Use precomputed exposure embeddings

README.md

@@ -68,9 +68,8 @@ python train_exposure_autoencoder.py \
   --val_eid_file ukb_val_eid.csv
 ```
 
-The best checkpoint contains both `model_state_dict`, an `encoder_state_dict`
-compatible with the default gated `TimesNetExposureEncoder`, and the channel
-normalization statistics needed when the encoder is attached to DeepHealth.
+The best checkpoint contains the encoder and normalization statistics needed
+to generate fixed exposure embeddings.
 Multi-GPU pretraining follows the main trainer interface: add
 `--data_parallel --gpu_ids 0,1,2,3`.
 For efficient multi-GPU training, launch one process per GPU with DDP:
@@ -84,13 +83,21 @@ torchrun --standalone --nproc_per_node=4 train_exposure_autoencoder.py \
 In DDP mode, `--batch_size` is the global batch size and must be divisible by
 the number of processes.
 
-The end-to-end next-step trainer supports the same DDP launch pattern:
+Encode every cached exposure window once:
+
+```bash
+python encode_exposure_cache.py \
+  --exposure_cache_dir ukb_exposure_cache \
+  --checkpoint runs/exposure_ae_RUN/best.pt
+```
+
+The next-step trainer reads `exposure_embeddings.npy` directly and does not
+run TimesNet:
 
 ```bash
 torchrun --standalone --nproc_per_node=4 train_next_step.py \
   --exposure_cache_dir ukb_exposure_cache \
-  --batch_size 128 \
-  --d_model 64
+  --batch_size 128
 ```
 Training-channel statistics are cached at
 `<exposure_cache_dir>/train_channel_stats.npz`; use