X commited on
Update app.py
Browse files
app.py
CHANGED
|
@@ -234,12 +234,16 @@ def train_ppo(episodes=300, steps_per_update=256, lr=3e-4):
|
|
| 234 |
# ==========================================
|
| 235 |
# 4. HF SPACE DEMO
|
| 236 |
# ==========================================
|
|
|
|
|
|
|
|
|
|
| 237 |
def create_demo():
|
| 238 |
print("🏗️ Обучение PPO-агента в бесконечном мире...")
|
| 239 |
agent, history = train_ppo(episodes=100)
|
| 240 |
agent.eval()
|
| 241 |
|
| 242 |
-
def run_exploration(n_steps=200):
|
|
|
|
| 243 |
env = InfiniteWorld(seed=random.randint(0, 99999))
|
| 244 |
obs = env.reset()
|
| 245 |
hidden = None
|
|
@@ -247,7 +251,7 @@ def create_demo():
|
|
| 247 |
positions = []
|
| 248 |
|
| 249 |
with torch.no_grad():
|
| 250 |
-
for _ in range(n_steps):
|
| 251 |
fig, ax = plt.subplots(figsize=(5, 5))
|
| 252 |
ax.imshow(obs)
|
| 253 |
ax.set_title(f"Pos: {env.agent_pos} | Chunks: {len(env.chunks)}")
|
|
@@ -262,26 +266,54 @@ def create_demo():
|
|
| 262 |
if done:
|
| 263 |
break
|
| 264 |
|
| 265 |
-
# Карта
|
| 266 |
-
|
| 267 |
-
|
| 268 |
-
|
| 269 |
-
|
| 270 |
-
|
| 271 |
-
|
| 272 |
-
|
|
|
|
| 273 |
|
| 274 |
return frames
|
| 275 |
|
|
|
|
| 276 |
with gr.Blocks(title="Infinite Builder Agent") as demo:
|
| 277 |
gr.Markdown("# 🌍 Бесконечный мир: PPO-агент с нуля")
|
| 278 |
gr.Markdown("Агент видит только локальный патч 16×16, имеет GRU-память и исследует процедурный мир.")
|
| 279 |
-
|
| 280 |
-
|
| 281 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
|
| 283 |
return demo
|
| 284 |
|
|
|
|
| 285 |
if __name__ == "__main__":
|
| 286 |
demo = create_demo()
|
| 287 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 234 |
# ==========================================
|
| 235 |
# 4. HF SPACE DEMO
|
| 236 |
# ==========================================
|
| 237 |
+
# ==========================================
|
| 238 |
+
# 4. HF SPACE DEMO (ИСПРАВЛЕНО ДЛЯ HF SPACES)
|
| 239 |
+
# ==========================================
|
| 240 |
def create_demo():
|
| 241 |
print("🏗️ Обучение PPO-агента в бесконечном мире...")
|
| 242 |
agent, history = train_ppo(episodes=100)
|
| 243 |
agent.eval()
|
| 244 |
|
| 245 |
+
def run_exploration(n_steps: int = 200):
|
| 246 |
+
"""Функция с явной типизацией для избежания бага schema в Gradio"""
|
| 247 |
env = InfiniteWorld(seed=random.randint(0, 99999))
|
| 248 |
obs = env.reset()
|
| 249 |
hidden = None
|
|
|
|
| 251 |
positions = []
|
| 252 |
|
| 253 |
with torch.no_grad():
|
| 254 |
+
for _ in range(min(n_steps, 500)): # Ограничиваем макс шаги
|
| 255 |
fig, ax = plt.subplots(figsize=(5, 5))
|
| 256 |
ax.imshow(obs)
|
| 257 |
ax.set_title(f"Pos: {env.agent_pos} | Chunks: {len(env.chunks)}")
|
|
|
|
| 266 |
if done:
|
| 267 |
break
|
| 268 |
|
| 269 |
+
# Карта траектории
|
| 270 |
+
if positions:
|
| 271 |
+
fig2, ax2 = plt.subplots(figsize=(6, 6))
|
| 272 |
+
xs, ys = zip(*positions)
|
| 273 |
+
ax2.scatter(xs, ys, c=range(len(positions)), cmap='viridis', s=1)
|
| 274 |
+
ax2.set_title("Траектория исследования")
|
| 275 |
+
ax2.set_aspect('equal')
|
| 276 |
+
frames.append(fig2)
|
| 277 |
+
plt.close(fig2)
|
| 278 |
|
| 279 |
return frames
|
| 280 |
|
| 281 |
+
# Явное указание типов компонентов устраняет баг schema
|
| 282 |
with gr.Blocks(title="Infinite Builder Agent") as demo:
|
| 283 |
gr.Markdown("# 🌍 Бесконечный мир: PPO-агент с нуля")
|
| 284 |
gr.Markdown("Агент видит только локальный патч 16×16, имеет GRU-память и исследует процедурный мир.")
|
| 285 |
+
|
| 286 |
+
steps_input = gr.Slider(
|
| 287 |
+
minimum=50,
|
| 288 |
+
maximum=500,
|
| 289 |
+
step=50,
|
| 290 |
+
value=200,
|
| 291 |
+
label="Количество шагов исследования"
|
| 292 |
+
)
|
| 293 |
+
btn = gr.Button("▶️ Запустить исследование", variant="primary")
|
| 294 |
+
gallery = gr.Gallery(
|
| 295 |
+
label="Процесс строительства и исследования",
|
| 296 |
+
columns=4,
|
| 297 |
+
height="auto",
|
| 298 |
+
object_fit="contain"
|
| 299 |
+
)
|
| 300 |
+
|
| 301 |
+
btn.click(
|
| 302 |
+
fn=run_exploration,
|
| 303 |
+
inputs=[steps_input],
|
| 304 |
+
outputs=[gallery]
|
| 305 |
+
)
|
| 306 |
|
| 307 |
return demo
|
| 308 |
|
| 309 |
+
|
| 310 |
if __name__ == "__main__":
|
| 311 |
demo = create_demo()
|
| 312 |
+
# КРИТИЧНО: Для HF Spaces обязательно server_name="0.0.0.0"
|
| 313 |
+
# share=False на Spaces, т.к. прокси сам маршрутизирует трафик
|
| 314 |
+
demo.launch(
|
| 315 |
+
server_name="0.0.0.0",
|
| 316 |
+
server_port=7860,
|
| 317 |
+
share=False,
|
| 318 |
+
show_api=False
|
| 319 |
+
)
|