🔧 Troubleshooting & Compatibility
🔴 1. Seeing a "FutureWarning: rcond" in Terminal?
If you are running on Jetson Nano (Ubuntu 18.04) or older NumPy versions, you might see a red warning text:
FutureWarning: rcond parameter will change to the default of machine precision...
The Fix:
This is harmless, but to make it clean:
- Open kill_shot_demo.py.
- Find
np.linalg.lstsq.
- Change
rcond=None to rcond=-1.
# Before
W_lin, _, _, _ = np.linalg.lstsq(X_lin, Y_target, rcond=None)
# After (Fix for Jetson/Old NumPy)
W_lin, _, _, _ = np.linalg.lstsq(X_lin, Y_target, rcond=-1)
---
## ⚠️ IMPORTANT: Running via SSH (Headless Mode)
If you are running this code on a **Jetson Nano** or **Raspberry Pi** via **SSH** (e.g., using VS Code Remote), `plt.show()` will fail because there is no display attached.
**How to fix it:**
1. Open kill_shot_demo.py.
2. Change the visualization backend at the top:
import matplotlib
matplotlib.use('Agg') # Add this line before importing pyplot
import matplotlib.pyplot as plt
3. Replace the show command at the bottom:
# plt.show() <-- Comment this out
plt.savefig('kill_shot_result.png') # <-- Save instead of show
print("Graph saved as kill_shot_result.png")
4. Check the file list. Your victory graph is now an image file! 🖼️
### 🔴 2. ⏳ VS Code "Connection Timed Out" / "Setting up Server" Hangs
The Reality: The Jetson Nano runs Ubuntu 18.04. The latest VS Code (v1.86+) requires a newer glibc library that Ubuntu 18.04 does not support. You cannot use the latest VS Code.
The Fix (The Time Machine):
1. Downgrade VS Code on your PC/Mac to Version 1.85.
This is the last version that supports the Jetson Nano OS.
2. Disable Auto-Update:
Go to Settings -> Search update -> Set mode to none.
3. Clean up the mess:
SSH into your Jetson (via terminal) and delete the corrupted server files:
rm -rf ~/.vscode-server
4. Connect again. It should work now.
### 🔴 3 🌐 "Host Not Found" / IP Conflicts
Problem: Your PC cannot ping the Jetson Nano (Request timed out), even though both are plugged in.
The Culprit: Subnet Mismatch.
Your PC might be on 192.168.88.x
Your Jetson might be stuck on a static IP 192.168.8.x (Different neighborhood!)
The Fix:
1. Connect a monitor/keyboard to the Jetson directly.
2. Go to Network Settings -> Wired -> IPv4.
3. Change Method to "Automatic (DHCP)" (or set a Static IP in the same range as your PC, e.g., 192.168.88.200).
4. Unplug and replug the Ethernet cable.
5. Check the new IP with ifconfig and use that IP in VS Code.
Note: Getting high-level math to run on low-level hardware is never easy. But once you clear these hurdles, you have a $99 supercomputer. Good luck.
🔧 Troubleshooting & Compatibility
🔴 1. Seeing a "FutureWarning: rcond" in Terminal?
If you are running on Jetson Nano (Ubuntu 18.04) or older NumPy versions, you might see a red warning text:
The Fix:
This is harmless, but to make it clean:
np.linalg.lstsq.rcond=Nonetorcond=-1.