Upload MiniHack ReMDM planner checkpoints and results

README.md

@@ -17,7 +17,7 @@ tags:
 Trained weights accompanying *The Double Intractability of Reinforcement Learning for Discrete Diffusion Planners*: a remasking discrete diffusion model
 (ReMDM) used as an action-sequence planner in
 [MiniHack](https://github.com/facebookresearch/minihack), together with the BFS
-oracle rollouts that supervise it.
+oracle rollouts that supervise it, and the results reported in the paper.
 
 Code, configs and evaluation harness: https://github.com/mathisweil/minihack-ReMDM-planner
 
@@ -26,7 +26,7 @@ Code, configs and evaluation harness: https://github.com/mathisweil/minihack-ReM
 | Path | Role | Environment | Architecture | Selected at | Training | Size |
 |---|---|---|---|---|---|---|
 | `checkpoints/offline/Minihack-Offline-Diffusion-BC-100M` | Diffusion planner (offline BC) | `MiniHack` | 4L, d_model 256, 4 heads, horizon 64, 5M params | gradient step 50,000 | 102,400,000 sample-equivalents | 100 MB |
-| `checkpoints/online/Minihack-Online-Diffusion-DAgger-6M` | Diffusion planner (online DAgger) | `MiniHack` | 4L, d_model 256, 4 heads, horizon 64, 5M params | iteration 563 | 5,654,965 env steps | 100 MB |
+| `checkpoints/online/Minihack-Online-Diffusion-DAgger-100M` | Diffusion planner (online DAgger) | `MiniHack` | 4L, d_model 256, 4 heads, horizon 64, 5M params | iteration 563 | 5,654,965 env steps | 100 MB |
 
 Each checkpoint ships the `.pth` training state it was published from (weights,
 EMA shadow, optimiser, scheduler, and for the DAgger run the curriculum and RNG
@@ -38,9 +38,6 @@ Weights are PyTorch training states with a `safetensors` export of the EMA
 weights alongside, and the paths above mirror the source repository so a
 snapshot can be dropped straight into a working copy.
 
-`results/` holds the evaluation and ablation tables reported in the paper, as
-produced by `experiments/rl_finetuning`. Figures and raw logs stay in the code
-repository.
 
 ## Download
 
@@ -54,7 +51,7 @@ snapshot_download(repo_id="mathisweil/remdm-minihack-checkpoints", local_dir="."
 snapshot_download(
     repo_id="mathisweil/remdm-minihack-checkpoints",
     local_dir=".",
-    allow_patterns="checkpoints/online/Minihack-*/**",
+    allow_patterns="checkpoints/offline/Minihack-Offline-Diffusion-BC-100M/**",
 )
 ```
 
@@ -63,9 +60,10 @@ snapshot_download(
 From a clone of the code repository, after downloading into it:
 
 ```bash
-DIR=checkpoints/online/Minihack-OnlineDiffusion-DAgger-123M
+DIR=checkpoints/offline/Minihack-Offline-Diffusion-BC-100M
 uv run python main.py --mode inference \
-    --config $DIR/config_iter600.yaml --checkpoint $DIR/iter600.pth
+    --config $DIR/config.yaml --checkpoint $DIR/offline_step50000.pth \
+    --output results/inference/eval.json
 ```
 
 Programmatic loading uses `src.models.denoiser.make_model` with the checkpoint's
@@ -76,9 +74,9 @@ from safetensors.torch import load_file
 from src.config import load_config
 from src.models.denoiser import make_model
 
-cfg = load_config(f"{DIR}/config_iter600.yaml")
+cfg = load_config("checkpoints/offline/Minihack-Offline-Diffusion-BC-100M/config.yaml")
 model = make_model(cfg)
-model.load_state_dict(load_file(f"{DIR}/model.safetensors"))
+model.load_state_dict(load_file("checkpoints/offline/Minihack-Offline-Diffusion-BC-100M/model.safetensors"))
 model.eval()
 ```
 
@@ -97,14 +95,13 @@ strategy, schedule and sampling settings, are in the per-checkpoint config
 snapshots listed above, which are the authoritative record.
 
 Both models are best-checkpoint selections rather than final-step dumps: each
-trainer evaluates every periodic checkpoint on 50 episodes per environment
-using EMA weights, and the highest-scoring one is published (the metric behind that selection is not recorded in this release).
-Directory names encode the sample-equivalents the published model consumed
-(gradient steps x batch size, rounded); file names carry each trainer's own
-counter, DAgger iterations online and gradient steps offline. The offline run
-was given the DAgger-matched budget of 60,000 gradient steps and its best
-checkpoint fell at 40,000, so the two published models sit at different points
-on a matched budget.
+trainer evaluates every periodic checkpoint on its configured number of
+episodes per environment using EMA weights, and the highest-scoring one is
+published (the metric behind that selection is not recorded in this release). Directory names encode the sample-equivalents the
+published model consumed (gradient steps x batch size, rounded); file names
+carry each trainer's own counter, DAgger iterations online and gradient steps
+offline. Each checkpoint's `selection.json` records the configured budget it
+was drawn from and the step it was selected at.
 
 ## Limitations