gitadmin/foxhunt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: fix Task 14 path and Task 15 EvaluationEngine API in plan

Browse Source 
- Task 14: evaluate_baseline is at crates/ml/examples/evaluate_baseline.rs
  (not bin/fxt/src/commands/), fix cargo check command and git add path
- Task 15: process_bar_factored takes (usize, &OHLCVBarF32, &FactoredAction)
  not (f64, usize, f64, f64), fix test to use correct types

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
jgrusewski
2026-03-11 09:37:33 +01:00
parent dce318b47f
commit b704af9112
1 changed files with 22 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
docs/plans/2026-03-11-cuda-backtest-gpu-residency-plan.md
View File

@@ -2028,16 +2028,16 @@ bitonic sort. Single download of extended WindowMetrics struct."
Replace CPU backtest path in the standalone evaluation binary.
**Files:**
- Modify: `bin/fxt/src/commands/evaluate_baseline.rs` (the evaluate_baseline binary entry point)
- Test: `SQLX_OFFLINE=true cargo check -p fxt --features cuda`
- Modify: `crates/ml/examples/evaluate_baseline.rs` (declared as `[[example]]` in `crates/ml/Cargo.toml:230`)
- Test: `SQLX_OFFLINE=true cargo check -p ml --example evaluate_baseline`
- [ ] **Step 1: Add GPU evaluation path to evaluate_baseline**
In `bin/fxt/src/commands/evaluate_baseline.rs`, at the evaluation entry point where the
In `crates/ml/examples/evaluate_baseline.rs`, at the evaluation entry point where the
`EvaluationEngine` is currently constructed, add a GPU path before the CPU path:
```rust
use crate::cuda_pipeline::gpu_backtest_evaluator::{GpuBacktestEvaluator, GpuBacktestConfig};
use ml::cuda_pipeline::gpu_backtest_evaluator::{GpuBacktestEvaluator, GpuBacktestConfig};
// Try GPU evaluation first
#[cfg(feature = "cuda")]
@@ -2076,13 +2076,13 @@ if let Ok(device) = Device::new_cuda(0) {
- [ ] **Step 2: Run check**
Run: `SQLX_OFFLINE=true cargo check -p fxt`
Run: `SQLX_OFFLINE=true cargo check -p ml --example evaluate_baseline`
Expected: PASS (compiles with and without cuda feature)
- [ ] **Step 3: Commit**
```bash
git add bin/fxt/src/commands/evaluate_baseline.rs
git add crates/ml/examples/evaluate_baseline.rs
git commit -m "feat(eval): wire GPU backtester into evaluate_baseline binary
Standalone evaluation now uses CUDA backtest kernel when available.
@@ -2109,6 +2109,8 @@ use ml::cuda_pipeline::gpu_backtest_evaluator::{
GpuBacktestConfig, GpuBacktestEvaluator, WindowMetrics,
};
use ml_dqn::evaluation::engine::EvaluationEngine;
use ml_dqn::evaluation::metrics::OHLCVBarF32;
use ml_core::common::action::FactoredAction;
use candle_core::{Device, Tensor, DType};
/// Generate deterministic synthetic price data (random walk with drift).
@@ -2184,15 +2186,22 @@ async fn test_gpu_vs_cpu_backtest_agreement() {
).expect("GPU evaluation");
// 3. Run CPU backtest with same always-Flat action
use ml_core::common::action::{ExposureLevel, OrderType, Urgency};
let flat_action = FactoredAction::new(ExposureLevel::Flat, OrderType::Market, Urgency::Normal);
let mut cpu_engines: Vec<EvaluationEngine> = vec![
EvaluationEngine::new(100_000.0),
EvaluationEngine::new(100_000.0),
EvaluationEngine::new_with_kelly(100_000.0, 1.0),
EvaluationEngine::new_with_kelly(100_000.0, 1.0),
];
let all_closes = [&closes_1, &closes_2];
for (engine, closes) in cpu_engines.iter_mut().zip(all_closes.iter()) {
for close in closes.iter() {
// Action 2 = Flat (no trading)
engine.process_bar_factored(*close, 2, 0.1, 0.0001);
let all_prices = [&prices_1, &prices_2];
for (engine, window_prices) in cpu_engines.iter_mut().zip(all_prices.iter()) {
for (bar_idx, ohlc) in window_prices.iter().enumerate() {
let bar = OHLCVBarF32 {
timestamp: bar_idx as i64,
open: ohlc[0], high: ohlc[1], low: ohlc[2], close: ohlc[3],
volume: 1000.0,
};
engine.process_bar_factored(bar_idx, &bar, &flat_action);
}
}