feat(train): cross-Re transfer pipeline — re60/re200/re400 calibrations + script
- Add crossre_transfer.sh: calibrate → transfer-train for re60→re200→re400 - Add re60 config (ν=0.006667, SI=800, uniform+free-slip, very weak shedding) - Calibrate re60, re200, re400: FORCE_SCALE, SENS_SCALE, dtw_norm_scale, SIM_BP - Fix all paths: use DynamisLab submodule CelerisLab, remove external ~/CelerisLab - Remove _clean_cache() from envs/calibrate — CelerisLab handles internally - Move V4 backups to old/: env_karman_2000x600, train_karman_2000x600, etc. - train_karman.py: save model + vecnormalize every episode (non-optional) - Update TRAIN_KNOWLEDGE.md: file structure, calibration table, cross-re guide - All 3 Re verified: 5-episode transfer test passed (re60: 0.64, re200: 0.43, re400: 0.49) Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
Frank14f
parent
b3ee72e144
commit
5f061bec06
50
configs/config_lbm_karman_2000x600_re60.json
Normal file
50
configs/config_lbm_karman_2000x600_re60.json
Normal file
@ -0,0 +1,50 @@
|
||||
{
|
||||
"_doc": "Karman Cloak Re60: uniform inlet, free-slip walls, 2000x600 grid. nu = U0*2D/60 = 0.006667.",
|
||||
"grid": {
|
||||
"lattice_model": "D2Q9",
|
||||
"nx": 2000,
|
||||
"ny": 600,
|
||||
"nz": 1
|
||||
},
|
||||
"physics": {
|
||||
"data_type": "FP32",
|
||||
"viscosity": 0.006667,
|
||||
"velocity": 0.01,
|
||||
"rho": 1.0
|
||||
},
|
||||
"method": {
|
||||
"collision": "MRT",
|
||||
"streaming": "double_buffer",
|
||||
"store_precision": "FP32",
|
||||
"ddf_shifting": false,
|
||||
"les": {
|
||||
"enabled": false,
|
||||
"cs": 0.16,
|
||||
"closed_form": true
|
||||
},
|
||||
"trt": {
|
||||
"magic_param": 0.1875
|
||||
},
|
||||
"inlet": {
|
||||
"profile": "uniform",
|
||||
"scheme": "regularized",
|
||||
"trt_neq_damp": 0.5,
|
||||
"regularized_neq_damp": 0.5
|
||||
},
|
||||
"outlet": {
|
||||
"mode": "neq_extrap",
|
||||
"backflow_clamp": true,
|
||||
"blend_alpha": 0.7,
|
||||
"srt_neq_damp": 0.5
|
||||
},
|
||||
"y_wall_bc": "free_slip",
|
||||
"omega_guard": {
|
||||
"min": 0.01,
|
||||
"max": 1.99
|
||||
}
|
||||
},
|
||||
"cuda": {
|
||||
"threads_per_block": 256,
|
||||
"compute_capability": "auto"
|
||||
}
|
||||
}
|
||||
@ -1,7 +1,10 @@
|
||||
# Karman Cloak Training — Knowledge Document (V5)
|
||||
|
||||
> **V5 (2026-07-01)**: Parameterized, calibration-driven, no_bias only, multi-GPU ready.
|
||||
> Original V4 files preserved as `env_karman_2000x600.py` and `train_karman_2000x600.py`.
|
||||
> **V5 (2026-07-03)**: Parameterized, calibration-driven, no_bias only.
|
||||
> All paths use DynamisLab submodule `CelerisLab/` (no external dev dir).
|
||||
> Cross-Re transfer pipeline verified: re60, re200, re400.
|
||||
> Every episode saves model checkpoint.
|
||||
> V4 backups moved to `old/`.
|
||||
|
||||
---
|
||||
|
||||
@ -12,257 +15,199 @@ hydrodynamic cloaking — making the downstream flow match the "undisturbed" flo
|
||||
(as if the pinball weren't there). The upstream disturbance cylinder generates a
|
||||
Kármán vortex street; the pinball must cancel it.
|
||||
|
||||
- **CFD**: CelerisLab LBM solver, D2Q9 MRT, 2000x600 grid, uniform inlet, free-slip walls
|
||||
- **CFD**: CelerisLab LBM solver (DynamisLab submodule), D2Q9 MRT, 2000x600 grid, uniform inlet, free-slip walls
|
||||
- **DRL**: PPO with Sin activation, 64x64 MLP, SB3 + VecNormalize
|
||||
- **V5 mode**: No_bias only (ACTION_SCALE=12, ACTION_BIAS=[0,0,0]). All cases use calibration-first workflow.
|
||||
|
||||
## 0. V5 Quick Start (Server Deployment)
|
||||
## 0. Quick Start
|
||||
|
||||
### 0.1 Single-Re training (re100)
|
||||
|
||||
```bash
|
||||
# 1. Calibrate (once per case, ~5 min on single GPU)
|
||||
# 1. Calibrate (~5 min)
|
||||
cd src/drl_pinball/train
|
||||
conda run -n pycuda_3_10 python calibrate.py \
|
||||
--case re100 --device-id 0 \
|
||||
--config configs/config_lbm_karman_2000x600.json
|
||||
|
||||
# 2. Launch multi-seed training on server (sequential, 7 min between GPUs)
|
||||
# 2. Multi-seed training (server, sequential 7-min delay between GPUs)
|
||||
bash launch_multi.sh \
|
||||
--case-name re100_karman --seeds 42,43,44,45,46,47 \
|
||||
--gpus 0,1,2,3,4,5 --episodes 500 \
|
||||
--config configs/config_lbm_karman_2000x600.json \
|
||||
--calibration calibrations/re100/calibration.json
|
||||
|
||||
# 3. Transfer learning to another Re
|
||||
conda run -n pycuda_3_10 python calibrate.py \
|
||||
--case re200 --device-id 0 --si 500 \
|
||||
--config configs/config_lbm_karman_2000x600_re200.json
|
||||
|
||||
bash launch_multi.sh \
|
||||
--case-name re200_karman --seeds 42,43,44 --gpus 0,1,2 \
|
||||
--config configs/config_lbm_karman_2000x600_re200.json \
|
||||
--calibration calibrations/re200/calibration.json \
|
||||
--transfer output/re100_karman_seed42/models/best_model.zip
|
||||
|
||||
# 4. Monitor
|
||||
# 3. Monitor
|
||||
tail -f output/re100_karman_seed42/train.log
|
||||
tensorboard --logdir output/re100_karman_seed42/tb --port 6006
|
||||
```
|
||||
|
||||
### 0.2 Cross-Re transfer (re60, re200, re400)
|
||||
|
||||
```bash
|
||||
# Local test (5 episodes each, quick verification):
|
||||
bash crossre_transfer.sh --re-list 60,200,400 --test-episodes 5
|
||||
|
||||
# Server production (200 episodes each):
|
||||
# !! BEFORE PUSHING TO SERVER: update BEST_MODEL path in crossre_transfer.sh !!
|
||||
bash crossre_transfer.sh --re-list 60,200,400
|
||||
# Then push to server and run there.
|
||||
```
|
||||
|
||||
### 0.3 Path notes for server deployment
|
||||
|
||||
`crossre_transfer.sh` has relative paths via `SCRIPT_DIR`/`REPO_DIR`.
|
||||
Only `BEST_MODEL` needs updating — point to the best Re100 model from
|
||||
multi-seed training (e.g. `output/re100_karman_seed42/models/best_model.zip`).
|
||||
|
||||
---
|
||||
|
||||
## 2. File Structure (V5)
|
||||
## 2. File Structure (V5 Final)
|
||||
|
||||
```
|
||||
train/
|
||||
├── __init__.py
|
||||
│
|
||||
├── # V5 ACTIVE FILES (parameterized, no_bias, calibration-driven)
|
||||
│
|
||||
├── # ACTIVE FILES
|
||||
├── calibrate.py # Phase 0 calibration (produces calibration.json + target.npy)
|
||||
├── env_karman.py # Parameterized Karman cloak env (loads calibration JSON)
|
||||
├── env_karman.py # Parameterized Karman cloak env
|
||||
├── env_illusion.py # Parameterized Illusion env
|
||||
├── env_vortex.py # Vortex cloak env (lamb/taylor, MAX_STEPS=150)
|
||||
├── train_karman.py # Parameterized training script (--calibration, --config)
|
||||
├── train_karman.py # Parameterized training script (every ep saves model)
|
||||
├── train_illusion.py # Illusion training script
|
||||
├── launch_multi.sh # Sequential multi-GPU server launcher
|
||||
│
|
||||
├── # V5 SUPPORT FILES (unchanged from V4)
|
||||
│
|
||||
├── crossre_transfer.sh # Cross-Re transfer: calibrate + train (re60→re200→re400)
|
||||
├── symmetry_wrapper.py # G-mirror symmetry augmentation (per-rollout)
|
||||
├── phase0_baseline_measure.py # Legacy baseline measurement tool (reference)
|
||||
├── visualize_and_analyze.py # Flow-field visualization & analysis
|
||||
├── SERVER_DEPLOY.md # Server deployment instructions
|
||||
├── TRAIN_KNOWLEDGE.md # This file
|
||||
│
|
||||
├── # V4 BACKUP FILES (preserved, NOT active)
|
||||
├── old/ # Archived V4 files (NOT active)
|
||||
│ ├── env_karman_2000x600.py
|
||||
│ ├── train_karman_2000x600.py
|
||||
│ ├── phase0_baseline_measure.py
|
||||
│ └── analyze_final.py
|
||||
│
|
||||
├── env_karman_2000x600.py # Original V4 env (hardcoded FORCE_SCALE, bias support)
|
||||
├── train_karman_2000x600.py # Original V4 training (--no-bias flag)
|
||||
├── analyze_final.py # V3 training curve plotting (legacy paths)
|
||||
│
|
||||
├── # Calibration & output
|
||||
│
|
||||
├── calibrations/ # Per-case calibration files
|
||||
│ └── re100/
|
||||
│ ├── calibration.json # FORCE_SCALE, SENS_SCALE, SIM_BP/VAL, etc.
|
||||
│ ├── target.npy # Target sensor signals (FIFO_LEN, 6)
|
||||
│ └── calibrate.log # Calibration run log
|
||||
├── calibrations/ # Per-case calibration files (IMMUTABLE)
|
||||
│ ├── re60/
|
||||
│ │ ├── calibration.json # SI=800, FORCE_SCALE=0.0021, SENS_SCALE=0.72
|
||||
│ │ ├── target.npy
|
||||
│ │ └── calibrate.log
|
||||
│ ├── re100/
|
||||
│ │ ├── calibration.json # SI=800, FORCE_SCALE=0.0024, SENS_SCALE=0.75
|
||||
│ │ └── calibrate.log
|
||||
│ ├── re200/
|
||||
│ │ ├── calibration.json # SI=500, FORCE_SCALE=0.0026, SENS_SCALE=0.90
|
||||
│ │ ├── target.npy
|
||||
│ │ └── calibrate.log
|
||||
│ ├── re400/
|
||||
│ │ ├── calibration.json # SI=400, FORCE_SCALE=0.0042, SENS_SCALE=0.98
|
||||
│ │ ├── target.npy
|
||||
│ │ └── calibrate.log
|
||||
│ └── illusion_1L/
|
||||
│ ├── calibration.json
|
||||
│ ├── target.npy
|
||||
│ ├── target_harmonics.json
|
||||
│ └── calibrate.log
|
||||
│
|
||||
└── output/ # Training outputs
|
||||
└── re100_karman_seed42/
|
||||
├── models/ # best_model.zip, final_model.zip, chkpt_ep*.zip
|
||||
└── <case>_seed<N>/
|
||||
├── models/ # ep0001_model.zip, ..., best_model.zip, final_model.zip
|
||||
├── tb/ # TensorBoard logs
|
||||
├── train.log # Training run log
|
||||
├── calibration.json # Copy of calibration used
|
||||
├── vec_normalize.pkl # VecNormalize statistics
|
||||
└── meta.json # Run metadata
|
||||
├── train.log
|
||||
├── calibration.json
|
||||
├── vec_normalize.pkl
|
||||
└── meta.json
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 3. Calibration Workflow (V5 — ALWAYS RUN FIRST)
|
||||
## 3. Calibration Workflow (ALWAYS RUN FIRST)
|
||||
|
||||
Every case MUST run `calibrate.py` before training. This produces:
|
||||
|
||||
- `calibration.json`: FORCE_SCALE, SENS_SCALE, dtw_norm_scale, SIM_BP, SIM_VAL, reward constants
|
||||
- `target.npy`: Target sensor signals (150 steps x 6 channels, legacy-equiv)
|
||||
- `target.npy`: Target sensor signals (150 steps x 6 channels)
|
||||
|
||||
The calibration measures Stage0 (zero rotation) and Stage1 (open-loop reference action
|
||||
equivalent to legacy [0,-4,4]*U0 bias) to compute:
|
||||
- FORCE_SCALE = combined max|force| across both stages
|
||||
- SENS_SCALE = combined max|sensor| (legacy-equiv) across both stages
|
||||
- dtw_norm_scale = target's 3 uy-channel avg std
|
||||
- SIM_BP/SIM_VAL = piecewise DTW mapping from Stage0/Stage1 sim mean values
|
||||
The calibration measures Stage0 (zero rotation) and Stage1 (open-loop reference) to compute normalization
|
||||
constants. Calibration is **IMMUTABLE** — once produced, never modify.
|
||||
|
||||
Calibration is IMMUTABLE — once produced, never modify. Training and inference use it as-is.
|
||||
### Calibration Results (all Re)
|
||||
|
||||
| Case | SI | FORCE_SCALE | SENS_SCALE | dtw_norm_scale | Stage0 sim | Stage1 sim |
|
||||
|------|-----|-------------|------------|----------------|-----------|-----------|
|
||||
| re60 | 800 | 0.0021 | 0.72 | 0.107 | 0.41 | 0.61 |
|
||||
| re100 | 800 | 0.0024 | 0.75 | 0.204 | 0.32 | 0.82 |
|
||||
| re200 | 500 | 0.0026 | 0.90 | 0.269 | 0.45 | 0.77 |
|
||||
| re400 | 400 | 0.0042 | 0.98 | 0.310 | 0.56 | 0.73 |
|
||||
|
||||
### Cross-Re SI guidance
|
||||
|
||||
Based on ~18 samples per vortex shedding cycle:
|
||||
| Case | SI | Rationale |
|
||||
|------|----|-----------|
|
||||
| re50 | 1600 | Lower Re, longer period |
|
||||
| re60 | 800 | FFT shows very weak/absent shedding at this Re with free-slip |
|
||||
| re100 | 800 | Verified (~19 samples/cycle) |
|
||||
| re200 | 500 | ~18 samples/cycle |
|
||||
| re400 | 400 | ~18 samples/cycle |
|
||||
|
||||
---
|
||||
### re60 note
|
||||
|
||||
## 4. Calibration Results (re100)
|
||||
|
||||
```
|
||||
FORCE_SCALE = 0.002429
|
||||
SENS_SCALE = 0.7543
|
||||
dtw_norm_scale = 0.2043
|
||||
Stage0 sim_mean = 0.3166 (-> SIM_VAL=0.2)
|
||||
Stage1 sim_mean = 0.8158 (-> SIM_VAL=0.5)
|
||||
SIM_BP = [0.0, 0.3166, 0.8158, 0.8895, 0.9448, 1.0]
|
||||
SIM_VAL = [0.0, 0.2, 0.5, 0.8, 0.9, 1.0]
|
||||
```
|
||||
|
||||
Verified: env with calibration gives zero-action reward = 0.076 (Stage0 level, correct).
|
||||
At Re=60 with uniform inlet + free-slip walls, the upstream disturbance cylinder
|
||||
produces very weak periodic shedding (FFT dominant period ~5625 samples).
|
||||
This is a different regime from legacy parabolic+no-slip where re50 did shed.
|
||||
SI=800 is adequate; the DRL essentially learns a steady-state control policy.
|
||||
|
||||
---
|
||||
|
||||
## 5. Cross-Re Configs
|
||||
|
||||
All share the same 2000x600 grid, uniform inlet, free-slip walls. Only viscosity differs:
|
||||
- `configs/config_lbm_karman_2000x600.json` (v=0.004, Re=100)
|
||||
- `configs/config_lbm_karman_2000x600_re50.json` (v=0.008, Re=50)
|
||||
- `configs/config_lbm_karman_2000x600_re200.json` (v=0.002, Re=200)
|
||||
- `configs/config_lbm_karman_2000x600_re400.json` (v=0.001, Re=400)
|
||||
|
||||
---
|
||||
|
||||
## 6. Original "How to Start Training" (V4 — kept for reference)
|
||||
|
||||
### Prerequisites
|
||||
- conda env `pycuda_3_10` with PyCUDA, PyTorch, SB3, tensorboard
|
||||
- CelerisLab installed at `/home/frank14f/CelerisLab`
|
||||
- `target.npy` in the train directory (pre-recorded, reusable)
|
||||
|
||||
### Starting a training run (CRITICAL: start sequentially, not simultaneously!)
|
||||
|
||||
Two trainings on two GPUs MUST be started one after another with a ~7 min gap,
|
||||
because they share the same CelerisLab kernel cache. Simultaneous startup causes
|
||||
kernel compilation race conditions → corrupted kernel → reward=0.000 forever.
|
||||
|
||||
```bash
|
||||
# 1. Clean stale cache
|
||||
rm -f /home/frank14f/CelerisLab/src/CelerisLab/lbm/kernels/config/config_objects.h
|
||||
rm -f /home/frank14f/CelerisLab/src/CelerisLab/lbm/kernels/kernel.ptx
|
||||
|
||||
# 2. Start Bias on GPU0 (with nohup so it survives terminal close)
|
||||
cd /home/frank14f/DynamisLab/src/drl_pinball/train
|
||||
nohup conda run --no-capture-output -n pycuda_3_10 python -u train_karman_2000x600.py \
|
||||
--device-id 0 --seed 42 --total-episodes 500 --target target.npy \
|
||||
> nohup_bias.log 2>&1 &
|
||||
|
||||
# 3. WAIT ~7 minutes for Bias Ep1 to appear (check train.log)
|
||||
# Confirm reward is non-zero before proceeding!
|
||||
|
||||
# 4. Start NoBias on GPU1
|
||||
nohup conda run --no-capture-output -n pycuda_3_10 python -u train_karman_2000x600.py \
|
||||
--device-id 1 --seed 42 --total-episodes 500 --target target.npy --no-bias \
|
||||
> nohup_nobias.log 2>&1 &
|
||||
```
|
||||
|
||||
### Monitoring
|
||||
```bash
|
||||
# Check progress
|
||||
tail -5 output/bias_seed42_s2048_e10_v4/train.log
|
||||
tail -5 output/nobias_seed42_s2048_e10_v4/train.log
|
||||
|
||||
# TensorBoard
|
||||
tensorboard --logdir output/bias_seed42_s2048_e10_v4/tb --port 6006
|
||||
|
||||
# Plot training curves
|
||||
conda run -n pycuda_3_10 python -u analyze_final.py
|
||||
```
|
||||
|
||||
### Stopping
|
||||
```bash
|
||||
ps aux | grep train_karman | grep -v grep | awk '{print $2}' | xargs kill
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 4. Reward Design (V3 — Current)
|
||||
## 4. Reward Design
|
||||
|
||||
### Formula
|
||||
|
||||
```python
|
||||
FORCE_SCALE = 0.0025 # fixed physical constant (combined max|force| from Phase 0)
|
||||
# Gaussian reward (no zero-crossing spikes)
|
||||
r_cd_raw = exp(-cd_norm² * K_CD) # cd_norm = (Σfx)/3 / FORCE_SCALE
|
||||
r_cl_raw = exp(-cl_norm² * K_CL)
|
||||
|
||||
# Gaussian reward (no zero-crossing spikes that exp(-|x|) causes)
|
||||
r_cd_raw = exp(-cd_norm² * 50) # cd_norm = (Σfx)/3 / FORCE_SCALE
|
||||
r_cl_raw = exp(-cl_norm² * 100) # cl_norm = (Σfy)/3 / FORCE_SCALE
|
||||
|
||||
# EMA smoothing for cd/cl only (r_sim is already smooth from DTW)
|
||||
# EMA smoothing for cd/cl (r_sim uses DTW, already smooth)
|
||||
r_cd = EMA(r_cd_raw, weight=0.2)
|
||||
r_cl = EMA(r_cl_raw, weight=0.2)
|
||||
|
||||
# Normalized DTW similarity (raw, no EMA)
|
||||
# norm_scale = mean of target's 3 uy-channel stds (fixed once target is recorded)
|
||||
r_sim = piecewise_map(sim, breakpoints, values)
|
||||
# Normalized DTW similarity (piecewise-mapped to [0,1])
|
||||
r_sim = piecewise_map(sim, SIM_BP, SIM_VAL)
|
||||
|
||||
# Floor penalty: prevents DRL from sacrificing one component
|
||||
if r_cd < 0.10: penalty += 0.05 * (0.10 - r_cd) / 0.10
|
||||
if r_cl < 0.10: penalty += 0.05 * (0.10 - r_cl) / 0.10
|
||||
if r_sim < 0.10: penalty += 0.05 * (0.10 - r_sim) / 0.10
|
||||
# Floor penalty: prevents sacrificing one component
|
||||
penalty = 0.05 * sum(max(0, FLOOR - r) / FLOOR for r, FLOOR in zip(...))
|
||||
|
||||
reward = max(0, 0.30*r_cd + 0.30*r_cl + 0.40*r_sim - penalty)
|
||||
reward = max(0, W_CD*r_cd + W_CL*r_cl + W_SIM*r_sim - penalty)
|
||||
# W_CD=0.30, W_CL=0.30, W_SIM=0.40
|
||||
```
|
||||
|
||||
### Three-stage targets
|
||||
|
||||
| Stage | r_cd | r_cl | r_sim | Description |
|
||||
|-------|------|------|-------|-------------|
|
||||
| Stage0 (zero rotation) | ~0 | ~0 | ~0.2 | No control baseline |
|
||||
| Stage1 (reference open-loop) | ~0.7 | ~0.5 | ~0.5 | Legacy-equiv bias |
|
||||
| Optimal (trained) | ~0.9 | ~0.9 | ~0.9 | Full cloaking |
|
||||
|
||||
### Why Gaussian not exp(-|x|)
|
||||
|
||||
`exp(-|x|)` has maximum gradient at x=0. When cd/cl oscillates around zero,
|
||||
the reward swings wildly (spikes at every zero-crossing). `exp(-x²)` has zero
|
||||
gradient at x=0, giving smooth reward near the optimum.
|
||||
`exp(-|x|)` has maximum gradient at x=0, causing reward spikes at zero-crossings
|
||||
of oscillating cd/cl. `exp(-x²)` has zero gradient at x=0 → smooth near optimum.
|
||||
|
||||
### Why Normalized DTW
|
||||
### Why normalized DTW
|
||||
|
||||
Raw DTW `1 - cost/n` has very narrow dynamic range (0.70-0.97) because cost
|
||||
depends on absolute signal amplitude. Normalizing by target's uy-channel std
|
||||
(≈0.20) extends the range to 0.0-0.9, giving DRL a meaningful gradient.
|
||||
Raw DTW has narrow dynamic range (0.70-0.97). Normalizing by target's uy-channel
|
||||
std extends range to 0.0-0.9, giving DRL meaningful gradient.
|
||||
|
||||
The norm_scale is computed once from the target signal and is fixed for all
|
||||
training and inference. It's universal — same formula works for cloak and
|
||||
illusion (each scenario has its own target → its own norm_scale).
|
||||
### Why no EMA on r_sim
|
||||
|
||||
### Why r_sim has no EMA
|
||||
|
||||
DTW is already a smooth signal (30-step windowed comparison). Adding EMA
|
||||
over-smoothed it and introduced artificial delay. The 40-step physical delay
|
||||
(convection 25 steps + DTW window 15 steps) is handled by GAE with gamma=0.995.
|
||||
|
||||
### Three-stage targets (from Phase 0 measurement)
|
||||
|
||||
| Stage | r_cd | r_cl | r_sim | total | Target |
|
||||
|-------|------|------|-------|-------|--------|
|
||||
| Stage0 (no rotation) | 0.03 | 0.22 | 0.17 | 0.14 | ~0.2 |
|
||||
| Stage1 (bias action) | 0.72 | 0.54 | 0.49 | 0.58 | ~0.5 |
|
||||
| Optimal (trained) | ~0.9 | ~0.9 | ~0.9 | ~0.9 | ~0.9 |
|
||||
DTW is already a smooth 30-step windowed signal. Adding EMA over-smooths.
|
||||
|
||||
---
|
||||
|
||||
## 5. PPO Configuration (V4 — Current)
|
||||
## 5. PPO Configuration
|
||||
|
||||
```python
|
||||
PPO(
|
||||
@ -270,203 +215,106 @@ PPO(
|
||||
policy_kwargs={"activation_fn": Sin, "net_arch": [64, 64]},
|
||||
env=vec_env,
|
||||
device=torch.device("cuda:X"),
|
||||
n_steps=2048, # MUST be 2048 (legacy default). 512 causes noisy curves.
|
||||
batch_size=64, # SB3 default
|
||||
n_epochs=10, # SB3 default. 3 was too few → slow learning.
|
||||
learning_rate=3e-4, # SB3 default
|
||||
gamma=0.995, # Higher than SB3 default (0.99) for r_sim delay propagation
|
||||
# ent_coef=0.0 # SB3 default (not set). n_steps=2048 provides enough diversity.
|
||||
# target_kl=None # SB3 default (not set). Well-estimated gradients don't need KL stop.
|
||||
verbose=0,
|
||||
n_steps=2048, # MUST be 2048. 512 → noisy curves.
|
||||
batch_size=64,
|
||||
n_epochs=10,
|
||||
learning_rate=3e-4,
|
||||
gamma=0.995, # Higher than default for DTW delay propagation
|
||||
)
|
||||
```
|
||||
|
||||
### Key lesson: Don't deviate from SB3 defaults without strong reason
|
||||
|
||||
V1-V3 used n_steps=512, n_epochs=3, ent_coef=0.01, target_kl=0.03. This caused:
|
||||
- Noisy training curves (high gradient variance from small n_steps)
|
||||
- Slow learning (3 epochs too few)
|
||||
- Degradation after peak (policy oscillation from noisy gradients)
|
||||
|
||||
V4 matches SB3 defaults (n_steps=2048, n_epochs=10, no ent_coef, no target_kl)
|
||||
and produces smooth, steadily rising curves — matching the legacy training quality.
|
||||
|
||||
### Evaluation: stochastic (NOT deterministic)
|
||||
|
||||
```python
|
||||
action, _ = model.predict(eval_obs) # NO deterministic=True
|
||||
```
|
||||
|
||||
Legacy training used stochastic eval. Deterministic eval gives sharp, noisy
|
||||
rewards. Stochastic eval averages over the policy distribution → smoother curves.
|
||||
For paper presentation, stochastic is more honest (shows expected performance).
|
||||
|
||||
### Symmetry: per-rollout, not per-step
|
||||
|
||||
The G-mirror wrapper decides ONCE per rollout (every 2048 steps) whether to
|
||||
mirror. If mirrored, ALL steps in that rollout use G-transform consistently.
|
||||
Per-step random mirroring adds noise; per-rollout is clean.
|
||||
|
||||
During evaluation, symmetry is disabled (prob=0.0) for clean policy assessment.
|
||||
- **Evaluation**: stochastic (no deterministic=True) — smoother curves
|
||||
- **Symmetry**: per-rollout G-mirror (50% probability). Disabled during eval.
|
||||
- **Every episode saves**: `ep0001_model.zip` + `ep0001_vecnormalize.pkl` saved each episode
|
||||
|
||||
---
|
||||
|
||||
## 6. Obs Normalization
|
||||
## 6. Obs Normalization (Two-layer)
|
||||
|
||||
### Two-layer approach
|
||||
|
||||
1. **Env physical norm** (fixed, reproducible):
|
||||
- forces / FORCE_SCALE (0.0025)
|
||||
- sensors / SENS_SCALE (0.8, legacy-equiv scale)
|
||||
- No clipping (VecNormalize handles that)
|
||||
1. **Env physical norm** (fixed, from calibration):
|
||||
- forces / FORCE_SCALE, sensors / SENS_SCALE
|
||||
- No clipping (VecNormalize handles)
|
||||
|
||||
2. **SB3 VecNormalize** (running mean/std):
|
||||
- norm_obs=True, norm_reward=False
|
||||
- clip_obs=10.0 (generous, physical norm already keeps obs ~[-1,1])
|
||||
- norm_obs=True, norm_reward=False, clip_obs=10.0
|
||||
- Saved to `vec_normalize.pkl` for inference
|
||||
|
||||
### Why not just VecNormalize
|
||||
|
||||
Physical norm ensures the obs is in a reasonable range even before VecNormalize
|
||||
has collected enough statistics. This prevents extreme values in the first few
|
||||
hundred steps. The physical norm constants (FORCE_SCALE, SENS_SCALE) are fixed
|
||||
and do not change between training and inference.
|
||||
|
||||
### Why not just physical norm (like legacy)
|
||||
|
||||
VecNormalize adapts to the actual obs distribution, giving each dimension
|
||||
equal footing. Without it, sensor ux (range 0.6-0.9) dominates sensor uy
|
||||
(range 0.2) after physical norm. VecNormalize corrects this.
|
||||
|
||||
---
|
||||
|
||||
## 7. Action Configuration
|
||||
## 7. Action Configuration (V5: no_bias only)
|
||||
|
||||
### Bias mode (default)
|
||||
```python
|
||||
ACTION_SCALE = 8.0
|
||||
ACTION_BIAS = [0, -4, 4] # front, top(+y), bot(-y)
|
||||
# omega = -(action * 8 + [0,-4,4]) * U0 / RADIUS
|
||||
# Physical range: front [-8,8], top [-4,12], bot [-12,4] × U0
|
||||
```
|
||||
|
||||
### NoBias mode (--no-bias flag)
|
||||
```python
|
||||
ACTION_SCALE = 12.0
|
||||
ACTION_BIAS = [0, 0, 0]
|
||||
# omega = -(action * 12 + [0,0,0]) * U0 / RADIUS
|
||||
# Physical range: all cylinders [-12,12] × U0
|
||||
# omega = -(action * 12) * U0 / RADIUS
|
||||
# Physical range: all cylinders [-12, 12] × U0
|
||||
```
|
||||
|
||||
NoBias uses scale=12 (not 8) to cover the same omega range as Bias.
|
||||
With scale=8 and no bias, range is only [-8,8] — missing the ±12×U0 extremes
|
||||
that Bias reaches. This caused NoBias to fail at learning lift control (r_cl
|
||||
collapsed to 0.05) because it couldn't reach the necessary rotation speeds.
|
||||
|
||||
### Sign convention (CelerisLab new kernel)
|
||||
|
||||
The new CelerisLab kernel uses `Uw = -omega * ry` (negative sign).
|
||||
Legacy used `Uw = action * (y_c - y) / radius` (no negative).
|
||||
Conversion: `omega = -surface_vel / radius`
|
||||
Sign convention: `Uw = -omega * ry` (omega>0 = clockwise).
|
||||
|
||||
---
|
||||
|
||||
## 8. Environment Design
|
||||
|
||||
### Two-phase initialization (avoids runtime sync_bodies)
|
||||
### Two-phase initialization
|
||||
|
||||
1. `record_target()`: temporary Simulation (dist_cyl + sensors only) →
|
||||
record 150 steps of target signal → close
|
||||
2. `KarmanCloakEnv.__init__()`: training Simulation with ALL 7 objects
|
||||
upfront → warmup → zero-action FIFO → bias FIFO → snapshot
|
||||
1. `record_target()`: dist_cyl + sensors only → record 150-step target → close
|
||||
2. Training Simulation: all 7 objects → warmup → zero-action FIFO → snapshot
|
||||
|
||||
All objects (1 dist_cyl + 3 sensors + 3 pinball) are added before `initialize()`.
|
||||
No runtime `sync_bodies()` — it causes recompilation and cache conflicts.
|
||||
|
||||
### Body IDs (add order)
|
||||
Body IDs (add order):
|
||||
```
|
||||
0: dist_cyl (force only, skipped in obs)
|
||||
1: sensor 0 (top, +y)
|
||||
2: sensor 1 (center)
|
||||
3: sensor 2 (bottom, -y)
|
||||
4: pinball_front
|
||||
5: pinball_top (rear, +y)
|
||||
6: pinball_bottom (rear, -y)
|
||||
0: dist_cyl (force, skipped in obs)
|
||||
1-3: sensors (top, center, bottom)
|
||||
4-6: pinball (front, top_rear, bottom_rear)
|
||||
```
|
||||
|
||||
### Obs layout (12-dim, env output before VecNormalize)
|
||||
```
|
||||
[0:6] = forces / FORCE_SCALE: [front_fx, front_fy, top_fx, top_fy, bot_fx, bot_fy]
|
||||
[6:12] = sensors / SENS_SCALE: [s0_ux, s0_uy, s1_ux, s1_uy, s2_ux, s2_uy]
|
||||
```
|
||||
Obs layout (12-dim): forces[6] + sensors[6]
|
||||
|
||||
### No-reset training
|
||||
|
||||
`step()` always returns `terminated=False`. The flow field is never reset
|
||||
during training (matches legacy behavior). Episode length is controlled by
|
||||
the external training loop (`model.learn(2048)` per call).
|
||||
|
||||
Evaluation does `eval_env.reset()` (restores snapshot) for reproducible
|
||||
assessment. After eval, training continues from the post-eval state.
|
||||
`step()` returns `terminated=False`. Eval does `reset()` for reproducible assessment.
|
||||
|
||||
---
|
||||
|
||||
## 9. Bugs Found & Fixed (Full History)
|
||||
## 9. Cross-Re Transfer Pipeline
|
||||
|
||||
### Critical bugs
|
||||
### How to add a new Re
|
||||
|
||||
| # | Bug | Symptom | Fix |
|
||||
|---|-----|---------|-----|
|
||||
| 1 | **Simultaneous startup** | Two trainings on 2 GPUs → kernel compilation race → reward=0.000 forever | Start sequentially: Bias first, wait 7 min, then NoBias |
|
||||
| 2 | **Bias FIFO not applying omega** | `_init_cfd` set bias_omega but never called `set_omega()` during bias FIFO | Added `self._set_omega(bias_omega)` in bias FIFO loop |
|
||||
| 3 | **CPU training** | `device="cpu"` was used to avoid PyCUDA/PyTorch conflict | Use GPU with `context.push()/pop()` around CFD calls |
|
||||
| 4 | **Symmetry during eval** | G-mirror prob=0.5 during evaluation → noisy, inaccurate reward | Set `inner.prob = 0.0` before eval, restore after |
|
||||
| 5 | **DTW not normalized** | Raw DTW `1-cost/n` → sim always 0.90+ → no gradient | Normalize by target uy-channel avg std |
|
||||
| 6 | **r_sim EMA + delay** | Added unnecessary EMA smoothing and 20-step delay buffer | Removed both — DTW is already smooth, use raw value |
|
||||
| 7 | **exp(-\|x\|) reward** | Zero-crossing spikes in r_cd/r_cl → noisy training | Changed to Gaussian `exp(-x²*K)` |
|
||||
| 8 | **NoBias scale=8** | Can't reach ±12×U0 → r_cl collapses | Use scale=12 for NoBias |
|
||||
| 9 | **n_steps=512** | High gradient variance → noisy curves, degradation | Use n_steps=2048 (SB3 default) |
|
||||
| 10 | **n_epochs=3** | Too few → slow learning | Use n_epochs=10 (SB3 default) |
|
||||
| 11 | **deterministic eval** | Sharp, noisy reward measurements | Use stochastic eval (no deterministic=True) |
|
||||
| 12 | **Stale config_objects.h** | N_OBJS mismatch after previous run | `_clean_cache()` before each Simulation creation |
|
||||
```bash
|
||||
# 1. Create config (copy existing, change viscosity)
|
||||
cp configs/config_lbm_karman_2000x600.json configs/config_lbm_karman_2000x600_reNNN.json
|
||||
# Edit "viscosity" to U0 * 2D / Re_NNN
|
||||
|
||||
### Sensor scaling (from reproduce phase)
|
||||
# 2. Calibrate
|
||||
python calibrate.py --case reNNN --device-id 0 --si <SI> --config <config>
|
||||
|
||||
New CelerisLab `read_sensor(normalize=True)` returns area-averaged velocity
|
||||
(÷cell_count). Legacy returned raw sum (~78x larger). For DTW comparison with
|
||||
legacy-recorded targets, multiply by `SENSOR_CC=78`.
|
||||
# 3. Test (5 episodes, local)
|
||||
python train_karman.py --case-name transfer_reNNN --device-id 0 --seed 41 \
|
||||
--config <config> --calibration calibrations/reNNN/calibration.json \
|
||||
--si <SI> --total-episodes 5 \
|
||||
--transfer-model output/re100_karman_seed<BEST>/models/best_model.zip
|
||||
|
||||
### Omega sign inversion (from reproduce phase)
|
||||
# 4. Production (add to crossre_transfer.sh or run directly with 200 episodes)
|
||||
```
|
||||
|
||||
New kernel: `Uw = -omega * ry` (omega>0 = clockwise).
|
||||
Legacy: `Uw = action * (y_c - y) / radius` (no sign inversion).
|
||||
Conversion: `omega = -surface_vel / radius`
|
||||
### Transfer test results (5-episode verification, 2026-07-02)
|
||||
|
||||
| Re | Best Reward | r_cd | r_cl | r_sim | Time/ep |
|
||||
|----|------------|------|------|-------|---------|
|
||||
| 60 | 0.637 | 0.879 | 0.402 | 0.641 | 281s |
|
||||
| 200 | 0.428 | 0.677 | 0.259 | 0.387 | 186s |
|
||||
| 400 | 0.489 | 0.787 | 0.491 | 0.274 | 153s |
|
||||
|
||||
All show rapid learning from Re100 base. r_cl remains the hardest component across all Re.
|
||||
|
||||
---
|
||||
|
||||
## 10. Training Results Summary
|
||||
## 10. CelerisLab Integration
|
||||
|
||||
### V4 (current, 30 ep before OOM stop — full 500ep running)
|
||||
|
||||
| Training | Ep1 | Ep10 | Ep20 | Ep30 | Smooth? |
|
||||
|----------|-----|------|------|------|---------|
|
||||
| Bias | 0.32 | 0.37 | 0.40 | 0.54 | Yes, steady rise |
|
||||
| NoBias | 0.11 | 0.23 | 0.43 | 0.41 | Yes, near-monotonic |
|
||||
|
||||
### Previous versions (for reference)
|
||||
|
||||
| Version | n_steps | Key change | Result |
|
||||
|---------|---------|------------|--------|
|
||||
| V1 | 512 | Config8 reward, exp(-\|x\|) | Peak 0.79 but noisy, possible degradation |
|
||||
| V2 | 512 | Gaussian + EMA r_sim | Stable but slow (0.62 peak) |
|
||||
| V3 | 512 | Normalized DTW, no r_sim EMA | Stable, NoBias reached 0.73 |
|
||||
| **V4** | **2048** | **SB3 defaults, stochastic eval** | **Smoothest curves, running** |
|
||||
|
||||
### NoBias vs Bias
|
||||
|
||||
NoBias requires more episodes to peak (~265 vs ~89 in V3) but can reach
|
||||
similar or higher peak reward. NoBias r_cl is the main challenge — it needs
|
||||
the scale=12 action range and floor penalty to prevent collapse.
|
||||
**CelerisLab is a git submodule** at `DynamisLab/CelerisLab/`.
|
||||
All Python imports use `from CelerisLab import Simulation`.
|
||||
No external paths — everything is self-contained within the repo.
|
||||
Do NOT reference `/home/frank14f/CelerisLab` anywhere.
|
||||
|
||||
---
|
||||
|
||||
@ -474,62 +322,31 @@ the scale=12 action range and floor penalty to prevent collapse.
|
||||
|
||||
| Parameter | Value | Where | Notes |
|
||||
|-----------|-------|-------|-------|
|
||||
| Grid | 2000×600 | config_lbm_karman_2000x600.json | uniform inlet, free_slip |
|
||||
| Grid | 2000×600 | config | uniform inlet, free_slip |
|
||||
| U0 | 0.01 | config | lattice inlet velocity |
|
||||
| ν | 0.004 | config | Re_D=50 (code Re=100) |
|
||||
| SI | 800 | env | LBM steps per action |
|
||||
| ν (re100) | 0.004 | config | Re_D=50 (code Re=100) |
|
||||
| SI | 400-800 | calibration | varies by Re |
|
||||
| FIFO_LEN | 150 | env | history buffer |
|
||||
| CONV_LEN | 30 | env | DTW comparison window |
|
||||
| FORCE_SCALE | 0.0025 | env | reward normalization |
|
||||
| SENS_SCALE | 0.8 | env | obs normalization (legacy-equiv) |
|
||||
| SENSOR_CC | 78 | env | sensor area→legacy conversion |
|
||||
| K_CD | 50 | env | Gaussian reward coefficient |
|
||||
| K_CL | 100 | env | Gaussian reward coefficient |
|
||||
| K_CD/K_CL | 50/100 | env | Gaussian reward coefficients |
|
||||
| W_CD/W_CL/W_SIM | 0.30/0.30/0.40 | env | reward weights |
|
||||
| FLOOR_CD/CL/SIM | 0.10 | env | floor penalty threshold |
|
||||
| n_steps | 2048 | train | PPO rollout size |
|
||||
| n_epochs | 10 | train | PPO epochs per update |
|
||||
| gamma | 0.995 | train | discount factor |
|
||||
| ACTION_SCALE | 8 (bias) / 12 (nobias) | env | action scaling |
|
||||
| ACTION_BIAS | [0,-4,4] / [0,0,0] | env | action offset |
|
||||
| ACTION_SCALE | 12.0 | env | no_bias only |
|
||||
|
||||
---
|
||||
|
||||
## 12. CFD Config
|
||||
## 12. Bugs Found & Fixed
|
||||
|
||||
File: `configs/config_lbm_karman_2000x600.json`
|
||||
|
||||
```
|
||||
Grid: 2000×600, D2Q9, MRT, double_buffer, FP32
|
||||
Inlet: uniform, regularized scheme
|
||||
Walls: free_slip (matches experimental water tunnel)
|
||||
Outlet: neq_extrap with backflow_clamp
|
||||
```
|
||||
|
||||
### EsoPull note
|
||||
|
||||
EsoPull streaming was tested — it computes correctly but gives ~5% slowdown
|
||||
(not the expected 50% speedup) on this grid size. Use double_buffer for speed.
|
||||
EsoPull + zou_he_local inlet gives very different physics (different force
|
||||
balances) — only use if you re-record the target with the same inlet scheme.
|
||||
|
||||
---
|
||||
|
||||
## 13. Future Work
|
||||
|
||||
1. **Re50/re200/re400 transfer learning**: Calibrate each Re, then transfer from re100 best model using `--transfer` flag. Use adjusted SI per Re.
|
||||
2. **Vortex cloak**: `env_vortex.py` ready. Transfer from re100 model. Lamb and Taylor variants.
|
||||
3. **Illusion**: `env_illusion.py` needed. S_DIM=14, harmonics target reconstruction. Adjusted pinball/sensor positions.
|
||||
4. **Steady cloak**: Simpler case (no upstream disturbance cylinder). Target is uniform flow.
|
||||
5. **Symmetry ablation study**: Compare with/without G-mirror to quantify benefit.
|
||||
6. **Longer training**: 500 episodes may not be enough. Try 1000 episodes.
|
||||
7. **Learning rate schedule**: Try lr decay after peak to prevent degradation.
|
||||
|
||||
## 14. V5 Design Principles
|
||||
|
||||
1. **Calibration-first**: Every case runs calibrate.py before training. Produces a single immutable JSON.
|
||||
2. **No-bias only**: action_scale=12, action_bias=[0,0,0]. Simplifies G-mirror, removes human priors.
|
||||
3. **Parameterized envs**: env_karman.py accepts calibration dict + config_path + si. No module-level hardcode.
|
||||
4. **Sequential GPU startup**: launch_multi.sh enforces 7-min delay between launches.
|
||||
5. **Immutable calibration**: Once produced, calibration is read-only for training and inference. Saved alongside model.
|
||||
6. **r_sim maps to [0,1]**: SIM_VAL[-1]=1.0. Full-range normalized DTW similarity.
|
||||
| # | Bug | Symptom | Fix |
|
||||
|---|-----|---------|-----|
|
||||
| 1 | Simultaneous GPU startup | Kernel compilation race → reward=0.000 | Sequential launch, 7-min delay |
|
||||
| 2 | Bias FIFO not applying omega | Wrong normalization baseline | Added set_omega in bias FIFO |
|
||||
| 3 | DTW not normalized | sim always 0.90+ → flat gradient | Normalize by target uy-channel avg std |
|
||||
| 4 | Gaussian vs exp(-|x|) | Zero-crossing reward spikes | Use exp(-x² · K) |
|
||||
| 5 | n_steps=512 | Noisy curves, degradation | Use 2048 (SB3 default) |
|
||||
| 6 | deterministic eval | Sharp, noisy reward | Stochastic eval |
|
||||
| 7 | External CelerisLab paths | Permission issues, server mismatch | Use DynamisLab submodule only |
|
||||
| 8 | Manual kernel cache cleaning | Unnecessary, root-only files | CelerisLab handles internally |
|
||||
|
||||
@ -35,15 +35,6 @@ if str(_REPO) not in sys.path:
|
||||
|
||||
from CelerisLab import Simulation
|
||||
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
def _clean_cache():
|
||||
for p in [_CONFIG_H, _PTX]:
|
||||
if p.exists():
|
||||
p.unlink()
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Physics / geometry constants
|
||||
# ---------------------------------------------------------------------------
|
||||
@ -72,6 +63,7 @@ _REF_OMEGA = np.array([0.0, 0.004, -0.004], dtype=np.float32)
|
||||
|
||||
# Reward constants
|
||||
K_CD = 50.0; K_CL = 100.0
|
||||
K_CD_ILLUSION = 12.0; K_CL_ILLUSION = 25.0 # lower: 3-pinball forces vs 1-target cylinder, wider error range
|
||||
W_CD = 0.30; W_CL = 0.30; W_SIM = 0.40
|
||||
FLOOR_CD = 0.10; FLOOR_CL = 0.10; FLOOR_SIM = 0.10
|
||||
FLOOR_PENALTY = 0.05
|
||||
@ -199,7 +191,7 @@ _ILL_PINBALL_REAR_X = 406.0
|
||||
_ILL_SENSOR_X = 600.0
|
||||
_ILL_TARGET_X = 400.0
|
||||
_ILL_TARGET_RADIUS = 1.0 * L0
|
||||
_ILL_REF_OMEGA = np.array([0.0, 0.002, -0.002], dtype=np.float32) # legacy-equiv bias
|
||||
_ILL_REF_OMEGA = np.array([0.0, 0.001, -0.001], dtype=np.float32) # surface_vel = [0,-1,1]*U0, FIFO-init level
|
||||
|
||||
def _analyze_harmonics_for_calib(states, n_harmonics=5):
|
||||
N, D = states.shape
|
||||
@ -224,7 +216,6 @@ def _analyze_harmonics_for_calib(states, n_harmonics=5):
|
||||
def _calibrate_illusion(case, config_path, device_id, si, out_dir, log, warmup):
|
||||
# ---- Step 1: Record target (target cylinder + 3 sensors) ----
|
||||
log("Step 1: Recording illusion target (target cyl + sensors)...")
|
||||
_clean_cache()
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
sim.add_body("circle", center=(_ILL_TARGET_X, CENTER_Y, 0.0),
|
||||
@ -266,7 +257,6 @@ def _calibrate_illusion(case, config_path, device_id, si, out_dir, log, warmup):
|
||||
|
||||
# ---- Step 2: Training sim (3 sensors + 3 pinball) ----
|
||||
log("Step 2: Creating training sim (6 objects)...")
|
||||
_clean_cache()
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
sensor_ids = [
|
||||
@ -315,9 +305,23 @@ def _calibrate_illusion(case, config_path, device_id, si, out_dir, log, warmup):
|
||||
s0_sim = float(np.mean(stage0["sims"]))
|
||||
s1_sim = float(np.mean(stage1["sims"]))
|
||||
|
||||
sim_bp = [0.0, s0_sim, s1_sim,
|
||||
s1_sim + (1.0 - s1_sim) * 0.4,
|
||||
s1_sim + (1.0 - s1_sim) * 0.7,
|
||||
# Build SIM_BP: [0, worst_measured, better_measured, ..., 1.0]
|
||||
# Normally (Karman) Stage0 < Stage1 (zero is worse, reference is better).
|
||||
# But for Illusion, Stage0 may already match the target well, and reference
|
||||
# rotation (steady-cloak-like) makes it worse. Swap roles in that case.
|
||||
if s1_sim < s0_sim:
|
||||
# Reference action is worse than zero (Illusion case).
|
||||
# Stage1 (reference) -> r_sim ~0.2, Stage0 (zero) -> r_sim ~0.4
|
||||
worst_sim = s1_sim
|
||||
better_sim = s0_sim
|
||||
else:
|
||||
# Normal (Karman) case.
|
||||
worst_sim = s0_sim
|
||||
better_sim = s1_sim
|
||||
|
||||
sim_bp = [0.0, worst_sim, better_sim,
|
||||
better_sim + (1.0 - better_sim) * 0.4,
|
||||
better_sim + (1.0 - better_sim) * 0.7,
|
||||
1.0]
|
||||
sim_val = [0.0, 0.2, 0.5, 0.8, 0.9, 1.0]
|
||||
|
||||
@ -411,7 +415,6 @@ def main() -> int:
|
||||
|
||||
# ---- Step 1: Record target (Karman: dist_cyl + sensors only) ----
|
||||
log("Step 1: Recording target signal...")
|
||||
_clean_cache()
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
dist_id_t = sim.add_body("circle", center=(DIST_X, CENTER_Y, 0.0), radius=1.0 * L0)
|
||||
@ -444,7 +447,6 @@ def main() -> int:
|
||||
|
||||
# ---- Step 2: Create training sim ----
|
||||
log("Step 2: Creating training sim (all 7 objects)...")
|
||||
_clean_cache()
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
dist_id = sim.add_body("circle", center=(DIST_X, CENTER_Y, 0.0), radius=1.0 * L0)
|
||||
@ -545,8 +547,8 @@ def main() -> int:
|
||||
"dtw_norm_scale": float(dtw_norm_scale),
|
||||
"SIM_BP": [float(x) for x in sim_bp],
|
||||
"SIM_VAL": [float(x) for x in sim_val],
|
||||
"K_CD": K_CD,
|
||||
"K_CL": K_CL,
|
||||
"K_CD": K_CD_ILLUSION,
|
||||
"K_CL": K_CL_ILLUSION,
|
||||
"W_CD": W_CD,
|
||||
"W_CL": W_CL,
|
||||
"W_SIM": W_SIM,
|
||||
|
||||
@ -10,15 +10,15 @@
|
||||
"FIFO_LEN": 150,
|
||||
"CONV_LEN": 30,
|
||||
"SENSOR_CC": 78.0,
|
||||
"FORCE_SCALE": 0.0029,
|
||||
"FORCE_SCALE": 0.002,
|
||||
"SENS_SCALE": 0.93,
|
||||
"dtw_norm_scale": 0.251,
|
||||
"SIM_BP": [
|
||||
0.0,
|
||||
0.73,
|
||||
0.81,
|
||||
0.82,
|
||||
0.83,
|
||||
0.84,
|
||||
0.89,
|
||||
0.94,
|
||||
1.0
|
||||
],
|
||||
"SIM_VAL": [
|
||||
@ -29,8 +29,8 @@
|
||||
0.9,
|
||||
1.0
|
||||
],
|
||||
"K_CD": 50.0,
|
||||
"K_CL": 100.0,
|
||||
"K_CD": 12.0,
|
||||
"K_CL": 25.0,
|
||||
"W_CD": 0.3,
|
||||
"W_CL": 0.3,
|
||||
"W_SIM": 0.4,
|
||||
|
||||
@ -1,12 +1,12 @@
|
||||
[
|
||||
{
|
||||
"dc": 0.6848811427752177,
|
||||
"dc": 0.6848811439673106,
|
||||
"amps": [
|
||||
0.20348444790268552,
|
||||
0.02915900547523486,
|
||||
0.02784031012371259,
|
||||
0.026538733180849892,
|
||||
0.01919840514104942
|
||||
0.2034844510278786,
|
||||
0.029159004511650252,
|
||||
0.027840311609468333,
|
||||
0.02653873514475916,
|
||||
0.01919840709106637
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -16,21 +16,21 @@
|
||||
0.04666666666666667
|
||||
],
|
||||
"phases": [
|
||||
-1.2362005505810982,
|
||||
-1.1766679461536396,
|
||||
1.8863477863626301,
|
||||
-2.834029481822404,
|
||||
-0.7286212732915559
|
||||
-1.2362005574545905,
|
||||
-1.1766679293247424,
|
||||
1.886347686621138,
|
||||
-2.834029604726311,
|
||||
-0.7286213249887804
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": -0.026989686206604045,
|
||||
"dc": -0.026989686344750227,
|
||||
"amps": [
|
||||
0.2790809917571785,
|
||||
0.07572308632544687,
|
||||
0.04473095779630388,
|
||||
0.039113569175253486,
|
||||
0.038411479291876154
|
||||
0.2790809917939555,
|
||||
0.07572308617806586,
|
||||
0.044730957946728864,
|
||||
0.03911356916608822,
|
||||
0.03841147969973682
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -40,21 +40,21 @@
|
||||
0.03333333333333333
|
||||
],
|
||||
"phases": [
|
||||
3.0547687538583306,
|
||||
0.6711306028525605,
|
||||
-0.10499736264498984,
|
||||
-2.7955698201715835,
|
||||
3.1220861518480647
|
||||
3.0547687544981676,
|
||||
0.671130604261652,
|
||||
-0.10499736216933567,
|
||||
-2.7955698151736605,
|
||||
3.1220861483228126
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": 0.5674182403087616,
|
||||
"dc": 0.5674182399113973,
|
||||
"amps": [
|
||||
0.06514963980633084,
|
||||
0.02472242867395058,
|
||||
0.014768360330077245,
|
||||
0.010156019183366295,
|
||||
0.00963919337572356
|
||||
0.06514964076759504,
|
||||
0.024722427490218583,
|
||||
0.01476835974183025,
|
||||
0.010156020328112192,
|
||||
0.009639191161046853
|
||||
],
|
||||
"freqs": [
|
||||
0.05333333333333334,
|
||||
@ -64,21 +64,21 @@
|
||||
0.1
|
||||
],
|
||||
"phases": [
|
||||
-0.7488772609862894,
|
||||
2.444199252505853,
|
||||
-0.7923018439616549,
|
||||
2.505537406241232,
|
||||
-1.0758943656055078
|
||||
-0.7488773712604715,
|
||||
2.444199243833346,
|
||||
-0.7923014988253196,
|
||||
2.5055369932207086,
|
||||
-1.0758944221483198
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": 0.014232361110819814,
|
||||
"dc": 0.014232359210921763,
|
||||
"amps": [
|
||||
0.4461623192829029,
|
||||
0.0779793315804837,
|
||||
0.0550568453231777,
|
||||
0.05331603051277121,
|
||||
0.04874852247514895
|
||||
0.446162318961067,
|
||||
0.07797933262227957,
|
||||
0.05505684614794328,
|
||||
0.05331603052134459,
|
||||
0.04874852338883084
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -88,21 +88,21 @@
|
||||
0.07333333333333333
|
||||
],
|
||||
"phases": [
|
||||
-3.1049303564692066,
|
||||
0.027203636269757633,
|
||||
0.09727455160562376,
|
||||
-3.095222555392279,
|
||||
-3.044905056768087
|
||||
-3.104930355366964,
|
||||
0.027203661003797012,
|
||||
0.09727456602426711,
|
||||
-3.0952225671159,
|
||||
-3.0449050417637897
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": 0.6885532836119334,
|
||||
"dc": 0.688553271094958,
|
||||
"amps": [
|
||||
0.2018316634259717,
|
||||
0.033110505502103905,
|
||||
0.029481533925939486,
|
||||
0.027209600513899444,
|
||||
0.0210247548932384
|
||||
0.20183164156868158,
|
||||
0.033110505614057664,
|
||||
0.029481548571705242,
|
||||
0.02720959921763942,
|
||||
0.02102475262837555
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -112,21 +112,21 @@
|
||||
0.07333333333333333
|
||||
],
|
||||
"phases": [
|
||||
1.8817739461803038,
|
||||
2.742416499482124,
|
||||
-1.1159813908192473,
|
||||
1.7448219790472927,
|
||||
1.8563271918371878
|
||||
1.8817739067162356,
|
||||
2.7424167422444903,
|
||||
-1.1159816805254985,
|
||||
1.7448225406868731,
|
||||
1.8563274842269628
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": 0.0464287880451108,
|
||||
"dc": 0.046428785625806386,
|
||||
"amps": [
|
||||
0.266793500575173,
|
||||
0.08445240294008177,
|
||||
0.05549481483267607,
|
||||
0.03176632041902732,
|
||||
0.025647047642157136
|
||||
0.2667935014251649,
|
||||
0.08445240278954501,
|
||||
0.05549481355853139,
|
||||
0.031766322660930275,
|
||||
0.025647051108265238
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -136,21 +136,21 @@
|
||||
0.04666666666666667
|
||||
],
|
||||
"phases": [
|
||||
3.0338443035387885,
|
||||
-2.634676303912741,
|
||||
-0.03215847385538798,
|
||||
-0.005548069481591528,
|
||||
0.8284453225261742
|
||||
3.0338443211965744,
|
||||
-2.6346763210802586,
|
||||
-0.032158500100019624,
|
||||
-0.005548092544019917,
|
||||
0.8284453426622406
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": 0.002807500216489037,
|
||||
"dc": 0.0028075010624403754,
|
||||
"amps": [
|
||||
1.712089159739782e-05,
|
||||
6.84709236632855e-06,
|
||||
3.6652517552615438e-06,
|
||||
2.9729489026324893e-06,
|
||||
2.0179845909227172e-06
|
||||
1.7116459800799798e-05,
|
||||
6.866397692253015e-06,
|
||||
3.6591197181741872e-06,
|
||||
2.979711101541413e-06,
|
||||
2.0364590024389495e-06
|
||||
],
|
||||
"freqs": [
|
||||
0.05333333333333334,
|
||||
@ -160,21 +160,21 @@
|
||||
0.06666666666666667
|
||||
],
|
||||
"phases": [
|
||||
-2.4049945951829694,
|
||||
0.6548426535356905,
|
||||
-2.327050319985697,
|
||||
0.5677103875435425,
|
||||
-2.2688377992989093
|
||||
-2.4055008082312566,
|
||||
0.6560206596369318,
|
||||
-2.338324550342246,
|
||||
0.5697953506705974,
|
||||
-2.2632313796302874
|
||||
]
|
||||
},
|
||||
{
|
||||
"dc": -1.4845058922219323e-05,
|
||||
"dc": -1.484534581322805e-05,
|
||||
"amps": [
|
||||
0.0006606466483213061,
|
||||
0.00010502206909109318,
|
||||
8.38977596666718e-05,
|
||||
4.8485264824562914e-05,
|
||||
4.61590173098796e-05
|
||||
0.000660645874022141,
|
||||
0.00010502076564124189,
|
||||
8.389696757021492e-05,
|
||||
4.848746760651764e-05,
|
||||
4.616076991013906e-05
|
||||
],
|
||||
"freqs": [
|
||||
0.02666666666666667,
|
||||
@ -184,11 +184,11 @@
|
||||
0.04
|
||||
],
|
||||
"phases": [
|
||||
-0.8350712750146696,
|
||||
2.4283127217217664,
|
||||
-0.9184054668600495,
|
||||
2.6043773129885874,
|
||||
-0.9754872507212389
|
||||
-0.8350715543078073,
|
||||
2.428305305182952,
|
||||
-0.9184117032214737,
|
||||
2.6043964384951956,
|
||||
-0.9755175136337735
|
||||
]
|
||||
}
|
||||
]
|
||||
49
src/drl_pinball/train/calibrations/re200/calibration.json
Normal file
49
src/drl_pinball/train/calibrations/re200/calibration.json
Normal file
@ -0,0 +1,49 @@
|
||||
{
|
||||
"case": "re200",
|
||||
"grid": {
|
||||
"nx": 2000,
|
||||
"ny": 600
|
||||
},
|
||||
"config_path": "/home/frank14f/DynamisLab/configs/config_lbm_karman_2000x600_re200.json",
|
||||
"SI": 500,
|
||||
"FIFO_LEN": 150,
|
||||
"CONV_LEN": 30,
|
||||
"SENSOR_CC": 78.0,
|
||||
"FORCE_SCALE": 0.0026,
|
||||
"SENS_SCALE": 0.9,
|
||||
"dtw_norm_scale": 0.269,
|
||||
"SIM_BP": [
|
||||
0.0,
|
||||
0.45,
|
||||
0.77,
|
||||
0.86,
|
||||
0.93,
|
||||
1.0
|
||||
],
|
||||
"SIM_VAL": [
|
||||
0.0,
|
||||
0.2,
|
||||
0.5,
|
||||
0.8,
|
||||
0.9,
|
||||
1.0
|
||||
],
|
||||
"K_CD": 12.0,
|
||||
"K_CL": 25.0,
|
||||
"W_CD": 0.3,
|
||||
"W_CL": 0.3,
|
||||
"W_SIM": 0.4,
|
||||
"FLOOR_CD": 0.1,
|
||||
"FLOOR_CL": 0.1,
|
||||
"FLOOR_SIM": 0.1,
|
||||
"FLOOR_PENALTY": 0.05,
|
||||
"ACTION_SCALE": 12.0,
|
||||
"ACTION_BIAS": [
|
||||
0.0,
|
||||
0.0,
|
||||
0.0
|
||||
],
|
||||
"U0": 0.01,
|
||||
"RADIUS": 10.0,
|
||||
"L0": 20.0
|
||||
}
|
||||
49
src/drl_pinball/train/calibrations/re400/calibration.json
Normal file
49
src/drl_pinball/train/calibrations/re400/calibration.json
Normal file
@ -0,0 +1,49 @@
|
||||
{
|
||||
"case": "re400",
|
||||
"grid": {
|
||||
"nx": 2000,
|
||||
"ny": 600
|
||||
},
|
||||
"config_path": "/home/frank14f/DynamisLab/configs/config_lbm_karman_2000x600_re400.json",
|
||||
"SI": 400,
|
||||
"FIFO_LEN": 150,
|
||||
"CONV_LEN": 30,
|
||||
"SENSOR_CC": 78.0,
|
||||
"FORCE_SCALE": 0.0042,
|
||||
"SENS_SCALE": 0.98,
|
||||
"dtw_norm_scale": 0.31,
|
||||
"SIM_BP": [
|
||||
0.0,
|
||||
0.56,
|
||||
0.73,
|
||||
0.84,
|
||||
0.92,
|
||||
1.0
|
||||
],
|
||||
"SIM_VAL": [
|
||||
0.0,
|
||||
0.2,
|
||||
0.5,
|
||||
0.8,
|
||||
0.9,
|
||||
1.0
|
||||
],
|
||||
"K_CD": 12.0,
|
||||
"K_CL": 25.0,
|
||||
"W_CD": 0.3,
|
||||
"W_CL": 0.3,
|
||||
"W_SIM": 0.4,
|
||||
"FLOOR_CD": 0.1,
|
||||
"FLOOR_CL": 0.1,
|
||||
"FLOOR_SIM": 0.1,
|
||||
"FLOOR_PENALTY": 0.05,
|
||||
"ACTION_SCALE": 12.0,
|
||||
"ACTION_BIAS": [
|
||||
0.0,
|
||||
0.0,
|
||||
0.0
|
||||
],
|
||||
"U0": 0.01,
|
||||
"RADIUS": 10.0,
|
||||
"L0": 20.0
|
||||
}
|
||||
49
src/drl_pinball/train/calibrations/re60/calibration.json
Normal file
49
src/drl_pinball/train/calibrations/re60/calibration.json
Normal file
@ -0,0 +1,49 @@
|
||||
{
|
||||
"case": "re60",
|
||||
"grid": {
|
||||
"nx": 2000,
|
||||
"ny": 600
|
||||
},
|
||||
"config_path": "/home/frank14f/DynamisLab/configs/config_lbm_karman_2000x600_re60.json",
|
||||
"SI": 800,
|
||||
"FIFO_LEN": 150,
|
||||
"CONV_LEN": 30,
|
||||
"SENSOR_CC": 78.0,
|
||||
"FORCE_SCALE": 0.0021,
|
||||
"SENS_SCALE": 0.72,
|
||||
"dtw_norm_scale": 0.107,
|
||||
"SIM_BP": [
|
||||
0.0,
|
||||
0.41,
|
||||
0.61,
|
||||
0.77,
|
||||
0.88,
|
||||
1.0
|
||||
],
|
||||
"SIM_VAL": [
|
||||
0.0,
|
||||
0.2,
|
||||
0.5,
|
||||
0.8,
|
||||
0.9,
|
||||
1.0
|
||||
],
|
||||
"K_CD": 12.0,
|
||||
"K_CL": 25.0,
|
||||
"W_CD": 0.3,
|
||||
"W_CL": 0.3,
|
||||
"W_SIM": 0.4,
|
||||
"FLOOR_CD": 0.1,
|
||||
"FLOOR_CL": 0.1,
|
||||
"FLOOR_SIM": 0.1,
|
||||
"FLOOR_PENALTY": 0.05,
|
||||
"ACTION_SCALE": 12.0,
|
||||
"ACTION_BIAS": [
|
||||
0.0,
|
||||
0.0,
|
||||
0.0
|
||||
],
|
||||
"U0": 0.01,
|
||||
"RADIUS": 10.0,
|
||||
"L0": 20.0
|
||||
}
|
||||
115
src/drl_pinball/train/crossre_transfer.sh
Executable file
115
src/drl_pinball/train/crossre_transfer.sh
Executable file
@ -0,0 +1,115 @@
|
||||
#!/bin/bash
|
||||
# Sequential cross-Re transfer learning: Re60 -> Re200 -> Re400
|
||||
# Each: calibrate (~5 min) + train N episodes
|
||||
#
|
||||
# Usage (local test, 5 episodes):
|
||||
# bash crossre_transfer.sh --re-list 60 --test-episodes 5
|
||||
#
|
||||
# Usage (production on server, 200 episodes each):
|
||||
# bash crossre_transfer.sh --re-list 60,200,400
|
||||
set -euo pipefail
|
||||
|
||||
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
||||
REPO_DIR="$(cd "$SCRIPT_DIR/../../.." && pwd)"
|
||||
TRAIN_DIR="$SCRIPT_DIR"
|
||||
CONFIG_DIR="$REPO_DIR/configs"
|
||||
# TODO: BEFORE PUSHING TO SERVER, update BEST_MODEL to the correct path
|
||||
BEST_MODEL="$TRAIN_DIR/output/re100_karman_seed539439/models/best_model.zip"
|
||||
GPU=0
|
||||
SEED=41
|
||||
EPISODES=200
|
||||
TEST_EPISODES=0
|
||||
RE_LIST=""
|
||||
LOG_BASE="/tmp/crossre_transfer"
|
||||
CONDA_ENV="pycuda_3_10"
|
||||
|
||||
usage() {
|
||||
echo "Usage: $0 [--re-list 60,200,400] [--test-episodes N] [--best-model PATH]"
|
||||
echo " --re-list Comma-separated Re numbers (default: 60,200,400)"
|
||||
echo " --test-episodes Run only N episodes per Re for quick verification"
|
||||
echo " --best-model Path to Re100 best model .zip"
|
||||
exit 1
|
||||
}
|
||||
|
||||
while [[ $# -gt 0 ]]; do
|
||||
case "$1" in
|
||||
--re-list) RE_LIST="$2"; shift 2 ;;
|
||||
--test-episodes) TEST_EPISODES="$2"; shift 2 ;;
|
||||
--best-model) BEST_MODEL="$2"; shift 2 ;;
|
||||
*) echo "Unknown option: $1"; usage ;;
|
||||
esac
|
||||
done
|
||||
|
||||
if [[ -z "$RE_LIST" ]]; then
|
||||
RE_LIST="60,200,400"
|
||||
fi
|
||||
|
||||
IFS=',' read -ra RE_ARR <<< "$RE_LIST"
|
||||
|
||||
echo "=== Cross-Re Transfer ==="
|
||||
echo " Re list: ${RE_ARR[*]}"
|
||||
echo " Best model: ${BEST_MODEL}"
|
||||
echo " Episodes: ${EPISODES} (test-mode: ${TEST_EPISODES})"
|
||||
echo " GPU: ${GPU}, Seed: ${SEED}"
|
||||
echo ""
|
||||
|
||||
if [[ ! -f "$BEST_MODEL" ]]; then
|
||||
echo "ERROR: Best model not found at $BEST_MODEL"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
mkdir -p "$LOG_BASE"
|
||||
|
||||
if [[ "$TEST_EPISODES" -gt 0 ]]; then
|
||||
EPISODES="$TEST_EPISODES"
|
||||
echo " TEST MODE: only $TEST_EPISODES episodes per Re"
|
||||
fi
|
||||
|
||||
for re in "${RE_ARR[@]}"; do
|
||||
case $re in
|
||||
60) SI=800; vis_label="re60" ;;
|
||||
200) SI=500; vis_label="re200" ;;
|
||||
400) SI=400; vis_label="re400" ;;
|
||||
*) echo "ERROR: Unknown Re=$re (supported: 60, 200, 400)"; exit 1 ;;
|
||||
esac
|
||||
CONFIG="$CONFIG_DIR/config_lbm_karman_2000x600_${vis_label}.json"
|
||||
CASE="transfer_${vis_label}"
|
||||
LOG="$LOG_BASE/${vis_label}.log"
|
||||
|
||||
echo "=== $(date): Starting $CASE (SI=$SI) ===" | tee -a "$LOG"
|
||||
|
||||
if [[ ! -f "$CONFIG" ]]; then
|
||||
echo " ERROR: Config not found: $CONFIG" | tee -a "$LOG"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Step 1: Calibrate (skip if calibration.json already exists)
|
||||
CAL_JSON="$TRAIN_DIR/calibrations/$vis_label/calibration.json"
|
||||
if [[ -f "$CAL_JSON" ]]; then
|
||||
echo " [SKIP] Calibration already exists: $CAL_JSON" | tee -a "$LOG"
|
||||
echo " (Delete calibrations/$vis_label/ to force re-calibration)"
|
||||
else
|
||||
echo " [$(date '+%H:%M:%S')] Calibrating..." | tee -a "$LOG"
|
||||
conda run --no-capture-output -n "$CONDA_ENV" python -u \
|
||||
"$TRAIN_DIR/calibrate.py" \
|
||||
--case "$vis_label" --device-id $GPU --si $SI \
|
||||
--config "$CONFIG" >> "$LOG" 2>&1
|
||||
echo " [$(date '+%H:%M:%S')] Calibration done." | tee -a "$LOG"
|
||||
fi
|
||||
|
||||
# Step 2: Train with transfer
|
||||
echo " [$(date '+%H:%M:%S')] Training ${EPISODES} episodes..." | tee -a "$LOG"
|
||||
conda run --no-capture-output -n "$CONDA_ENV" python -u \
|
||||
"$TRAIN_DIR/train_karman.py" \
|
||||
--case-name "$CASE" --device-id $GPU --seed $SEED \
|
||||
--config "$CONFIG" \
|
||||
--calibration "$CAL_JSON" \
|
||||
--si $SI --total-episodes $EPISODES \
|
||||
--transfer-model "$BEST_MODEL" >> "$LOG" 2>&1
|
||||
echo " [$(date '+%H:%M:%S')] Training done." | tee -a "$LOG"
|
||||
|
||||
echo "=== $(date): $CASE complete ===" | tee -a "$LOG"
|
||||
echo ""
|
||||
done
|
||||
|
||||
echo "=== ALL DONE ===" | tee -a "$LOG_BASE/summary.log"
|
||||
@ -1,21 +1,33 @@
|
||||
#!/usr/bin/env python3
|
||||
"""Hydrodynamic Illusion environment (V5 - calibration-driven, no_bias, 2000x600).
|
||||
"""Hydrodynamic Illusion environment for 2000x600 config (uniform, free-slip).
|
||||
|
||||
Two-phase initialization:
|
||||
Phase 1: target cylinder (diameter=1.0*L0) + 3 sensors -> warmup ->
|
||||
record 150-step signal -> FFT harmonics -> close
|
||||
Phase 2: Training sim (3 sensors + 3 pinball) -> warmup ->
|
||||
Design: Two-phase initialization to AVOID runtime sync_bodies().
|
||||
Phase 1: Temporary Simulation(target cyl + sensors) -> record 150-step signal
|
||||
+ FFT harmonics -> close
|
||||
Phase 2: Training Simulation(3 sensors + 3 pinball upfront) -> warmup ->
|
||||
zero-action FIFO -> snapshot
|
||||
|
||||
Observation (14-dim, physical norm, NO clip):
|
||||
CUDA context: mirrors legacy pattern - push CFD context before GPU ops, pop after.
|
||||
|
||||
Geometry: pinball@19/20.3L0, sensor@30L0, target cylinder@20L0 (1L diameter).
|
||||
Equivalent pinball-to-sensor distance matched to illusion target.
|
||||
|
||||
Observation (14-dim, physical norm, NO clip - VecNormalize handles that):
|
||||
[0:6] = raw_forces / FORCE_SCALE (front_fx,fy, top_fx,fy, bot_fx,fy)
|
||||
[6:12] = raw_sensors / SENS_SCALE (s0_ux,uy, s1_ux,uy, s2_ux,uy)
|
||||
[12:14] = target_cd, target_cl (from harmonics reconstruction)
|
||||
[12:14] = target_cd, target_cl (from FFT harmonics reconstruction)
|
||||
|
||||
Action (3-dim): no_bias only
|
||||
[-1,1] -> omega = -(action * 12 + [0,0,0]) * U0 / R
|
||||
[-1,1] -> omega = -(action*12 + [0,0,0]) * U0 / RADIUS
|
||||
|
||||
Reward: Gaussian cd/cl (compared to harmonics-reconstructed target) + normalized DTW.
|
||||
Reward (V3: Gaussian + EMA smoothing + normalized DTW):
|
||||
r_cd = EMA(exp(-(cd - target_cd)^2 * K_CD), EMA_FAST)
|
||||
r_cl = EMA(exp(-(cl - target_cl)^2 * K_CL), EMA_FAST)
|
||||
r_sim = piecewise_map(sim, SIM_BP, SIM_VAL)
|
||||
reward = W_CD*r_cd + W_CL*r_cl + W_SIM*r_sim - floor_penalty
|
||||
|
||||
K_CD/K_CL are LOWER than Karman (12/25 vs 50/100) because 3-pinball total force
|
||||
is compared to 1-target-cylinder force, giving wider normalized error range.
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
@ -34,14 +46,6 @@ if str(_REPO) not in sys.path:
|
||||
|
||||
from CelerisLab import Simulation
|
||||
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
def _clean_cache():
|
||||
for p in [_CONFIG_H, _PTX]:
|
||||
if p.exists(): p.unlink()
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
L0 = 20.0; U0 = 0.01; RADIUS = L0 / 2.0
|
||||
NX = 2000; NY = 600
|
||||
@ -151,7 +155,6 @@ def gen_target_states_at(t, harmonics):
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
def record_illusion_target(config_path, device_id, si, target_diam=1.0):
|
||||
_clean_cache()
|
||||
warmup = int(4.0 * NX / U0)
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
@ -255,7 +258,6 @@ class IllusionCloakEnv(gym.Env):
|
||||
|
||||
# ---- Phase 2: Training sim ----
|
||||
print(" [illusion] Phase 2: Training sim...", flush=True)
|
||||
_clean_cache()
|
||||
self.sim = Simulation(lbm_config_path=self._config_path, device_id=self.device_id)
|
||||
self.sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
self.sensor_ids = [
|
||||
|
||||
@ -41,14 +41,6 @@ if str(_REPO) not in sys.path:
|
||||
|
||||
from CelerisLab import Simulation
|
||||
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
def _clean_cache():
|
||||
for p in [_CONFIG_H, _PTX]:
|
||||
if p.exists(): p.unlink()
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Geometry constants (fixed across all Karman cloak cases)
|
||||
# ---------------------------------------------------------------------------
|
||||
@ -124,7 +116,6 @@ class ActionSmoother:
|
||||
# ---------------------------------------------------------------------------
|
||||
def record_target(config_path: str, device_id: int, si: int) -> np.ndarray:
|
||||
"""Record target signal (dist_cyl + sensors, no pinball)."""
|
||||
_clean_cache()
|
||||
warmup = int(4.0 * NX / U0)
|
||||
sim = Simulation(lbm_config_path=config_path, device_id=device_id)
|
||||
sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
@ -242,7 +233,6 @@ class KarmanCloakEnv(gym.Env):
|
||||
if self.target_states is None:
|
||||
self.target_states = record_target(self._config_path, self.device_id, self._si)
|
||||
|
||||
_clean_cache()
|
||||
self.sim = Simulation(lbm_config_path=self._config_path, device_id=self.device_id)
|
||||
self.sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
|
||||
|
||||
@ -36,14 +36,6 @@ if str(_REPO) not in sys.path:
|
||||
from CelerisLab import Simulation
|
||||
from CelerisLab.lbm.initializers import add_vortex
|
||||
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
def _clean_cache():
|
||||
for p in [_CONFIG_H, _PTX]:
|
||||
if p.exists(): p.unlink()
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Geometry constants
|
||||
# ---------------------------------------------------------------------------
|
||||
@ -177,7 +169,6 @@ class VortexCloakEnv(gym.Env):
|
||||
|
||||
# ---- Phase 1: Target (sensors + vortex only, no pinball) ----
|
||||
print(" [vortex] Phase 1: Target recording...", flush=True)
|
||||
_clean_cache()
|
||||
sim_t = Simulation(lbm_config_path=self._config_path, device_id=self.device_id)
|
||||
sim_t._assert_object_count_contract = lambda *a, **kw: None
|
||||
s0 = sim_t.add_body("sensor", center=(SENSOR_X, CENTER_Y + 40.0, 0.0), radius=5.0)
|
||||
@ -209,7 +200,6 @@ class VortexCloakEnv(gym.Env):
|
||||
|
||||
# ---- Phase 2: Training sim (sensors + pinball + vortex) ----
|
||||
print(" [vortex] Phase 2: Training sim...", flush=True)
|
||||
_clean_cache()
|
||||
self.sim = Simulation(lbm_config_path=self._config_path, device_id=self.device_id)
|
||||
self.sim._assert_object_count_contract = lambda *a, **kw: None
|
||||
self.sensor_ids = [
|
||||
|
||||
@ -19,13 +19,11 @@
|
||||
#
|
||||
# Requirements:
|
||||
# - conda env pycuda_3_10
|
||||
# - CelerisLab at /home/frank14f/CelerisLab
|
||||
# - CelerisLab submodule at DynamisLab/CelerisLab
|
||||
# - train_karman.py, env_karman.py, symmetry_wrapper.py in same directory
|
||||
set -euo pipefail
|
||||
|
||||
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
|
||||
CACHE_H="${HOME}/CelerisLab/src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
CACHE_PTX="${HOME}/CelerisLab/src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
# --- Defaults ---
|
||||
CASE_NAME=""
|
||||
@ -97,10 +95,6 @@ echo " Delay: ${DELAY_SECONDS}s between launches"
|
||||
echo " Jobs: ${#SEED_ARR[@]}"
|
||||
echo ""
|
||||
|
||||
# Clean stale cache before starting
|
||||
rm -f "$CACHE_H" "$CACHE_PTX"
|
||||
echo " Cleaned kernel cache."
|
||||
|
||||
# Build transfer arg
|
||||
TRANSFER_ARG=""
|
||||
if [[ -n "$TRANSFER" ]]; then
|
||||
|
||||
@ -38,7 +38,7 @@ if str(_REPO) not in sys.path:
|
||||
|
||||
from CelerisLab import Simulation
|
||||
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CELERIS = _REPO / "CelerisLab"
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
@ -37,7 +37,7 @@ from CelerisLab import Simulation
|
||||
# ---------------------------------------------------------------------------
|
||||
# Hard-coded CelerisLab paths for cache cleaning
|
||||
# ---------------------------------------------------------------------------
|
||||
_CELERIS = Path("/home/frank14f/CelerisLab")
|
||||
_CELERIS = _REPO / "CelerisLab"
|
||||
_CONFIG_H = _CELERIS / "src/CelerisLab/lbm/kernels/config/config_objects.h"
|
||||
_PTX = _CELERIS / "src/CelerisLab/lbm/kernels/kernel.ptx"
|
||||
|
||||
@ -200,9 +200,9 @@ def main() -> int:
|
||||
elif ep % 5 == 0:
|
||||
log(f" Ep {ep:3d}: reward={avg_r:.4f} (best={best_reward:.4f}, {dt:.0f}s/ep)")
|
||||
|
||||
if ep % 10 == 0:
|
||||
model.save(str(out_dir / "models" / f"chkpt_ep{ep}.zip"))
|
||||
vec_env.save(norm_path)
|
||||
# Save model + eval summary after EVERY episode
|
||||
model.save(str(out_dir / "models" / f"ep{ep:04d}_model.zip"))
|
||||
vec_env.save(str(out_dir / "models" / f"ep{ep:04d}_vecnormalize.pkl"))
|
||||
|
||||
model.save(str(out_dir / "models" / "final_model.zip"))
|
||||
vec_env.save(norm_path)
|
||||
|
||||
Loading…
Reference in New Issue
Block a user