Spaces:
Running
Running
File size: 19,992 Bytes
eb04905 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 | """
app.py β Streamlit frontend for the Smart MCQ Solver.
Run locally:
streamlit run src/app.py
"""
import sys
import os
import numpy as np
# ββ Ensure the project root is on the Python path so `src.*` imports work ββββ
_SRC_DIR = os.path.dirname(os.path.abspath(__file__))
_PROJECT_ROOT = os.path.dirname(_SRC_DIR)
if _PROJECT_ROOT not in sys.path:
sys.path.insert(0, _PROJECT_ROOT)
import streamlit as st
from src.config import (
DEVICE, ID2LABEL, BAR_COLORS, EXAMPLES,
MODEL_SOURCE,
)
from src.model import load_roberta_model, predict_mcq
# ββ Page config βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.set_page_config(
page_title="Smart MCQ Solver",
page_icon="π§ ",
layout="wide",
initial_sidebar_state="expanded",
)
# ββ Session State Init ββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
for k in ["q_prompt", "opt_A", "opt_B", "opt_C", "opt_D", "opt_E"]:
if k not in st.session_state:
st.session_state[k] = ""
def load_example():
"""Callback to update text areas when an example is selected."""
choice = st.session_state.ex_choice_box
if choice != "(none)":
idx = int(choice.split()[-1]) - 1
ex = EXAMPLES[idx]
st.session_state.q_prompt = ex["prompt"]
st.session_state.opt_A = ex["A"]
st.session_state.opt_B = ex["B"]
st.session_state.opt_C = ex["C"]
st.session_state.opt_D = ex["D"]
st.session_state.opt_E = ex["E"]
else:
st.session_state.q_prompt = ""
st.session_state.opt_A = ""
st.session_state.opt_B = ""
st.session_state.opt_C = ""
st.session_state.opt_D = ""
st.session_state.opt_E = ""
# ββ Custom CSS ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.markdown("""
<style>
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&display=swap');
html, body, [class*="css"] {
font-family: 'Inter', sans-serif;
}
.stApp {
background: linear-gradient(135deg, #0f0c29, #302b63, #24243e);
color: #e0e0e0;
}
/* ββ Glass card βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.glass-card {
background: rgba(255, 255, 255, 0.05);
border: 1px solid rgba(255, 255, 255, 0.1);
border-radius: 16px;
padding: 24px;
backdrop-filter: blur(16px);
margin-bottom: 16px;
box-shadow: 0 4px 30px rgba(0, 0, 0, 0.1);
}
/* ββ Hero ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.hero-title {
font-size: 2.8rem;
font-weight: 800;
background: linear-gradient(90deg, #a78bfa, #60a5fa, #34d399);
-webkit-background-clip: text;
-webkit-text-fill-color: transparent;
background-clip: text;
text-align: center;
margin-bottom: 4px;
letter-spacing: -0.02em;
}
.hero-sub {
text-align: center;
color: #94a3b8;
font-size: 1.05rem;
margin-bottom: 32px;
font-weight: 400;
}
/* ββ Prediction badge βββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.pred-badge {
display: inline-block;
font-size: 4rem;
font-weight: 800;
background: linear-gradient(135deg, #a78bfa, #60a5fa);
-webkit-background-clip: text;
-webkit-text-fill-color: transparent;
background-clip: text;
line-height: 1.2;
}
.pred-label {
font-size: 0.85rem;
color: #64748b;
text-transform: uppercase;
letter-spacing: 0.12em;
font-weight: 700;
}
/* ββ Probability bars βββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.prob-row {
display: flex;
align-items: center;
margin: 8px 0;
gap: 12px;
}
.prob-label {
font-size: 1.05rem;
font-weight: 700;
width: 28px;
text-align: center;
color: #e2e8f0;
}
.prob-track {
flex: 1;
height: 12px;
background: rgba(255,255,255,0.06);
border-radius: 12px;
overflow: hidden;
}
.prob-fill {
height: 100%;
border-radius: 12px;
transition: width 0.6s cubic-bezier(0.22, 1, 0.36, 1);
}
.prob-pct {
font-size: 0.9rem;
color: #94a3b8;
font-weight: 500;
width: 56px;
text-align: right;
}
/* ββ Chips βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.chip {
display: inline-block;
padding: 8px 18px;
border-radius: 999px;
font-weight: 600;
font-size: 0.95rem;
margin: 4px;
box-shadow: 0 2px 10px rgba(0,0,0,0.1);
}
.chip-1 { background: linear-gradient(135deg,#a78bfa,#7c3aed); color:#fff; }
.chip-2 { background: rgba(96,165,250,0.15); color:#60a5fa; border:1px solid rgba(96,165,250,0.4); }
.chip-3 { background: rgba(52,211,153,0.1); color:#34d399; border:1px solid rgba(52,211,153,0.3); }
/* ββ Section headers ββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.section-header {
font-size: 0.85rem;
font-weight: 700;
text-transform: uppercase;
letter-spacing: 0.15em;
color: #a78bfa;
margin-bottom: 12px;
}
/* ββ Text inputs ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.stTextArea textarea, .stTextInput input {
background: rgba(255,255,255,0.04) !important;
border: 1px solid rgba(255,255,255,0.1) !important;
border-radius: 12px !important;
color: #f1f5f9 !important;
font-family: 'Inter', sans-serif !important;
font-size: 0.95rem !important;
line-height: 1.5 !important;
padding: 12px 16px !important;
transition: all 0.2s ease;
}
.stTextArea textarea:focus, .stTextInput input:focus {
background: rgba(255,255,255,0.08) !important;
border-color: #a78bfa !important;
box-shadow: 0 0 0 3px rgba(167,139,250,0.2) !important;
}
/* ββ Sidebar ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
section[data-testid="stSidebar"] {
background: rgba(15, 12, 41, 0.95) !important;
border-right: 1px solid rgba(255,255,255,0.05);
}
/* ββ Button βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.stButton > button {
background: linear-gradient(135deg, #8b5cf6, #3b82f6) !important;
color: white !important;
border: none !important;
border-radius: 12px !important;
font-weight: 600 !important;
font-size: 1.1rem !important;
padding: 0.8rem 2rem !important;
width: 100% !important;
transition: all 0.2s ease !important;
letter-spacing: 0.02em;
box-shadow: 0 4px 14px rgba(59, 130, 246, 0.3) !important;
}
.stButton > button:hover {
transform: translateY(-2px) !important;
box-shadow: 0 6px 20px rgba(59, 130, 246, 0.4) !important;
}
.stButton > button:active {
transform: translateY(0px) !important;
}
/* ββ Selectbox ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.stSelectbox > div > div {
background: rgba(255,255,255,0.05) !important;
border: 1px solid rgba(255,255,255,0.1) !important;
border-radius: 10px !important;
color: #e2e8f0 !important;
}
/* ββ Metric card ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.metric-card {
background: rgba(255,255,255,0.03);
border: 1px solid rgba(255,255,255,0.08);
border-radius: 12px;
padding: 16px;
text-align: center;
transition: transform 0.2s ease;
}
.metric-card:hover {
transform: translateY(-2px);
background: rgba(255,255,255,0.05);
}
.metric-value {
font-size: 1.4rem;
font-weight: 700;
color: #e2e8f0;
}
.metric-desc {
font-size: 0.75rem;
color: #94a3b8;
margin-top: 6px;
text-transform: uppercase;
letter-spacing: 0.05em;
font-weight: 600;
}
/* ββ Status pill ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ */
.status-pill {
display: inline-block;
padding: 6px 14px;
border-radius: 999px;
font-size: 0.75rem;
font-weight: 700;
letter-spacing: 0.05em;
}
.status-ok { background: rgba(52,211,153,0.1); color: #34d399; border: 1px solid rgba(52,211,153,0.3); }
.status-warn { background: rgba(251,191,36,0.1); color: #fbbf24; border: 1px solid rgba(251,191,36,0.3); }
</style>
""", unsafe_allow_html=True)
# ββ HTML render helpers βββββββββββββββββββββββββββββββββββββββββββββββββββββββ
def render_prob_bars(probs: np.ndarray, highlight_idx: int) -> str:
html = ""
for i, p in enumerate(probs):
width = max(int(p * 100), 1)
color = BAR_COLORS[i]
bold = "font-weight:800;" if i == highlight_idx else "font-weight:600;"
label_color = color if i == highlight_idx else "#94a3b8"
html += f"""
<div class="prob-row">
<span class="prob-label" style="color:{label_color};{bold}">{ID2LABEL[i]}</span>
<div class="prob-track">
<div class="prob-fill" style="width:{width}%;background:{color};"></div>
</div>
<span class="prob-pct">{p*100:.1f}%</span>
</div>"""
return html
def render_top3_chips(probs: np.ndarray) -> str:
top3 = np.argsort(-probs)[:3]
chip_classes = ["chip chip-1", "chip chip-2", "chip chip-3"]
rank_labels = ["π₯", "π₯", "π₯"]
html = ""
for rank, idx in enumerate(top3):
html += f'<span class="{chip_classes[rank]}">{rank_labels[rank]} {ID2LABEL[idx]} {probs[idx]*100:.1f}%</span>'
return html
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# SIDEBAR
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
with st.sidebar:
st.markdown("### π Quick Examples")
st.selectbox(
"Load an example question",
options=["(none)"] + [f"Example {i+1}" for i in range(len(EXAMPLES))],
key="ex_choice_box",
on_change=load_example,
label_visibility="collapsed"
)
st.divider()
st.markdown("### βΉοΈ About the Model")
source_label = "HF Hub" if MODEL_SOURCE == "hub" else "Local"
hw_label = "π’ GPU" if DEVICE == "cuda" else "π‘ CPU"
st.markdown(f"""
<div style="display:flex; flex-direction:column; gap:12px; margin-bottom:20px;">
<div class="metric-card">
<div class="metric-value">RoBERTa Base</div>
<div class="metric-desc">Architecture</div>
</div>
<div style="display:flex; gap:12px;">
<div class="metric-card" style="flex:1;">
<div class="metric-value">r=8</div>
<div class="metric-desc">LoRA Rank</div>
</div>
<div class="metric-card" style="flex:1;">
<div class="metric-value">3.4MB</div>
<div class="metric-desc">Adapter</div>
</div>
</div>
</div>
<div style="text-align:center;">
<span class="status-pill {'status-ok' if DEVICE=='cuda' else 'status-warn'}">
{hw_label} {DEVICE.upper()}
</span>
</div>
""", unsafe_allow_html=True)
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# MAIN PAGE
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.markdown('<div class="hero-title">π§ Smart MCQ Solver</div>', unsafe_allow_html=True)
st.markdown(
'<div class="hero-sub">Powered by LoRA fine-tuned RoBERTa Β· '
'IITM DL+GenAI T2-2026</div>',
unsafe_allow_html=True,
)
# ββ Load model (cached across reruns) βββββββββββββββββββββββββββββββββββββββββ
@st.cache_resource(show_spinner=False)
def _cached_load():
return load_roberta_model()
with st.spinner("β³ Loading model weights into memory..."):
try:
rob_model, rob_tok = _cached_load()
except Exception as err:
st.error(f"β Failed to load model: {err}")
st.stop()
# βββ Two-column layout ββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
left, right = st.columns([1.1, 0.9], gap="large")
# ββ INPUT COLUMN ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
with left:
st.markdown('<div class="glass-card">', unsafe_allow_html=True)
st.markdown('<div class="section-header">π The Question</div>', unsafe_allow_html=True)
st.text_area(
"Question / Prompt",
height=110,
placeholder="Enter your multiple-choice question here...",
label_visibility="collapsed",
key="q_prompt",
)
st.markdown("</div>", unsafe_allow_html=True)
# ββ Options βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.markdown('<div class="glass-card">', unsafe_allow_html=True)
st.markdown('<div class="section-header">π€ Options (AβE)</div>', unsafe_allow_html=True)
for lbl in ["A", "B", "C", "D", "E"]:
st.text_area(
f"Option {lbl}",
height=68,
placeholder=f"Type option {lbl}...",
key=f"opt_{lbl}",
)
st.markdown("</div>", unsafe_allow_html=True)
predict_clicked = st.button("β¨ Predict Answer", use_container_width=True)
# ββ RESULT COLUMN βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
with right:
if predict_clicked:
# Validation
prompt = st.session_state.q_prompt.strip()
opts = {
"A": st.session_state.opt_A.strip(),
"B": st.session_state.opt_B.strip(),
"C": st.session_state.opt_C.strip(),
"D": st.session_state.opt_D.strip(),
"E": st.session_state.opt_E.strip(),
}
missing = []
if not prompt: missing.append("Question")
for k, v in opts.items():
if not v: missing.append(f"Option {k}")
if missing:
st.error(f"β οΈ Please fill in: **{', '.join(missing)}**")
st.stop()
with st.spinner("π§ Thinking..."):
result = predict_mcq(prompt, opts, rob_model, rob_tok)
probs = result["probs"]
top1_letter = result["top1"]
top1_conf = result["confidence"] * 100
# ββ Top-1 prediction card βββββββββββββββββββββββββββββββββββββββββββββ
st.markdown(f"""
<div class="glass-card" style="text-align:center; padding: 40px 20px;">
<div class="pred-label">Top Prediction</div>
<div class="pred-badge">{top1_letter}</div>
<div style="color:#94a3b8; margin-top:8px; font-size:1.05rem;">
Confidence: <b style="color:#a78bfa">{top1_conf:.1f}%</b>
</div>
</div>
""", unsafe_allow_html=True)
# ββ Top-3 chips βββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.markdown('<div class="glass-card">', unsafe_allow_html=True)
st.markdown('<div class="section-header">π Top-3 Ranking</div>', unsafe_allow_html=True)
st.markdown(f"<div style='margin-top:12px;'>{render_top3_chips(probs)}</div>", unsafe_allow_html=True)
st.markdown("</div>", unsafe_allow_html=True)
# ββ Probability bars ββββββββββββββββββββββββββββββββββββββββββββββββββ
top1_idx = int(np.argmax(probs))
st.markdown('<div class="glass-card">', unsafe_allow_html=True)
st.markdown('<div class="section-header">π Detailed Probabilities</div>', unsafe_allow_html=True)
st.markdown(f"<div style='margin-top:16px;'>{render_prob_bars(probs, top1_idx)}</div>", unsafe_allow_html=True)
st.markdown("</div>", unsafe_allow_html=True)
else:
st.markdown("""
<div class="glass-card" style="text-align:center; padding:80px 28px; display:flex; flex-direction:column; justify-content:center; align-items:center; height:100%;">
<div style="font-size:3.5rem; margin-bottom:20px; filter: drop-shadow(0 0 10px rgba(167,139,250,0.5));">π―</div>
<div style="color:#e2e8f0; font-size:1.1rem; font-weight:500; margin-bottom:8px;">Ready for Inference</div>
<div style="color:#94a3b8; font-size:0.95rem; line-height:1.6; max-width:250px;">
Enter your question and options on the left, then click <b style="color:#a78bfa;">Predict Answer</b>.
</div>
<div style="margin-top:24px; padding-top:24px; border-top:1px solid rgba(255,255,255,0.1); color:#64748b; font-size:0.85rem;">
Need a test? Load an example from the sidebar.
</div>
</div>
""", unsafe_allow_html=True)
# ββ Footer ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
st.markdown("""
<div style="text-align:center; color:#475569; font-size:0.8rem; padding: 24px 0; margin-top:20px; border-top:1px solid rgba(255,255,255,0.05);">
Developed for <b>IITM DL+GenAI T2-2026</b>
</div>
""", unsafe_allow_html=True)
|