Initial upload: FaustineCNN best checkpoint + cutoffs + classes

README.md

@@ -0,0 +1,52 @@
+---
+license: mit
+tags:
+  - nilm
+  - energy-disaggregation
+  - uk-dale
+  - pytorch
+library_name: nilmbench
+datasets:
+  - Pybunny/nilmbench-ukdale
+language:
+  - en
+---
+
+# FaustineCNN — NILMbench baseline
+
+FaustineCNN trained on UK-DALE House 1 (sparse 6-second 16 kHz V/I frames) and
+evaluated on the dense House 2 set of 60,000 frames. This is the best-scoring
+baseline from the NILMbench paper (`MJ_{20W} = 0.504` raw,
+`0.559` with the recall-constrained cutoffs in `cutoffs.json`).
+
+## Files
+
+| File              | Purpose                                                              |
+| ----------------- | -------------------------------------------------------------------- |
+| `faustine_best.pt`| PyTorch state-dict, loadable with `nilmbench.models.FaustineCNN`.    |
+| `cutoffs.json`    | Recall-constrained per-class cutoffs `c_k` (calibrated on House 1).  |
+| `classes.json`    | Ordered category names (length 7).                                   |
+
+## Quick start
+
+```python
+import torch, json
+from huggingface_hub import hf_hub_download
+from nilmbench.models import FaustineCNN
+
+ckpt = hf_hub_download("Pybunny/nilmbench-faustine", "faustine_best.pt")
+classes = json.load(open(hf_hub_download("Pybunny/nilmbench-faustine", "classes.json")))
+
+model = FaustineCNN(n_categories=len(classes))
+model.load_state_dict(torch.load(ckpt, map_location="cpu"))
+model.eval()
+```
+
+## Citation
+
+NILMbench paper (2026). See https://github.com/Saharmgh/NILMbench for the
+full code and reproducible-figure scripts.
+
+## License
+
+MIT.

classes.json

@@ -0,0 +1,9 @@
+[
+  "always on",
+  "cooking",
+  "dishwasher",
+  "electronics & lighting",
+  "fridge",
+  "misc",
+  "washing machine"
+]

cutoffs.json

@@ -0,0 +1,44 @@
+{
+  "classes": [
+    "always on",
+    "cooking",
+    "dishwasher",
+    "electronics & lighting",
+    "fridge",
+    "misc",
+    "washing machine"
+  ],
+  "cutoffs_W": {
+    "always on": 3.0,
+    "cooking": 1474.240966796875,
+    "dishwasher": 104.46902465820312,
+    "electronics & lighting": 5.0,
+    "fridge": 61.15494155883789,
+    "misc": 11.852706909179688,
+    "washing machine": 14.989958763122559
+  },
+  "dense_after": {
+    "F1": 0.8084632287157288,
+    "Jaccard": 0.7267236904761906,
+    "MAE_W": 27.093177795410156,
+    "MF_20pct": 0.29055909090909093,
+    "MJ_abs_20W": 0.5597594444444445,
+    "MJ_rel_20pct": 0.26448138888888884,
+    "StateAcc_BalancedMacro": 0.645689942881223,
+    "StateAcc_ExactMatch": 0.3600833333333333,
+    "StateAcc_Hamming": 0.8766452380952381,
+    "TECA": 0.7393695116043091
+  },
+  "dense_before": {
+    "F1": 0.7871189225589227,
+    "Jaccard": 0.6958458333333335,
+    "MAE_W": 29.638530731201172,
+    "MF_20pct": 0.2634095466570467,
+    "MJ_abs_20W": 0.5025619841269842,
+    "MJ_rel_20pct": 0.23849805555555553,
+    "StateAcc_BalancedMacro": 0.6855679513646701,
+    "StateAcc_ExactMatch": 0.27818333333333334,
+    "StateAcc_Hamming": 0.8425142857142857,
+    "TECA": 0.7056184411048889
+  }
+}

faustine_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ba2aae21d5b86f70bd966599cb1fa5096437761421b6f2c4de3811cada31fd5
+size 1033238