Download 360 DBN files (36.3 MB) using Rust databento client

- Created data/examples/download_ml_training_data.rs using reqwest + Databento HTTP API
- Downloaded 90 days × 4 symbols (ES.FUT, NQ.FUT, ZN.FUT, 6E.FUT)
- Files saved to test_data/real/databento/ml_training/
- Total: 360 files, 15 MB compressed DBN format
- Used existing Rust pattern from download_nq_fut.rs
- API key loaded from .env file
- 100% success rate (360/360 files)
- Ready for ML training benchmarks

Next: Create simplified training benchmark for RTX 3050 Ti GPU measurements

check_subscription.py

@@ -0,0 +1,92 @@
+#!/usr/bin/env python3
+"""
+Check Databento subscription and data availability.
+"""
+import os
+import databento as db
+import sys
+
+def main():
+    api_key = os.environ.get("DATABENTO_API_KEY")
+    if not api_key:
+        print("ERROR: DATABENTO_API_KEY environment variable not set")
+        sys.exit(1)
+
+    client = db.Historical(api_key)
+
+    print("=" * 70)
+    print("Databento Subscription Check")
+    print("=" * 70)
+
+    # Get billing information
+    try:
+        print("\n1. Checking Account Status...")
+        # Note: There's no direct API for billing, but we can try to get cost for a known symbol
+        # Let's try ES (S&P 500 E-mini) which is very common
+
+        test_cases = [
+            ("XNAS.ITCH", "AAPL", "trades", "Nasdaq stocks"),
+            ("GLBX.MDP3", "ES", "ohlcv-1m", "CME E-mini S&P 500"),
+            ("GLBX.MDP3", "ES.FUT", "ohlcv-1m", "CME E-mini S&P 500 (FUT)"),
+            ("GLBX.MDP3", "ESH24", "ohlcv-1m", "CME E-mini S&P 500 (Mar 2024)"),
+        ]
+
+        print("\n2. Testing Data Access...\n")
+
+        for dataset, symbol, schema, description in test_cases:
+            print(f"Testing: {description}")
+            print(f"  Dataset: {dataset}, Symbol: {symbol}, Schema: {schema}")
+
+            try:
+                cost = client.metadata.get_cost(
+                    dataset=dataset,
+                    symbols=[symbol],
+                    schema=schema,
+                    start="2024-01-02",
+                    end="2024-01-05"
+                )
+                print(f"  ✅ ACCESS GRANTED - Cost: ${cost:.4f} (3 days)")
+            except Exception as e:
+                error_msg = str(e)
+                if "401" in error_msg or "unauthorized" in error_msg.lower():
+                    print(f"  ❌ UNAUTHORIZED - Need subscription")
+                elif "404" in error_msg or "not found" in error_msg.lower():
+                    print(f"  ❌ NOT FOUND - Symbol doesn't exist")
+                elif "symbology" in error_msg.lower():
+                    print(f"  ❌ SYMBOLOGY ERROR - Symbol can't be resolved")
+                else:
+                    print(f"  ❌ ERROR: {error_msg[:80]}")
+            print()
+
+        print("=" * 70)
+        print("Diagnosis")
+        print("=" * 70)
+        print("""
+Based on the results above:
+
+1. If ALL tests show UNAUTHORIZED:
+   → Your subscription doesn't include historical data access
+   → You may only have live data access
+
+2. If ONLY GLBX.MDP3 (CME) tests fail:
+   → Your subscription doesn't include CME futures data
+   → CME data requires a separate subscription tier
+
+3. If SYMBOLOGY ERROR appears:
+   → The symbol format is incorrect for that dataset
+   → Try checking Databento's symbology guide
+
+4. If some symbols work:
+   → Your subscription is working, but specific instruments need adjustment
+
+Recommendations:
+- Check your Databento account subscription at: https://databento.com/account
+- For CME futures data (6E, ES, etc.), verify you have GLBX.MDP3 access
+- Contact Databento support if you believe you should have access
+        """)
+
+    except Exception as e:
+        print(f"\nError checking subscription: {e}")
+
+if __name__ == "__main__":
+    main()