Standardization: Applied mandatory inference and validation templates

Dockerfile

@@ -1,5 +1,5 @@
 # Use official lightweight Python image
-# REBUILD_TIMESTAMP: 2026-04-07 18:58 (Final Phase 2 Sync)
+# REBUILD_TIMESTAMP: 2026-04-07 20:10 (Package Structure Fix)
 FROM python:3.10-slim
 
 # Set working directory
@@ -27,4 +27,4 @@ ENV HF_HOME=/tmp/.cache
 
 # Command to run the FastAPI server
 # This calls the main function in server/app.py which starts uvicorn
-CMD ["python", "server/app.py", "--host", "0.0.0.0", "--port", "8000"]
+CMD ["python", "drone_env/server/app.py", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -81,41 +81,39 @@ The framework supports three operational modes:
 The codebase follows a clean separation-of-concerns architecture across four distinct layers:
 
 ```
-drone_env/
-├── core/                        # Simulation Logic Layer
-│   ├── drone.py                 # Movement physics & battery drain
-│   ├── graders.py               # Unified scoring logic (0.01 - 0.99)
-│   ├── grid_generator.py        # Map generation logic
-│   ├── obstacles.py             # Collision & terrain detection
-│   ├── state_manager.py         # Episodic state management
-│   └── tasks.py                 # Mission difficulty configurations
-├── rl/                          # Intelligence Layer
-│   ├── model.py                 # Neural network architecture (DQN)
-│   ├── policy.py                # Action selection policies
-│   └── trainer.py               # Path analytics & learning engine
-├── server/                      # Interface Layer
-│   ├── app.py                   # FastAPI server & Grader discovery
-│   ├── grid_world_environment.py # Main simulation environment
-│   ├── map_generator.py         # Procedural map generation
-│   └── static/                  # Dashboard Assets
-│       ├── index.html           # Interactive dashboard layout
-│       ├── script.js            # Frontend logic & Mission modal
-│       ├── style.css            # Cyan/Amber aesthetic styles
-│       └── favicon.png          # SkyRelic Branding (favicon)
+.
+├── drone_env/                   # Core Package Subdirectory
+│   ├── core/                    # Simulation Logic Layer
+│   │   ├── drone.py             # Movement physics & battery drain
+│   │   ├── graders.py           # Unified scoring logic (0.01 - 0.99)
+│   │   ├── grid_generator.py    # Map generation logic
+│   │   ├── obstacles.py         # Collision & terrain detection
+│   │   ├── state_manager.py     # Episodic state management
+│   │   └── tasks.py             # Mission difficulty configurations
+│   ├── rl/                      # Intelligence Layer
+│   │   ├── model.py             # Neural network architecture (DQN)
+│   │   ├── policy.py            # Action selection policies
+│   │   └── trainer.py           # Path analytics & learning engine
+│   ├── server/                  # Interface Layer
+│   │   ├── app.py               # FastAPI server & Grader discovery
+│   │   ├── grid_world_environment.py # Main simulation environment
+│   │   ├── map_generator.py     # Procedural map generation
+│   │   └── static/              # Dashboard Assets
+│   ├── models.py                # Unified Pydantic data models
+│   ├── client.py                # CLI client for testing
+│   └── __init__.py              # Package initialization
 ├── data/                        # Persistence Layer
 │   ├── memory.json              # Historical episode logs (JSON)
 │   └── train.log                # Neural training logs
 ├── tests/                       # Validation Layer
 │   ├── test_api.py              # Endpoint integration tests
 │   └── test_env.py              # Physics & Grading unit tests
+├── inference.py                 # Standardized Inference Entry Point
+├── validate-submission.sh       # Submission validation script
 ├── openenv.yaml                 # Mission Manifest (Tasks & Graders)
 ├── pyproject.toml               # Python project & dependency config
-├── models.py                    # Unified Pydantic data models
-├── train.py                     # Neural training entry point
-├── inference.py                 # LLM-guided inference entry point
-├── client.py                    # CLI client for testing
 ├── Dockerfile                   # Deployment container manifest
-└── validate-submission.sh       # Submission validation script
+└── README.md                    # Documentation
 ```
 
 ### Component Interaction Flow
@@ -350,7 +348,7 @@ python train.py --task hard_delivery --episodes 5000
 ### Python SDK Client
 
 ```python
-from client import DroneEnvClient
+from drone_env.client import DroneEnvClient
 
 with DroneEnvClient("http://localhost:8000") as client:
     # Check server health
@@ -506,8 +504,8 @@ And a concise per-step user prompt with position, battery, target, and distance.
 ### Build & Run Locally
 
 ```bash
-# Build from the server/ directory
-docker build -t drone-env -f server/Dockerfile .
+# Build from the root directory
+docker build -t drone-env .
 
 # Run with health check
 docker run -p 8000:8000 \
@@ -553,7 +551,7 @@ HEALTHCHECK --interval=30s --timeout=3s \
   CMD curl -f http://localhost:8000/health || exit 1
 
 # Entrypoint
-CMD ["sh", "-c", "cd /app/env && uvicorn server.app:app --host 0.0.0.0 --port 8000"]
+CMD ["python", "drone_env/server/app.py", "--host", "0.0.0.0", "--port", "8000"]
 ```
 
 ---
@@ -567,7 +565,7 @@ spec_version: 1
 name: drone-env
 type: space
 runtime: fastapi
-app: server.app:app
+app: drone_env.server.app:app
 port: 8000
 tasks:
   - id: easy_delivery

data/memory.json

@@ -1,136 +1,4 @@
 [
-  {
-    "task": "drone_env.core.graders:grade_easy",
-    "steps": [
-      {
-        "step": 1,
-        "x": 9,
-        "y": 8,
-        "action": "RIGHT",
-        "reward": -0.25,
-        "battery": 0.9833
-      },
-      {
-        "step": 2,
-        "x": 9,
-        "y": 8,
-        "action": "RIGHT",
-        "reward": -0.25,
-        "battery": 0.9667
-      },
-      {
-        "step": 3,
-        "x": 9,
-        "y": 8,
-        "action": "RIGHT",
-        "reward": -0.25,
-        "battery": 0.95
-      },
-      {
-        "step": 4,
-        "x": 9,
-        "y": 8,
-        "action": "RIGHT",
-        "reward": -0.25,
-        "battery": 0.9333
-      },
-      {
-        "step": 5,
-        "x": 9,
-        "y": 8,
-        "action": "RIGHT",
-        "reward": -0.25,
-        "battery": 0.9167
-      }
-    ],
-    "grid_meta": {
-      "width": 10,
-      "height": 10
-    },
-    "delivery_positions": [
-      [
-        7,
-        3
-      ]
-    ],
-    "deliveries_done": 0,
-    "total_reward": -1.25,
-    "total_steps": 5
-  },
-  {
-    "task": "drone_env.core.graders:grade_easy",
-    "steps": [
-      {
-        "step": 1,
-        "x": 8,
-        "y": 6,
-        "action": "UP",
-        "reward": -0.05,
-        "battery": 0.9833
-      },
-      {
-        "step": 2,
-        "x": 9,
-        "y": 6,
-        "action": "RIGHT",
-        "reward": -0.05,
-        "battery": 0.9667
-      },
-      {
-        "step": 3,
-        "x": 9,
-        "y": 7,
-        "action": "DOWN",
-        "reward": -0.05,
-        "battery": 0.95
-      },
-      {
-        "step": 4,
-        "x": 8,
-        "y": 7,
-        "action": "LEFT",
-        "reward": -0.05,
-        "battery": 0.9333
-      },
-      {
-        "step": 5,
-        "x": 8,
-        "y": 7,
-        "action": "WAIT",
-        "reward": -0.1,
-        "battery": 0.9167
-      },
-      {
-        "step": 6,
-        "x": 8,
-        "y": 6,
-        "action": "UP",
-        "reward": -0.05,
-        "battery": 0.9
-      },
-      {
-        "step": 7,
-        "x": 9,
-        "y": 6,
-        "action": "RIGHT",
-        "reward": -0.05,
-        "battery": 0.8833
-      }
-    ],
-    "grid_meta": {
-      "width": 10,
-      "height": 10
-    },
-    "delivery_positions": [
-      [
-        3,
-        8
-      ]
-    ],
-    "deliveries_done": 0,
-    "total_reward": -0.4,
-    "total_steps": 7
-  },
   {
     "task": "drone_env.core.graders:grade_easy",
     "steps": [
@@ -17192,5 +17060,1121 @@
     "deliveries_done": 5,
     "total_reward": 13.5,
     "total_steps": 37
+  },
+  {
+    "task": "drone_env.core.graders:grade_easy",
+    "steps": [
+      {
+        "step": 1,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.9833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 2,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.9667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 3,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.95,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 4,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.9333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 5,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.9167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 6,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.9,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 7,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 8,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 9,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.85,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 10,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 11,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 12,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 13,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 14,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 15,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.75,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 16,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 17,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 18,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 19,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 20,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 21,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.65,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 22,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 23,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 24,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 25,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 26,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 27,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.55,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 28,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 29,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 30,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 31,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 32,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 33,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.45,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 34,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 35,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 36,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 37,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 38,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 39,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.35,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 40,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 41,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 42,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 43,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 44,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 45,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.25,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 46,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 47,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 48,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 49,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 50,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 51,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.15,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 52,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 53,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 54,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 55,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 56,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 57,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.05,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 58,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 59,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 60,
+        "x": 9,
+        "y": 0,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      }
+    ],
+    "grid_meta": {
+      "width": 10,
+      "height": 10
+    },
+    "delivery_positions": [
+      [
+        0,
+        7
+      ]
+    ],
+    "deliveries_done": 0,
+    "total_reward": 6.1,
+    "total_steps": 60
+  },
+  {
+    "task": "drone_env.core.graders:grade_easy",
+    "steps": [
+      {
+        "step": 1,
+        "x": 8,
+        "y": 7,
+        "action": "LEFT",
+        "reward": 0.1,
+        "battery": 0.9833,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 2,
+        "x": 8,
+        "y": 8,
+        "action": "DOWN",
+        "reward": 0.1,
+        "battery": 0.9667,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 3,
+        "x": 7,
+        "y": 8,
+        "action": "LEFT",
+        "reward": 0.1,
+        "battery": 0.95,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 4,
+        "x": 8,
+        "y": 8,
+        "action": "RIGHT",
+        "reward": 0.1,
+        "battery": 0.9333,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 5,
+        "x": 7,
+        "y": 8,
+        "action": "LEFT",
+        "reward": 0.1,
+        "battery": 0.9167,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 6,
+        "x": 8,
+        "y": 8,
+        "action": "RIGHT",
+        "reward": 0.1,
+        "battery": 0.9,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 7,
+        "x": 7,
+        "y": 8,
+        "action": "LEFT",
+        "reward": 0.1,
+        "battery": 0.8833,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 8,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 9,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.85,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 10,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 11,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 12,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.8,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 13,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 14,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 15,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.75,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 16,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 17,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 18,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.7,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 19,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 20,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 21,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.65,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 22,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 23,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 24,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.6,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 25,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 26,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 27,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.55,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 28,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 29,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 30,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.5,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 31,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 32,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 33,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.45,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 34,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 35,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 36,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.4,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 37,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 38,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 39,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.35,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 40,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 41,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 42,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.3,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 43,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 44,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 45,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.25,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 46,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 47,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 48,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.2,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 49,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 50,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 51,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.15,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 52,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 53,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1167,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 54,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.1,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 55,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0833,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 56,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0667,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 57,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.05,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 58,
+        "x": 7,
+        "y": 8,
+        "action": "WAIT",
+        "reward": 0.1,
+        "battery": 0.0333,
+        "message": "Drone is idling (WAIT)... \ud83d\udd0b"
+      },
+      {
+        "step": 59,
+        "x": 7,
+        "y": 7,
+        "action": "UP",
+        "reward": 0.1,
+        "battery": 0.0167,
+        "message": "On road \ud83d\udee3\ufe0f"
+      },
+      {
+        "step": 60,
+        "x": 8,
+        "y": 7,
+        "action": "RIGHT",
+        "reward": 0.1,
+        "battery": 0.0,
+        "message": "On road \ud83d\udee3\ufe0f"
+      }
+    ],
+    "grid_meta": {
+      "width": 10,
+      "height": 10
+    },
+    "delivery_positions": [
+      [
+        8,
+        5
+      ]
+    ],
+    "deliveries_done": 0,
+    "total_reward": 6.1,
+    "total_steps": 60
   }
 ]

drone_env/core/__init__.py

@@ -0,0 +1,4 @@
+# drone_env.core package
+from .graders import grade_easy, grade_medium, grade_hard
+
+__all__ = ["grade_easy", "grade_medium", "grade_hard"]

drone_env/rl/__init__.py

@@ -0,0 +1,6 @@
+# drone_env.rl package
+from .model import PathQNet
+from .policy import EpsilonGreedyPolicy
+from .trainer import PathLearner
+
+__all__ = ["PathQNet", "EpsilonGreedyPolicy", "PathLearner"]

{server → drone_env/server}/app.py

@@ -17,24 +17,16 @@ import time
 import json
 
 # Add root to sys.path for local imports
-BASE_DIR = Path(__file__).parent.parent
+BASE_DIR = Path(__file__).parent.parent.parent
 if str(BASE_DIR) not in sys.path:
     sys.path.insert(0, str(BASE_DIR))
 
-# Unified Imports - Root-relative
-# We use try/except to handle different execution environments (uv run vs direct python)
-try:
-    from drone_env.models import DroneAction, DroneObservation, DroneState
-    from drone_env.server.grid_world_environment import DroneDeliveryEnvironment
-    from drone_env.core.tasks import TASK_CONFIG
-    from drone_env.core.graders import GRADERS
-    from drone_env.rl.trainer import PathLearner, get_action_from_policy
-except ImportError:
-    from models import DroneAction, DroneObservation, DroneState
-    from server.grid_world_environment import DroneDeliveryEnvironment
-    from core.tasks import TASK_CONFIG
-    from core.graders import GRADERS
-    from rl.trainer import PathLearner, get_action_from_policy
+# Unified Imports - Canonical Package Paths
+from drone_env.models import DroneAction, DroneObservation, DroneState
+from drone_env.server.grid_world_environment import DroneDeliveryEnvironment
+from drone_env.core.tasks import TASK_CONFIG
+from drone_env.core.graders import GRADERS
+from drone_env.rl.trainer import PathLearner, get_action_from_policy
 
 app = FastAPI(
     title="Drone Delivery OpenEnv",

inference.py

@@ -1,96 +1,98 @@
+"""
+SkyRelic Drone Delivery: Standardized Inference Script
+===================================================
+MANDATORY
+- Environment variables: HF_TOKEN, API_BASE_URL, MODEL_NAME
+- STDOUT FORMAT: [START], [STEP], [END]
+- Participants must use OpenAI Client for all LLM calls.
+"""
+
 import asyncio
 import os
 import textwrap
-from typing import List, Optional
 import sys
+from typing import List, Optional
 from pathlib import Path
 
-# Add core directories to sys.path
+from openai import OpenAI
+
+# Local Project Imports
 ROOT_DIR = Path(__file__).parent
-sys.path.insert(0, str(ROOT_DIR))
+if str(ROOT_DIR) not in sys.path:
+    sys.path.insert(0, str(ROOT_DIR))
 
-# Manual .env loading helper (replaces python-dotenv for simplicity)
+# Load .env for local development convenience
 def load_dotenv(path: Path):
     if path.exists():
-        try:
-            for line in path.read_text().splitlines():
-                line = line.strip()
-                if not line or line.startswith("#"): continue
-                if "=" not in line: continue
-                k, v = line.split("=", 1)
-                os.environ[k.strip()] = v.strip().strip('"').strip("'")
-        except Exception as e:
-            print(f"[DEBUG] .env load error: {e}", file=sys.stderr)
-
+        for line in path.read_text().splitlines():
+            line = line.strip()
+            if not line or line.startswith("#") or "=" not in line: continue
+            k, v = line.split("=", 1)
+            os.environ[k.strip()] = v.strip().strip('"').strip("'")
 load_dotenv(ROOT_DIR / ".env")
 
-from openai import OpenAI
+# Use the established Drone Environment
 from drone_env.server.grid_world_environment import DroneDeliveryEnvironment
-from drone_env.models import DroneAction
+from drone_env.models import DroneAction, DroneObservation
 
-# MANDATORY Environment Variables
-# Defaults are set ONLY for API_BASE_URL and MODEL_NAME as per checklist
-HF_TOKEN = os.getenv("HF_TOKEN")
-API_BASE_URL = os.getenv("API_BASE_URL", "https://router.huggingface.co/v1/")
-MODEL_NAME = os.getenv("MODEL_NAME", "Qwen/Qwen2.5-7B-Instruct")
+# --- Configuration ------------------------------------------------------------
+IMAGE_NAME = os.getenv("IMAGE_NAME") or os.getenv("LOCAL_IMAGE_NAME")
+# Use a placeholder if no key is found to prevent initialization crash during local tests
+API_KEY = os.getenv("HF_TOKEN") or os.getenv("API_KEY") or os.getenv("OPENAI_API_KEY") or "EMPTY_KEY"
 
-# For local testing if no HF_TOKEN is provided but OPENAI_API_KEY is
-OPENAI_KEY = os.getenv("OPENAI_API_KEY") or os.getenv("API_KEY")
+# Defaults are set only for API_BASE_URL and MODEL_NAME as per requirements
+API_BASE_URL = os.getenv("API_BASE_URL") or "https://router.huggingface.co/v1"
+MODEL_NAME = os.getenv("MODEL_NAME") or "Qwen/Qwen2.5-7B-Instruct"
 
-# Final API_KEY selection (Must not have a hardcoded default)
-API_KEY = HF_TOKEN if HF_TOKEN else OPENAI_KEY
-
-# Optional variables
-DRONE_TASK = os.getenv("DRONE_TASK", "drone_env.core.graders:grade_easy")
+TASK_NAME = os.getenv("DRONE_TASK", "drone_env.core.graders:grade_easy")
 BENCHMARK = os.getenv("DRONE_BENCHMARK", "drone_env_v1")
-LOCAL_IMAGE_NAME = os.getenv("LOCAL_IMAGE_NAME") # Optional for Docker testing
 
-# Hyperparameters
 MAX_STEPS = 60
-TEMPERATURE = 0.0 # Set to 0 for deterministic navigation with Qwen
-MAX_TOKENS = 30   # Increased slightly for Qwen reasoning
-SUCCESS_SCORE_THRESHOLD = 0.5  # 50% score for success
+TEMPERATURE = 0.0
+MAX_TOKENS = 50
+SUCCESS_SCORE_THRESHOLD = 0.5
 
 SYSTEM_PROMPT = textwrap.dedent(
     """
-    You are a drone navigation AI. Your goal is to deliver all packages to their destinations.
-    Each step, you will see the drone's position, battery level, current target position, and distance.
-    Actions: UP, DOWN, LEFT, RIGHT, WAIT.
-    Format: Respond with exactly ONE action name in uppercase.
-    Example: UP
+    You are a drone navigation AI. Your goal is to deliver packages to targets.
+    Each turn, you receive the drone's position, battery, target, and distance.
+    Valid Actions: UP, DOWN, LEFT, RIGHT, WAIT.
+    Respond with exactly one action name in uppercase. No quotes.
     """
 ).strip()
 
+# --- Logging Helpers ---------------------------------------------------------
 def log_start(task: str, env: str, model: str) -> None:
     print(f"[START] task={task} env={env} model={model}", flush=True)
 
 def log_step(step: int, action: str, reward: float, done: bool, error: Optional[str]) -> None:
     error_val = error if error else "null"
     done_val = str(done).lower()
+    # Format reward to 2 decimal places as per requirement
     print(
         f"[STEP] step={step} action={action} reward={reward:.2f} done={done_val} error={error_val}",
         flush=True,
     )
 
 def log_end(success: bool, steps: int, score: float, rewards: List[float]) -> None:
+    # Format rewards to 2 decimal places as per requirement
     rewards_str = ",".join(f"{r:.2f}" for r in rewards)
-    # The format required by benchmarks
     print(f"[END] success={str(success).lower()} steps={steps} score={score:.3f} rewards={rewards_str}", flush=True)
 
-def build_user_prompt(obs) -> str:
+# --- Agent Logic -------------------------------------------------------------
+def build_user_prompt(obs: DroneObservation) -> str:
     return textwrap.dedent(
         f"""
         Pos: ({obs.drone_x}, {obs.drone_y})
         Battery: {obs.battery:.2f}
         Target: {obs.current_target}
         Distance: {obs.distance_to_target:.1f}
-        Message: {obs.message}
+        Status: {obs.message}
         Available Actions: UP, DOWN, LEFT, RIGHT, WAIT
-        Action?
         """
     ).strip()
 
-def get_model_action(client: OpenAI, obs) -> str:
+def get_model_action(client: OpenAI, obs: DroneObservation) -> str:
     user_prompt = build_user_prompt(obs)
     try:
         completion = client.chat.completions.create(
@@ -103,69 +105,67 @@ def get_model_action(client: OpenAI, obs) -> str:
             max_tokens=MAX_TOKENS,
             stream=False,
         )
-        action = (completion.choices[0].message.content or "").strip().upper()
-        # Validation for allowed actions
-        if action not in ["UP", "DOWN", "LEFT", "RIGHT", "WAIT"]:
+        text = (completion.choices[0].message.content or "").strip().upper()
+        if text not in ["UP", "DOWN", "LEFT", "RIGHT", "WAIT"]:
             return "WAIT"
-        return action
+        return text
     except Exception as exc:
-        # In case of API failure, log to stderr and return WAIT
         print(f"[DEBUG] Model request failed: {exc}", file=sys.stderr, flush=True)
         return "WAIT"
 
+# --- Run Loop ----------------------------------------------------------------
 async def main() -> None:
-    # Initialize OpenAI client according to benchmarks
     client = OpenAI(base_url=API_BASE_URL, api_key=API_KEY)
-    
-    # Initialize our environment locally
+
+    # Initialize environment locally
+    # Note: the sample template uses docker integration, but for local 
+    # hackathon development, we initialize the DroneDeliveryEnvironment directly.
     env = DroneDeliveryEnvironment()
 
+    history: List[str] = []
     rewards: List[float] = []
     steps_taken = 0
-    score_val = 0.0
+    score = 0.010
     success = False
 
-    log_start(task=DRONE_TASK, env=BENCHMARK, model=MODEL_NAME)
+    log_start(task=TASK_NAME, env=BENCHMARK, model=MODEL_NAME)
 
     try:
         # Initial reset
-        obs = env.reset(DroneAction(task_name=DRONE_TASK))
-        
-        # Use dynamic max_steps from environment config
-        current_max_steps = int(obs.max_steps) if obs.max_steps else MAX_STEPS
+        obs = env.reset(DroneAction(task_name=TASK_NAME))
         
-        for step in range(1, current_max_steps + 1):
+        # Max steps from environment or local constant
+        max_limit = int(obs.max_steps) if obs.max_steps else MAX_STEPS
+
+        for step in range(1, max_limit + 1):
             if obs.done:
                 break
 
-            # Get action from model
             action_str = get_model_action(client, obs)
 
-            # Execution in environment
+            # Step environment
             obs = env.step(DroneAction(direction=action_str))
 
-            reward = obs.reward_last
-            done = obs.done
-            error = None # Error field as per benchmark (normally handled via exceptions)
+            reward = float(obs.reward_last)
+            done = bool(obs.done)
+            error = None
 
             rewards.append(reward)
             steps_taken = step
-            
-            # Step logging
+
             log_step(step=step, action=action_str, reward=reward, done=done, error=error)
 
             if done:
                 break
 
-        # Calculate final metrics
-        score_val = float(obs.score)  # Graded score is already normalized to [0.01, 0.99]
-        success = (obs.deliveries_done == obs.deliveries_total) if obs.deliveries_total > 0 else False
+        # Final score (already normalized by the environment's grader [0.01, 0.99])
+        score = float(obs.score) if obs.score is not None else 0.010
+        success = (obs.deliveries_done == obs.deliveries_total) and obs.deliveries_total > 0
 
-    except Exception as e:
-        print(f"[DEBUG] Error during inference: {e}", file=sys.stderr, flush=True)
     finally:
-        # Always output the [END] line
-        log_end(success=success, steps=steps_taken, score=score_val, rewards=rewards)
+        # Mandatory close and log emission
+        log_end(success=success, steps=steps_taken, score=score, rewards=rewards)
+
 
 if __name__ == "__main__":
     asyncio.run(main())

openenv.yaml

@@ -2,7 +2,7 @@ spec_version: 1
 name: drone_env
 type: space
 runtime: fastapi
-app: server.app:app
+app: drone_env.server.app:app
 port: 8000
 tasks:
   - id: easy_delivery

pyproject.toml

@@ -34,5 +34,4 @@ server = "drone_env.server.app:main"
 
 [tool.setuptools]
 include-package-data = true
-packages = ["drone_env", "drone_env.server", "drone_env.core", "drone_env.rl"]
-package-dir = { "drone_env" = ".", "drone_env.server" = "server", "drone_env.core" = "core", "drone_env.rl" = "rl" }
+packages = ["drone_env", "drone_env.server", "drone_env.core", "drone_env.rl"]

validate-submission.sh

@@ -59,8 +59,7 @@ run_with_timeout() {
 
 portable_mktemp() {
   local prefix="${1:-validate}"
-  # Use current directory for temp file on Windows/bash compatibility
-  mktemp "${prefix}-XXXXXX" 2>/dev/null || mktemp
+  mktemp "${TMPDIR:-/tmp}/${prefix}-XXXXXX" 2>/dev/null || mktemp
 }
 
 CLEANUP_FILES=()