| html += '</div>' | |
| return html | |
| def build_map(agent_sp): | |
| agent_sp = agent_sp.reset_index(drop=True).copy() | |
| agent_sp["latitude"] += np.random.uniform(-0.0003, 0.0003, len(agent_sp)) | |
| agent_sp["longitude"] += np.random.uniform(-0.0003, 0.0003, len(agent_sp)) | |
| lat = agent_sp["latitude"].mean() | |
| lon = agent_sp["longitude"].mean() | |
| m = folium.Map(location=[lat, lon], zoom_start=12, tiles="CartoDB positron") | |
| coords = list(zip(agent_sp["latitude"], agent_sp["longitude"])) | |
| if len(coords) > 1: | |
| folium.PolyLine(coords, color="#cc000055", weight=1.5, opacity=0.4).add_to(m) | |
| n = len(agent_sp) | |
| for i, row in agent_sp.iterrows(): | |
| ratio = i / max(n - 1, 1) | |
| r = int(255 - ratio * (255 - 139)) | |
| g = int(204 * (1 - ratio) ** 2) | |
| b = 0 | |
| color = f"#{r:02x}{g:02x}{b:02x}" | |
| folium.CircleMarker( | |
| location=[row["latitude"], row["longitude"]], | |
| radius=7, color=color, fill=True, fill_color=color, fill_opacity=0.9, | |
| popup=folium.Popup( | |
| f"<b>#{i+1} {row['name']}</b><br>" | |
| f"{row['start_datetime'].strftime('%a %m/%d %H:%M')}<br>" | |
| f"{int(row['duration_min'])} min<br>{row['act_label']}", | |
| max_width=220 | |
| ) | |
| ).add_to(m) | |
| legend_html = """ | |
| <div style=" | |
| position:fixed; bottom:18px; left:18px; z-index:9999; | |
| background:rgba(255,255,255,0.92); border-radius:8px; | |
| padding:8px 12px; font-size:11px; font-family:sans-serif; | |
| box-shadow:0 1px 5px rgba(0,0,0,0.2); line-height:1.8; | |
| "> | |
| <div style="font-weight:600;margin-bottom:4px;">Stay Point Legend</div> | |
| <div style="display:flex;align-items:center;gap:6px;"> | |
| <svg width="60" height="10"> | |
| <defs><linearGradient id="lg" x1="0" x2="1" y1="0" y2="0"> | |
| <stop offset="0%" stop-color="#ffcc00"/> | |
| <stop offset="100%" stop-color="#8b0000"/> | |
| </linearGradient></defs> | |
| <rect width="60" height="10" rx="4" fill="url(#lg)"/> | |
| </svg> | |
| <span>Earlier → Later</span> | |
| </div> | |
| <div style="display:flex;align-items:center;gap:6px;margin-top:2px;"> | |
| <svg width="14" height="14"><circle cx="7" cy="7" r="5" fill="#cc4444" opacity="0.5"/></svg> | |
| <span>Movement path</span> | |
| </div> | |
| <div style="color:#999;font-size:10px;margin-top:2px;">Click dot for details</div> | |
| </div> | |
| """ | |
| m.get_root().html.add_child(folium.Element(legend_html)) | |
| m.get_root().width = "100%" | |
| m.get_root().height = "420px" | |
| return m._repr_html_() | |
| def build_demo_text(row): | |
| age = int(row["age"]) if row["age"] > 0 else "Unknown" | |
| return ( | |
| f"Age: {age} | " | |
| f"Sex: {SEX_MAP.get(int(row['sex']), row['sex'])} | " | |
| f"Race: {RACE_MAP.get(int(row['race']), row['race'])} | " | |
| f"Education: {EDU_MAP.get(int(row['education']), row['education'])} | " | |
| f"Income: {INC_MAP.get(int(row['hh_income']), row['hh_income'])}" | |
| ) | |
| # ── Callbacks ───────────────────────────────────────────────────────────────── | |
| def on_select(agent_id): | |
| agent_id = int(agent_id) | |
| agent_sp = sp[sp["agent_id"] == agent_id].sort_values("start_datetime") | |
| agent_demo = demo[demo["agent_id"] == agent_id].iloc[0] | |
| map_html = build_map(agent_sp) | |
| demo_text = build_demo_text(agent_demo) | |
| raw_text = build_mobility_summary(agent_sp) + "\n\n" + build_weekly_checkin(agent_sp) | |
| chain_html = render_chain("", "", "", status="idle") | |
| return map_html, raw_text, demo_text, chain_html | |
| def run_step(agent_id, step): | |
| """Reveal one stage per click. step: 0->1->2->done(-1)""" | |
| agent_id = int(agent_id) | |
| s1, s2, s3 = get_cot(agent_id) | |
| next_step = step + 1 | |
| if next_step == 1: | |
| html = render_chain(s1, "", "", status="running2") | |
| label = "▶ Stage 2: Behavioral Analysis" | |
| return html, 1, gr.update(value=label) | |
| elif next_step == 2: | |
| html = render_chain(s1, s2, "", status="running3") | |
| label = "▶ Stage 3: Demographic Inference" | |
| return html, 2, gr.update(value=label) | |
| else: | |
| html = render_chain(s1, s2, s3, status="done") | |
| return html, -1, gr.update(value="↺ Reset") | |
| def handle_btn(agent_id, step): | |
| if step == -1: | |
| html = render_chain("", "", "", status="idle") | |
| return html, 0, gr.update(value="▶ Stage 1: Feature Extraction") | |
| return run_step(agent_id, step) | |
| def on_select_reset(agent_id): | |
| agent_id_int = int(agent_id) | |
| agent_sp = sp[sp["agent_id"] == agent_id_int].sort_values("start_datetime") | |
| agent_demo = demo[demo["agent_id"] == agent_id_int].iloc[0] | |
| map_html = build_map(agent_sp) | |
| demo_text = build_demo_text(agent_demo) | |
| cot_entry = cot_by_agent.get(agent_id_int, {}) | |
| summary = build_mobility_summary(agent_sp) | |
| raw_full = cot_entry.get("weekly_checkin") or build_weekly_checkin(agent_sp) | |
| sep = "\n\n--- " | |
| parts = raw_full.split(sep) | |
| extra = len(parts) - 1 | |
| raw_text = parts[0] + (sep.join([""] + parts[1:2]) + ("\n\n... ({} more days)".format(extra - 1) if extra > 1 else "")) if extra > 0 else raw_full | |
| chain_html = render_chain("", "", "", status="idle") | |
| return map_html, summary, raw_text, demo_text, chain_html, 0, gr.update(value="▶ Stage 1: Feature Extraction") | |
| SHOWCASE_AGENTS = sample_agents[:6] | |
| def build_agent_cards(selected_id): | |
| selected_id = int(selected_id) | |
| parts = [] | |
| parts.append("<div style='display:grid;grid-template-columns:repeat(3,1fr);gap:10px;padding:4px 0;'>") | |
| for aid in SHOWCASE_AGENTS: | |
| row = demo[demo["agent_id"] == aid].iloc[0] | |
| age = int(row["age"]) if row["age"] > 0 else "?" | |
| sex = SEX_MAP.get(int(row["sex"]), "?") | |
| edu = EDU_MAP.get(int(row["education"]), "?") | |
| inc = INC_MAP.get(int(row["hh_income"]), "?") | |
| is_sel = (aid == selected_id) | |
| sel_style = "border:2px solid #c0392b;background:#fff8f8;box-shadow:0 2px 8px rgba(192,57,43,0.15);" | |
| nor_style = "border:1.5px solid #ddd;background:#fafafa;box-shadow:0 1px 3px rgba(0,0,0,0.06);" | |
| style = sel_style if is_sel else nor_style | |
| dot = "<span style='display:inline-block;width:8px;height:8px;border-radius:50%;background:#c0392b;margin-right:5px;'></span>" if is_sel else "" | |
| js = "var t=document.querySelector('#agent_hidden_input textarea');t.value='AID';t.dispatchEvent(new Event('input',{bubbles:true}));".replace("AID", str(aid)) | |
| parts.append( | |
| "<div onclick=\"" + js + "\" style=\"cursor:pointer;border-radius:10px;padding:10px 13px;transition:all 0.2s;" + style + "\">" | |
| "<div style='font-size:11px;font-weight:700;color:#c0392b;margin-bottom:3px;font-family:monospace;'>" + dot + "Agent #" + str(aid) + "</div>" | |
| "<div style='font-size:11px;color:#333;line-height:1.6;'>" | |
| "<b>Age:</b> " + str(age) + " <b>Sex:</b> " + sex + "<br>" | |
| "<b>Edu:</b> " + edu + "<br>" | |
| "<b>Income:</b> " + inc + "</div></div>" | |
| ) | |
| parts.append("</div>") | |
| return "".join(parts) | |
| # ── UI ──────────────────────────────────────────────────────────────────────── | |
| with gr.Blocks(title="HiCoTraj Demo") as app: | |
| gr.Markdown("## HiCoTraj — Trajectory Visualization & Hierarchical CoT Demo") | |
| gr.Markdown("*Zero-Shot Demographic Reasoning via Hierarchical Chain-of-Thought Prompting from Trajectory* · ACM SIGSPATIAL GeoGenAgent 2025") | |
| gr.Markdown(""" | |
| **Dataset:** NUMOSIM — a synthetic mobility dataset with realistic activity patterns across 6,000 agents. | |
| > Stanford C, Adari S, Liao X, et al. *NUMoSim: A Synthetic Mobility Dataset with Anomaly Detection Benchmarks.* ACM SIGSPATIAL Workshop on Geospatial Anomaly Detection, 2024. | |
| """) | |
| gr.Markdown("### Select Agent") | |
| agent_cards = gr.HTML(value=build_agent_cards(SHOWCASE_AGENTS[0])) | |
| agent_hidden = gr.Textbox( | |
| value=str(SHOWCASE_AGENTS[0]), | |
| visible=True, | |
| elem_id="agent_hidden_input", | |
| elem_classes=["hidden-input"] | |
| ) | |
| gr.HTML("<style>.hidden-input { display:none !important; }</style>") | |
| with gr.Row(): | |
| with gr.Column(scale=1): | |
| gr.Markdown("### Trajectory Map") | |
| map_out = gr.HTML() | |
| gr.Markdown("### NUMOSIM Raw Data") | |
| with gr.Tabs(): | |
| with gr.Tab("Summary"): | |
| summary_out = gr.Textbox(lines=10, interactive=False, label="", show_label=False) | |
| with gr.Tab("Raw Data"): | |
| raw_out = gr.Textbox(lines=10, interactive=False, label="", show_label=False) | |
| show_all_btn = gr.Button("Show All Days", size="sm", variant="secondary") | |
| with gr.Column(scale=1): | |
| gr.Markdown("### Hierarchical Chain-of-Thought Reasoning") | |
| step_state = gr.State(value=0) | |
| run_btn = gr.Button("▶ Stage 1: Feature Extraction", variant="primary") | |
| chain_out = gr.HTML(value=render_chain("", "", "", status="idle")) | |
| def on_agent_click(agent_id): | |
| cards_html = build_agent_cards(agent_id) | |
| map_html, summary, raw_text, _demo_text, chain_html, step, btn = on_select_reset(agent_id) | |
| return cards_html, map_html, summary, raw_text, chain_html, step, btn | |
| agent_hidden.change( | |
| fn=on_agent_click, inputs=agent_hidden, | |
| outputs=[agent_cards, map_out, summary_out, raw_out, chain_out, step_state, run_btn] | |
| ) | |
| def on_load(agent_id): | |
| map_html, summary, raw_text, _demo_text, chain_html, step, btn = on_select_reset(agent_id) | |
| return map_html, summary, raw_text, chain_html, step, btn | |
| app.load( | |
| fn=on_load, inputs=agent_hidden, | |
| outputs=[map_out, summary_out, raw_out, chain_out, step_state, run_btn] | |
| ) | |
| run_btn.click( | |
| fn=handle_btn, inputs=[agent_hidden, step_state], | |
| outputs=[chain_out, step_state, run_btn] | |
| ) | |
| def toggle_raw(agent_id, current_text): | |
| agent_id_int = int(agent_id) | |
| cot_entry = cot_by_agent.get(agent_id_int, {}) | |
| agent_sp = sp[sp["agent_id"] == agent_id_int].sort_values("start_datetime") | |
| raw_full = cot_entry.get("weekly_checkin") or build_weekly_checkin(agent_sp) | |
| if "more days" in current_text: | |
| return raw_full, gr.update(value="Show Less") | |
| else: | |
| sep = "\n\n--- " | |
| parts = raw_full.split(sep) | |
| extra = len(parts) - 1 | |
| short = parts[0] + (sep.join([""] + parts[1:2]) + ("\n\n... ({} more days)".format(extra - 1) if extra > 1 else "")) if extra > 0 else raw_full | |
| return short, gr.update(value="Show All Days") | |
| show_all_btn.click( | |
| fn=toggle_raw, inputs=[agent_hidden, raw_out], | |
| outputs=[raw_out, show_all_btn] | |
| ) | |
| app.launch(show_error=True, theme=gr.themes.Soft()) |