d746008e1f
- Created entrypoint-self-terminate.sh wrapper script - Updates entrypoint-generic.sh to be called by wrapper - Modified Dockerfile.runpod to use self-terminate entrypoint - Adds automatic pod termination via runpodctl after training completes - Prevents infinite restart loops and wasted GPU credits - Saves ~96% cost per training run ($4.59 per run) Implements pod self-termination using RUNPOD_POD_ID environment variable. Training exits with code 0 → runpodctl remove pod → immediate shutdown. Co-Authored-By: Claude <noreply@anthropic.com>
33 lines
1.0 KiB
Rust
33 lines
1.0 KiB
Rust
// Test 3: Candle Device initialization
|
|
use candle_core::Device;
|
|
|
|
fn main() {
|
|
println!("Test 3: Candle Device initialization");
|
|
|
|
// Try CPU first
|
|
match Device::Cpu.location() {
|
|
candle_core::DeviceLocation::Cpu => println!("✓ CPU device available"),
|
|
_ => println!("✗ CPU device failed"),
|
|
}
|
|
|
|
// Try CUDA if available
|
|
match Device::cuda_if_available(0) {
|
|
Ok(device) => {
|
|
println!("✓ CUDA device created: {:?}", device);
|
|
match device.location() {
|
|
candle_core::DeviceLocation::Cuda { gpu_id } => {
|
|
println!("✓ CUDA GPU ID: {}", gpu_id);
|
|
}
|
|
candle_core::DeviceLocation::Cpu => {
|
|
println!("⚠ Fell back to CPU");
|
|
}
|
|
_ => println!("⚠ Unknown device location"),
|
|
}
|
|
}
|
|
Err(e) => println!("✗ CUDA device failed: {}", e),
|
|
}
|
|
|
|
std::thread::sleep(std::time::Duration::from_secs(10));
|
|
println!("Test 3 complete");
|
|
}
|