Workflow to ensure codebase integrity

.github/workflows/restore_cost_benefit.yml

@@ -0,0 +1,97 @@
+name: Restore src/cost_benefit.py on every commit
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  restore-file:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v3
+
+      - name: Recreate src/cost_benefit.py
+        run: |
+          mkdir -p src
+          cat <<EOF > src/cost_benefit.py
+import argparse
+
+import subprocess
+
+import time
+
+import requests
+
+def detect_available_budget(runtime_env: str) -> int:
+    import torch
+    if "local" in runtime_env and torch.cuda.is_available():
+        total_vram_mb = torch.cuda.get_device_properties(0).total_memory // (1024 ** 2)
+        return min(total_vram_mb, 100)
+    else:
+        return 100
+
+def get_best_model(runtime_env: str, use_local_only=False, use_api_only=False) -> dict:
+    static_costs = {
+        "llama3.2": {"size": 20, "token_cost": 0.0001, "tokens_sec": 30, "type": "local"},
+        "mistral": {"size": 40, "token_cost": 0.0002, "tokens_sec": 50, "type": "local"},
+        "gemini-2.0-flash": {"size": 60, "token_cost": 0.0005, "tokens_sec": 60, "type": "api"},
+        "gemini-2.5-pro-preview-03-25": {"size": 80, "token_cost": 0.002, "tokens_sec": 45, "type": "api"}
+    }
+
+    def detect_available_budget(runtime_env: str) -> int:
+        import torch
+        if "local" in runtime_env and torch.cuda.is_available():
+            total_vram_mb = torch.cuda.get_device_properties(0).total_memory // (1024 ** 2)
+            return min(total_vram_mb, 100)
+        else:
+            return 100
+
+    budget = detect_available_budget(runtime_env)
+
+    best_model = None
+    best_speed = -1
+
+    for model, info in static_costs.items():
+        if info["size"] > budget:
+            continue
+        if use_local_only and info["type"] != "local":
+            continue
+        if use_api_only and info["type"] != "api":
+            continue
+        if info["tokens_sec"] > best_speed:
+            best_model = model
+            best_speed = info["tokens_sec"]
+
+    if not best_model:
+        return {
+            "model": "llama3.2",
+            "token_cost": static_costs["llama3.2"]["token_cost"],
+            "tokens_sec": static_costs["llama3.2"]["tokens_sec"],
+            "note": "Defaulted due to no models fitting filters"
+        }
+
+    return {
+        "model": best_model,
+        "token_cost": static_costs[best_model]["token_cost"],
+        "tokens_sec": static_costs[best_model]["tokens_sec"]
+    }
+EOF
+
+      - name: Commit and push if src/cost_benefit.py changed
+        run: |
+          git config user.name "github-actions"
+          git config user.email "github-actions@github.com"
+
+          if ! git diff --quiet src/cost_benefit.py; then
+            git add src/cost_benefit.py
+            git commit -m "Auto-restore src/cost_benefit.py"
+            git push
+          else
+            echo "No changes to src/cost_benefit.py"
+          fi