Planar Capacitance Model Space (QCap, QCap-Lite, QCap-Fast) from registry v2 (earlier contents: 906bc45ae05c)
Browse files- README.md +9 -9
- app.py +4 -4
- geomgnn/api.py +3 -3
- geomgnn/app_flagship.py +42 -31
- geomgnn/flagship.py +28 -2
- geomgnn/ui.py +1 -1
- models/accurate/info.json +1 -0
- models/balanced/info.json +1 -0
- models/fast/info.json +1 -0
- registry.yaml +35 -16
README.md
CHANGED
|
@@ -7,19 +7,19 @@ sdk: docker
|
|
| 7 |
app_port: 7860
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
-
short_description: 'Maxwell capacitance of planar layouts:
|
| 11 |
hf_oauth: true
|
| 12 |
---
|
| 13 |
|
| 14 |
# Planar Capacitance Model
|
| 15 |
|
| 16 |
-
The model of our paper *
|
| 17 |
|
| 18 |
- **Paper page:** [lfl-lab.github.io/SQuADDS/ml-paper](https://lfl-lab.github.io/SQuADDS/ml-paper/), with the explainer videos and the [tutorials](https://lfl-lab.github.io/SQuADDS/ml-paper/#tutorials)
|
| 19 |
- **Paper:** [lfl-lab.github.io/SQuADDS/ml-paper/paper](https://lfl-lab.github.io/SQuADDS/ml-paper/paper/)
|
| 20 |
- **Code:** [github.com/LFL-Lab/SQuADDS](https://github.com/LFL-Lab/SQuADDS)
|
| 21 |
- **Data:** [Q3D dataset](https://huggingface.co/datasets/SQuADDS/SQuADDS_PlanarCapacitance) and [synthetic dataset](https://huggingface.co/datasets/SQuADDS/SQuADDS_PlanarCapacitance-Synthetic), part of the [SQuADDS datasets](https://huggingface.co/collections/SQuADDS/squadds-datasets-6a65152b857f8efa5620880a)
|
| 22 |
-
- **Models:** [
|
| 23 |
- **API:** [SQuADDS ML inference API](https://huggingface.co/spaces/SQuADDS/squadds-ml-inference-api)
|
| 24 |
- **Point of contact:** [Sadman Ahmed Shanto](mailto:shanto@usc.edu)
|
| 25 |
|
|
@@ -33,13 +33,13 @@ The model of our paper *A physics-corrected graph neural network for the capacit
|
|
| 33 |
|
| 34 |
| model | how it works | error on 39 new layouts, % (MAE / RMSE, fF) | median time |
|
| 35 |
|---|---|---|---|
|
| 36 |
-
|
|
| 37 |
-
|
|
| 38 |
-
| Fast | a learned prior replaces the solves, then the networks correct it | 3.175 (0.583 / 0.824) | 0.24 s |
|
| 39 |
|
| 40 |
The 39 layouts are generated rings with a neighbour (3 or 4 conductors) that no part of any model saw in training. On them the extrapolated prior alone scores 1.765 % (MAE 0.275 / RMSE 0.376 fF). Times are on two CPUs of an Intel Xeon E5-2640 v4 cluster node.
|
| 41 |
|
| 42 |
-
Fast is for quick screening of the component families it was trained on. Use
|
| 43 |
|
| 44 |
## API
|
| 45 |
|
|
@@ -50,12 +50,12 @@ import os
|
|
| 50 |
from gradio_client import Client, handle_file
|
| 51 |
|
| 52 |
c = Client("shanto268/PlanarCapacitanceModel", oauth_token=os.environ["HF_TOKEN"]) # counts as you for the daily limit
|
| 53 |
-
r = c.predict(handle_file("layout.gds"), "1/0", "1/10", 0, "Si (cryogenic, training)", 11.45, 350, "
|
| 54 |
api_name="/predict_gds")
|
| 55 |
print(r["names"], r["matrix_fF"], r["timing_s"]["total_s"])
|
| 56 |
```
|
| 57 |
|
| 58 |
-
`model` is `
|
| 59 |
|
| 60 |
## Limits of this demo
|
| 61 |
|
|
|
|
| 7 |
app_port: 7860
|
| 8 |
app_file: app.py
|
| 9 |
pinned: false
|
| 10 |
+
short_description: 'Maxwell capacitance of planar layouts: QCap models'
|
| 11 |
hf_oauth: true
|
| 12 |
---
|
| 13 |
|
| 14 |
# Planar Capacitance Model
|
| 15 |
|
| 16 |
+
The model of our paper *QCap: a physics-corrected graph neural network for the capacitance matrix of any planar superconducting device layout*, trained on all 45,771 designs of our Q3D dataset, the test designs included. Upload a GDS layout and get the full Maxwell capacitance matrix of every conductor in it. Because the model has seen the test designs, it has no held-out score of its own. Its held-out accuracy, from the same model trained without the test designs, is in the paper.
|
| 17 |
|
| 18 |
- **Paper page:** [lfl-lab.github.io/SQuADDS/ml-paper](https://lfl-lab.github.io/SQuADDS/ml-paper/), with the explainer videos and the [tutorials](https://lfl-lab.github.io/SQuADDS/ml-paper/#tutorials)
|
| 19 |
- **Paper:** [lfl-lab.github.io/SQuADDS/ml-paper/paper](https://lfl-lab.github.io/SQuADDS/ml-paper/paper/)
|
| 20 |
- **Code:** [github.com/LFL-Lab/SQuADDS](https://github.com/LFL-Lab/SQuADDS)
|
| 21 |
- **Data:** [Q3D dataset](https://huggingface.co/datasets/SQuADDS/SQuADDS_PlanarCapacitance) and [synthetic dataset](https://huggingface.co/datasets/SQuADDS/SQuADDS_PlanarCapacitance-Synthetic), part of the [SQuADDS datasets](https://huggingface.co/collections/SQuADDS/squadds-datasets-6a65152b857f8efa5620880a)
|
| 22 |
+
- **Models:** [QCap models](https://huggingface.co/SQuADDS/QCap) (QCap, QCap-Lite and QCap-Fast), part of the [SQuADDS models](https://huggingface.co/collections/SQuADDS/squadds-models-6af8b7da67a22df62af374c6)
|
| 23 |
- **API:** [SQuADDS ML inference API](https://huggingface.co/spaces/SQuADDS/squadds-ml-inference-api)
|
| 24 |
- **Point of contact:** [Sadman Ahmed Shanto](mailto:shanto@usc.edu)
|
| 25 |
|
|
|
|
| 33 |
|
| 34 |
| model | how it works | error on 39 new layouts, % (MAE / RMSE, fF) | median time |
|
| 35 |
|---|---|---|---|
|
| 36 |
+
| QCap (default) | the extrapolated prior from two coarse method-of-moments solves, corrected by the networks | 1.05 (0.132 / 0.188) | 1.1 s |
|
| 37 |
+
| QCap-Lite | the same recipe on two coarser solves | 2.044 (0.258 / 0.348) | 0.56 s |
|
| 38 |
+
| QCap-Fast | a learned prior replaces the solves, then the networks correct it | 3.175 (0.583 / 0.824) | 0.24 s |
|
| 39 |
|
| 40 |
The 39 layouts are generated rings with a neighbour (3 or 4 conductors) that no part of any model saw in training. On them the extrapolated prior alone scores 1.765 % (MAE 0.275 / RMSE 0.376 fF). Times are on two CPUs of an Intel Xeon E5-2640 v4 cluster node.
|
| 41 |
|
| 42 |
+
QCap-Fast is for quick screening of the component families it was trained on. Use QCap otherwise.
|
| 43 |
|
| 44 |
## API
|
| 45 |
|
|
|
|
| 50 |
from gradio_client import Client, handle_file
|
| 51 |
|
| 52 |
c = Client("shanto268/PlanarCapacitanceModel", oauth_token=os.environ["HF_TOKEN"]) # counts as you for the daily limit
|
| 53 |
+
r = c.predict(handle_file("layout.gds"), "1/0", "1/10", 0, "Si (cryogenic, training)", 11.45, 350, "QCap",
|
| 54 |
api_name="/predict_gds")
|
| 55 |
print(r["names"], r["matrix_fF"], r["timing_s"]["total_s"])
|
| 56 |
```
|
| 57 |
|
| 58 |
+
`model` is `QCap`, `QCap-Lite` or `QCap-Fast` (any case). When ground and conductors share a layer, pass the ground's piece numbers as the last argument (`ground_pieces`, for example `"1"` or `"1,4"`) if the automatic pick is wrong.
|
| 59 |
|
| 60 |
## Limits of this demo
|
| 61 |
|
app.py
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
-
"""Flagship Space (deploy/GNN_SPACE.md "Flagship Space"): GDS -> full Maxwell capacitance matrix with three
|
| 2 |
-
|
| 3 |
(flagship_hub, flagship_tiers)."""
|
| 4 |
|
| 5 |
import os
|
|
@@ -19,7 +19,7 @@ os.environ.setdefault("TLSQ_RESPONSE_CACHE", "64")
|
|
| 19 |
os.environ.setdefault("TLSQ_RESPONSE_CACHE_MB", "1024")
|
| 20 |
os.environ.setdefault("TLSQ_EMBED_CACHE", os.path.join(os.environ.get("TMPDIR", "/tmp"), "tlsq_embed_cache"))
|
| 21 |
os.environ.setdefault("TLSQ_EMBED_CACHE_MB", "1024")
|
| 22 |
-
os.environ.setdefault("TLSQ_PRIOR_LEVEL_CACHE_MB", "256") # solved prior levels (
|
| 23 |
os.environ.pop("TLSQ_GNN_PRIOR_MODE", None) # every tier names its own solver mode (registry flagship_hub.prior_stacks)
|
| 24 |
|
| 25 |
HERE = Path(__file__).resolve().parent
|
|
@@ -76,7 +76,7 @@ def _preload_background():
|
|
| 76 |
|
| 77 |
|
| 78 |
def bake():
|
| 79 |
-
"""Image build (warm.py, after the import's preload): every example x
|
| 80 |
return G.bake(ex, BAKE)
|
| 81 |
|
| 82 |
|
|
|
|
| 1 |
+
"""Flagship Space (deploy/GNN_SPACE.md "Flagship Space"): GDS -> full Maxwell capacitance matrix with three models (QCap default,
|
| 2 |
+
QCap-Lite, QCap-Fast) retrained on all on-stack data, through geomgnn.flagship. Names, tiers and numbers come from registry.yaml
|
| 3 |
(flagship_hub, flagship_tiers)."""
|
| 4 |
|
| 5 |
import os
|
|
|
|
| 19 |
os.environ.setdefault("TLSQ_RESPONSE_CACHE_MB", "1024")
|
| 20 |
os.environ.setdefault("TLSQ_EMBED_CACHE", os.path.join(os.environ.get("TMPDIR", "/tmp"), "tlsq_embed_cache"))
|
| 21 |
os.environ.setdefault("TLSQ_EMBED_CACHE_MB", "1024")
|
| 22 |
+
os.environ.setdefault("TLSQ_PRIOR_LEVEL_CACHE_MB", "256") # solved prior levels (QCap and QCap-Lite share 1.5) and meshes (substrate changes)
|
| 23 |
os.environ.pop("TLSQ_GNN_PRIOR_MODE", None) # every tier names its own solver mode (registry flagship_hub.prior_stacks)
|
| 24 |
|
| 25 |
HERE = Path(__file__).resolve().parent
|
|
|
|
| 76 |
|
| 77 |
|
| 78 |
def bake():
|
| 79 |
+
"""Image build (warm.py, after the import's preload): every example x model into the response cache, pickled to BAKE."""
|
| 80 |
return G.bake(ex, BAKE)
|
| 81 |
|
| 82 |
|
geomgnn/api.py
CHANGED
|
@@ -53,9 +53,9 @@ CALIBRATED_SCORE = ("On 15 held-out designs the mutual entries were within 0.21-
|
|
| 53 |
"0.23 % (MAE 0.080 fF, RMSE 0.111 fF) on the training stack. The self terms were within 1.22 % (MAE up to 0.92 fF, "
|
| 54 |
"RMSE up to 1.07 fF).")
|
| 55 |
FAST_ROUTE = "fast: eps-rescaled training-stack prediction; thickness not modelled"
|
| 56 |
-
FAST_WARNING = ("Fast does not solve the prior at your stack. It rescales its 350 µm silicon answer by (1 + eps_r) / (1 + 11.45) and "
|
| 57 |
"ignores thickness. Off the stack we measured a median error of 6.4 % (MAE 1.7 fF, RMSE 2.4 fF), and 13 % (3.4 / 4.7 fF) "
|
| 58 |
-
"across Si thicknesses of 10-1000 µm. Use
|
| 59 |
"thickness and gets 0.48 % (0.13 / 0.18 fF) on the same solves.")
|
| 60 |
|
| 61 |
|
|
@@ -483,7 +483,7 @@ def physical_warning(chk: dict, C: np.ndarray) -> str | None:
|
|
| 483 |
if not bad:
|
| 484 |
return None
|
| 485 |
return ("physical check failed: " + ", ".join(bad) + ". This matrix is not physical, so do not use it. A model that solves the "
|
| 486 |
-
"physics, such as
|
| 487 |
|
| 488 |
|
| 489 |
def predict_gnn(prepared, name: str, reg, substrate="Si (cryogenic, training)", thickness_um: float = 350.0, model_dir=None,
|
|
|
|
| 53 |
"0.23 % (MAE 0.080 fF, RMSE 0.111 fF) on the training stack. The self terms were within 1.22 % (MAE up to 0.92 fF, "
|
| 54 |
"RMSE up to 1.07 fF).")
|
| 55 |
FAST_ROUTE = "fast: eps-rescaled training-stack prediction; thickness not modelled"
|
| 56 |
+
FAST_WARNING = ("QCap-Fast does not solve the prior at your stack. It rescales its 350 µm silicon answer by (1 + eps_r) / (1 + 11.45) and "
|
| 57 |
"ignores thickness. Off the stack we measured a median error of 6.4 % (MAE 1.7 fF, RMSE 2.4 fF), and 13 % (3.4 / 4.7 fF) "
|
| 58 |
+
"across Si thicknesses of 10-1000 µm. Use QCap for other substrates. It solves the prior at your eps_r and "
|
| 59 |
"thickness and gets 0.48 % (0.13 / 0.18 fF) on the same solves.")
|
| 60 |
|
| 61 |
|
|
|
|
| 483 |
if not bad:
|
| 484 |
return None
|
| 485 |
return ("physical check failed: " + ", ".join(bad) + ". This matrix is not physical, so do not use it. A model that solves the "
|
| 486 |
+
"physics, such as QCap, is the safer choice, especially outside the trained component families.")
|
| 487 |
|
| 488 |
|
| 489 |
def predict_gnn(prepared, name: str, reg, substrate="Si (cryogenic, training)", thickness_um: float = 350.0, model_dir=None,
|
geomgnn/app_flagship.py
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
-
"""Gradio front end of the flagship Space (deploy/GNN_SPACE.md "Flagship Space"; registry flagship_hub, flagship_tiers): three
|
| 2 |
-
one selector (
|
| 3 |
order as the GNN Space:
|
| 4 |
|
| 5 |
-
/models () -> served
|
| 6 |
/inspect (gds file) -> layer table, layer plot
|
| 7 |
/prepare (gds, layers table, crop, margin_gaps, margin_um, select) -> preview, warnings, summary, simplified GDS
|
| 8 |
/predict (gds, layers table, crop, margin_gaps, margin_um, select, substrate, custom_eps, thickness_um, model) -> matrix, CSV, JSON, report
|
| 9 |
/predict_gds (gds, ground "1/0", conductor "1/10", margin_gaps, substrate, custom_eps, thickness_um, model, ground_pieces "") -> result dict;
|
| 10 |
-
model =
|
| 11 |
(default "": detected automatically, see result warnings / layout info)
|
| 12 |
|
| 13 |
Every endpoint and the UI share the demo's limits (geomgnn.access): anonymous visitors run the bundled examples only, a signed-in user
|
|
@@ -54,11 +54,19 @@ def served(reg=None) -> list[str]:
|
|
| 54 |
return sorted(ks, key=lambda k: ORDER.index(k) if k in ORDER else len(ORDER))
|
| 55 |
|
| 56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
def resolve(reg, model) -> str:
|
| 58 |
-
|
| 59 |
-
if
|
| 60 |
-
|
| 61 |
-
|
|
|
|
|
|
|
|
|
|
| 62 |
|
| 63 |
|
| 64 |
def tiers_md(reg) -> str:
|
|
@@ -91,7 +99,7 @@ def check_thickness(thickness_um) -> float:
|
|
| 91 |
|
| 92 |
def settings(reg, name, substrate, thickness_um) -> tuple:
|
| 93 |
"""The response-cache settings of a request (with the prepared layout, the cache key)."""
|
| 94 |
-
return ("
|
| 95 |
|
| 96 |
|
| 97 |
def _predict(lay, model, substrate, thickness_um, w=None, own=False):
|
|
@@ -216,7 +224,7 @@ def _api_layout(f, ground_layers="1/0", conductor_layers="1/10", margin_gaps=0.0
|
|
| 216 |
|
| 217 |
|
| 218 |
def api_predict_gds(f, ground_layers="1/0", conductor_layers="1/10", margin_gaps=0.0, substrate=SI,
|
| 219 |
-
custom_eps=11.45, thickness_um=350.0, model="
|
| 220 |
oauth_token: gr.OAuthToken | None = None, request: gr.Request = None):
|
| 221 |
t = time.perf_counter()
|
| 222 |
lay, res = _served(AC.who(profile, request, oauth_token), f, lambda: _api_layout(f, ground_layers, conductor_layers, margin_gaps,
|
|
@@ -228,7 +236,7 @@ def api_predict_gds(f, ground_layers="1/0", conductor_layers="1/10", margin_gaps
|
|
| 228 |
|
| 229 |
|
| 230 |
def bake(examples_dir, path) -> dict:
|
| 231 |
-
"""Image build: every bundled example x
|
| 232 |
into the response cache, pickled to path. Both paths prepare the same layout, so they share one entry per example and tier."""
|
| 233 |
from . import respcache
|
| 234 |
reg = load_registry()
|
|
@@ -236,9 +244,9 @@ def bake(examples_dir, path) -> dict:
|
|
| 236 |
for f in sorted(Path(examples_dir).glob("*.gds")):
|
| 237 |
table = inspect_fn(str(f))[0]
|
| 238 |
gstate = GU.check_fn(str(f), table)[4]
|
| 239 |
-
for
|
| 240 |
-
api_predict_gds(str(f), model=
|
| 241 |
-
predict_fn(str(f), table, False, 0, None, "", SI, 11.45, 350,
|
| 242 |
out = {"entries": respcache.dump(path), "s": round(time.perf_counter() - t, 1)}
|
| 243 |
print("bake " + json.dumps(out), flush=True)
|
| 244 |
return out
|
|
@@ -260,7 +268,7 @@ def load_bake(path, examples_dir) -> dict:
|
|
| 260 |
lay = _api_layout(str(Path(examples_dir) / "interdigital.gds"))
|
| 261 |
old = baked.get(respcache.key(lay, *settings(reg, tier, SI, 350.0)), (None,))[0]
|
| 262 |
if old is None or not _same(old, FS.predict(lay, tier, reg, substrate=SI, thickness_um=350.0)):
|
| 263 |
-
out = {"bake": "dropped", "reason": f"interdigital / {tier} differs from the baked result", "s": time.perf_counter() - t}
|
| 264 |
break
|
| 265 |
else:
|
| 266 |
out = {"bake": "loaded", "entries": len(items), "cache_entries": respcache.put_many(items), "s": time.perf_counter() - t}
|
|
@@ -274,14 +282,15 @@ def models_fn():
|
|
| 274 |
out = []
|
| 275 |
for k in served(reg):
|
| 276 |
t = FS.tiers(reg)[k]
|
| 277 |
-
out.append({"id":
|
| 278 |
"own": t.get("own") or {}, "b5": t.get("b5"), "b5_physics": t.get("b5_physics"), "time_space": t.get("time_space"), "time_x86": t.get("time_x86"),
|
| 279 |
"stack": t.get("stack"), "levels_ppg": [float(p) for p in t.get("ppg") or []], "default": k == h["default_arm"],
|
| 280 |
"max_conductors": int(h["max_conductors"]), "limits": limits()})
|
| 281 |
from . import paper as PA
|
| 282 |
return out + [{"geomgnn_version": __version__, "tlsq_infer_version": TLSQ_VERSION, "registry_version": reg["version"],
|
| 283 |
"training_layouts": h.get("training_layouts"),
|
| 284 |
-
"startup": dict(PA.STARTUP
|
|
|
|
| 285 |
|
| 286 |
|
| 287 |
def packages() -> dict:
|
|
@@ -316,7 +325,7 @@ PRIVACY_MD = ("We process uploaded layouts in memory and keep them there only wh
|
|
| 316 |
VIEWS = ["Layers", "Nets"]
|
| 317 |
EXAMPLE_NAMES = {"interdigital.gds": "interdigital capacitor", "lattice_1q.gds": "one-qubit grid", "lattice_5q.gds": "five-qubit grid",
|
| 318 |
"shielded_concentric.gds": "shielded concentric", "transmon_pocket.gds": "TransmonPocket", "xmon.gds": "Xmon"}
|
| 319 |
-
EXAMPLE_NOTES = {"lattice_1q.gds": "≈20 s on
|
| 320 |
LAYER_HELP = ("Set the role of each layer by picking it from the list or clicking its row. Use `ground` for the ground plane (one net), "
|
| 321 |
"`conductor` for metal islands (each connected body is one net, C1, C2, …), `ground+conductor` when both are on one layer "
|
| 322 |
"(we then find the ground) and `ignore` for markers, junctions and labels. By default 1/0 is ground and 1/10 is conductor.")
|
|
@@ -325,7 +334,7 @@ LAYER_HELP = ("Set the role of each layer by picking it from the list or clickin
|
|
| 325 |
def about_md(reg) -> str:
|
| 326 |
h = reg["flagship_hub"]
|
| 327 |
|
| 328 |
-
return f"""The models handle planar layouts only, i.e., one metal layer of zero thickness on a dielectric substrate. Each model computes a thin-film physics prior (Fast learns it instead) and corrects it with networks trained on Q3D simulations. This is a private preview while the paper is in preparation.
|
| 329 |
|
| 330 |
|
| 331 |
|
|
@@ -336,12 +345,12 @@ You can also call the demo from Python with gradio_client. "Use via API" at the
|
|
| 336 |
```python
|
| 337 |
from gradio_client import Client, handle_file
|
| 338 |
c = Client("{h['space']}", token=HF_TOKEN, oauth_token=HF_TOKEN) # token while the Space is private, oauth_token signs you in (your own layouts)
|
| 339 |
-
r = c.predict(handle_file("layout.gds"), "1/0", "1/10", 0, "Si (cryogenic, training)", 11.45, 350, "{h['default_arm']}",
|
| 340 |
api_name="/predict_gds")
|
| 341 |
r["names"], r["matrix_fF"]
|
| 342 |
```
|
| 343 |
|
| 344 |
-
The arguments are, in order, the GDS file, the ground layers, the conductor layers, the crop margin in gaps (0 keeps the training geometry), the substrate, eps_r (for a custom substrate), the thickness in µm, the model ({', '.join(
|
| 345 |
|
| 346 |
{PRIVACY_MD}
|
| 347 |
|
|
@@ -501,8 +510,8 @@ def stack_note(reg, substrate, custom_eps, thickness) -> str:
|
|
| 501 |
if on:
|
| 502 |
return ""
|
| 503 |
return UI.callout("warn", "Off the training stack (Si, eps_r 11.45, 350 µm)", [
|
| 504 |
-
f"
|
| 505 |
-
f"{A.TESTED_RANGE}. Fast only rescales its prediction, and it tells you so."])
|
| 506 |
|
| 507 |
|
| 508 |
def _api(fn, n_out):
|
|
@@ -514,11 +523,11 @@ def build_app(title: str, examples_dir=None):
|
|
| 514 |
reg = load_registry()
|
| 515 |
subs = list(reg.substrates()) + [CUSTOM]
|
| 516 |
ms = served(reg)
|
| 517 |
-
default = reg["flagship_hub"]["default_arm"]
|
| 518 |
ex = sorted(Path(examples_dir).glob("*.gds")) if examples_dir and Path(examples_dir).exists() else []
|
| 519 |
AC.register_examples(ex)
|
| 520 |
usage_stats()
|
| 521 |
-
label = {"accurate": "
|
| 522 |
with UI.blocks(title, delete_cache=(60, PURGE_AGE_S)) as demo:
|
| 523 |
gr.HTML(UI.header(reg["flagship_hub"]["title"], LEDE), container=False, padding=False, elem_classes="tl-flush")
|
| 524 |
with gr.Row():
|
|
@@ -558,7 +567,7 @@ def build_app(title: str, examples_dir=None):
|
|
| 558 |
gparts = GU.build()
|
| 559 |
gst = gparts[0]
|
| 560 |
with gr.Row(equal_height=False):
|
| 561 |
-
model = gr.Radio([(label.get(k, k), k) for k in ms], value=default, label="Model", interactive=True, elem_id="tl-tier",
|
| 562 |
scale=3, min_width=300)
|
| 563 |
substrate = gr.Dropdown(subs, value=subs[0], label="Substrate", scale=2, min_width=220)
|
| 564 |
custom_eps = gr.Number(value=11.45, label="eps_r", visible=False, scale=1, min_width=110)
|
|
@@ -567,8 +576,8 @@ def build_app(title: str, examples_dir=None):
|
|
| 567 |
with gr.Accordion("Models and their accuracy", open=False):
|
| 568 |
gr.Markdown(tiers_md(reg))
|
| 569 |
if "fast" in ms:
|
| 570 |
-
gr.Markdown("Use Fast only for quick screening of the component families it was trained on. On new designs it did "
|
| 571 |
-
"worse than the physics alone (see the table above), so use
|
| 572 |
pred_btn = gr.Button("Predict", variant="primary")
|
| 573 |
status = gr.HTML(elem_classes="tl-flush")
|
| 574 |
with gr.Column(visible=False) as result_box, gr.Row():
|
|
@@ -591,6 +600,7 @@ def build_app(title: str, examples_dir=None):
|
|
| 591 |
a_c = gr.Textbox("1/10")
|
| 592 |
a_m = gr.Number(0)
|
| 593 |
a_gp = gr.Textbox("")
|
|
|
|
| 594 |
a_out = gr.JSON()
|
| 595 |
a_btn = gr.Button("api")
|
| 596 |
m_btn = gr.Button("models")
|
|
@@ -605,9 +615,10 @@ def build_app(title: str, examples_dir=None):
|
|
| 605 |
prepared_gds = gr.File()
|
| 606 |
pins = [gds, layers, crop, margin, margin_um, select, substrate, custom_eps, thickness, model, gst]
|
| 607 |
pouts = [mat, csv_f, json_f, report, res_state]
|
|
|
|
| 608 |
vin = [view, vst, gds, layers, crop, margin, margin_um, select, gst]
|
| 609 |
vout = [img, svg_btn, view_msg, vst]
|
| 610 |
-
ui_predict = functools.wraps(predict_fn)(UI.predict_ui(predict_fn, lambda r: expected(reg, r["tier"]), note=True))
|
| 611 |
ui_predict.__name__ = "predict_fn_ui"
|
| 612 |
uouts = pouts + [status, result_box, mat_note]
|
| 613 |
light = dict(concurrency_id="light", concurrency_limit=1) # previews and file handling: never queued behind a solve
|
|
@@ -630,9 +641,9 @@ def build_app(title: str, examples_dir=None):
|
|
| 630 |
for ev in (crop.change, margin.change, margin_um.change, select.change):
|
| 631 |
ev(preview_fn, vin, [view] + vout, trigger_mode="always_last", **hidden)
|
| 632 |
pred_btn.click(ui_predict, pins, uouts, api_visibility="private", **solve)
|
| 633 |
-
p_btn.click(_api(predict_fn, 5),
|
| 634 |
UI.mark_stale(res_state, status, mat, report, [gds.change, model.change, substrate.change, custom_eps.change, thickness.change,
|
| 635 |
select.change, crop.change, margin.change, margin_um.change, edit.then])
|
| 636 |
-
a_btn.click(_api(api_predict_gds, 1), [gds, a_g, a_c, a_m, substrate, custom_eps, thickness,
|
| 637 |
m_btn.click(models_fn, None, m_out, api_name="models", **light)
|
| 638 |
return demo
|
|
|
|
| 1 |
+
"""Gradio front end of the flagship Space (deploy/GNN_SPACE.md "Flagship Space"; registry flagship_hub, flagship_tiers): three models in
|
| 2 |
+
one selector (QCap default, QCap-Lite, QCap-Fast), served through geomgnn.flagship (= the paper Space's code path). Endpoints and argument
|
| 3 |
order as the GNN Space:
|
| 4 |
|
| 5 |
+
/models () -> served models with their reference numbers
|
| 6 |
/inspect (gds file) -> layer table, layer plot
|
| 7 |
/prepare (gds, layers table, crop, margin_gaps, margin_um, select) -> preview, warnings, summary, simplified GDS
|
| 8 |
/predict (gds, layers table, crop, margin_gaps, margin_um, select, substrate, custom_eps, thickness_um, model) -> matrix, CSV, JSON, report
|
| 9 |
/predict_gds (gds, ground "1/0", conductor "1/10", margin_gaps, substrate, custom_eps, thickness_um, model, ground_pieces "") -> result dict;
|
| 10 |
+
model = QCap (default) | QCap-Lite | QCap-Fast (any case, "-" or "_"); ground_pieces = e.g. "1,4" when ground and conductors share one layer
|
| 11 |
(default "": detected automatically, see result warnings / layout info)
|
| 12 |
|
| 13 |
Every endpoint and the UI share the demo's limits (geomgnn.access): anonymous visitors run the bundled examples only, a signed-in user
|
|
|
|
| 54 |
return sorted(ks, key=lambda k: ORDER.index(k) if k in ORDER else len(ORDER))
|
| 55 |
|
| 56 |
|
| 57 |
+
def names(reg) -> list[str]:
|
| 58 |
+
"""The served models' names (QCap, QCap-Lite, QCap-Fast), default first."""
|
| 59 |
+
return [FS.name(reg, k) for k in served(reg)]
|
| 60 |
+
|
| 61 |
+
|
| 62 |
def resolve(reg, model) -> str:
|
| 63 |
+
"""Tier key of a model name (QCap / QCap-Lite / QCap-Fast, any case; empty = the default)."""
|
| 64 |
+
if not str(model or "").strip():
|
| 65 |
+
return reg["flagship_hub"]["default_arm"]
|
| 66 |
+
k = FS.key(reg, model)
|
| 67 |
+
if k not in served(reg):
|
| 68 |
+
raise ValueError(f"Model {model!r} is not available here. Choose one of {', '.join(names(reg))}.")
|
| 69 |
+
return k
|
| 70 |
|
| 71 |
|
| 72 |
def tiers_md(reg) -> str:
|
|
|
|
| 99 |
|
| 100 |
def settings(reg, name, substrate, thickness_um) -> tuple:
|
| 101 |
"""The response-cache settings of a request (with the prepared layout, the cache key)."""
|
| 102 |
+
return ("qcap", FS.name(reg, name), str(substrate), float(thickness_um), reg["version"])
|
| 103 |
|
| 104 |
|
| 105 |
def _predict(lay, model, substrate, thickness_um, w=None, own=False):
|
|
|
|
| 224 |
|
| 225 |
|
| 226 |
def api_predict_gds(f, ground_layers="1/0", conductor_layers="1/10", margin_gaps=0.0, substrate=SI,
|
| 227 |
+
custom_eps=11.45, thickness_um=350.0, model="QCap", ground_pieces="", profile: gr.OAuthProfile | None = None,
|
| 228 |
oauth_token: gr.OAuthToken | None = None, request: gr.Request = None):
|
| 229 |
t = time.perf_counter()
|
| 230 |
lay, res = _served(AC.who(profile, request, oauth_token), f, lambda: _api_layout(f, ground_layers, conductor_layers, margin_gaps,
|
|
|
|
| 236 |
|
| 237 |
|
| 238 |
def bake(examples_dir, path) -> dict:
|
| 239 |
+
"""Image build: every bundled example x model on the training stack, through /predict_gds and the UI Predict path (an example click),
|
| 240 |
into the response cache, pickled to path. Both paths prepare the same layout, so they share one entry per example and tier."""
|
| 241 |
from . import respcache
|
| 242 |
reg = load_registry()
|
|
|
|
| 244 |
for f in sorted(Path(examples_dir).glob("*.gds")):
|
| 245 |
table = inspect_fn(str(f))[0]
|
| 246 |
gstate = GU.check_fn(str(f), table)[4]
|
| 247 |
+
for m in names(reg):
|
| 248 |
+
api_predict_gds(str(f), model=m)
|
| 249 |
+
predict_fn(str(f), table, False, 0, None, "", SI, 11.45, 350, m, gstate)
|
| 250 |
out = {"entries": respcache.dump(path), "s": round(time.perf_counter() - t, 1)}
|
| 251 |
print("bake " + json.dumps(out), flush=True)
|
| 252 |
return out
|
|
|
|
| 268 |
lay = _api_layout(str(Path(examples_dir) / "interdigital.gds"))
|
| 269 |
old = baked.get(respcache.key(lay, *settings(reg, tier, SI, 350.0)), (None,))[0]
|
| 270 |
if old is None or not _same(old, FS.predict(lay, tier, reg, substrate=SI, thickness_um=350.0)):
|
| 271 |
+
out = {"bake": "dropped", "reason": f"interdigital / {FS.name(reg, tier)} differs from the baked result", "s": time.perf_counter() - t}
|
| 272 |
break
|
| 273 |
else:
|
| 274 |
out = {"bake": "loaded", "entries": len(items), "cache_entries": respcache.put_many(items), "s": time.perf_counter() - t}
|
|
|
|
| 282 |
out = []
|
| 283 |
for k in served(reg):
|
| 284 |
t = FS.tiers(reg)[k]
|
| 285 |
+
out.append({"id": t["paper_name"], "title": t["title"], "summary": t.get("summary"), "warning": t.get("warning"),
|
| 286 |
"own": t.get("own") or {}, "b5": t.get("b5"), "b5_physics": t.get("b5_physics"), "time_space": t.get("time_space"), "time_x86": t.get("time_x86"),
|
| 287 |
"stack": t.get("stack"), "levels_ppg": [float(p) for p in t.get("ppg") or []], "default": k == h["default_arm"],
|
| 288 |
"max_conductors": int(h["max_conductors"]), "limits": limits()})
|
| 289 |
from . import paper as PA
|
| 290 |
return out + [{"geomgnn_version": __version__, "tlsq_infer_version": TLSQ_VERSION, "registry_version": reg["version"],
|
| 291 |
"training_layouts": h.get("training_layouts"),
|
| 292 |
+
"startup": dict(PA.STARTUP, **({"default": FS.name(reg, PA.STARTUP["default"])} if PA.STARTUP.get("default") in FS.tiers(reg) else {})),
|
| 293 |
+
"packages": packages()}]
|
| 294 |
|
| 295 |
|
| 296 |
def packages() -> dict:
|
|
|
|
| 325 |
VIEWS = ["Layers", "Nets"]
|
| 326 |
EXAMPLE_NAMES = {"interdigital.gds": "interdigital capacitor", "lattice_1q.gds": "one-qubit grid", "lattice_5q.gds": "five-qubit grid",
|
| 327 |
"shielded_concentric.gds": "shielded concentric", "transmon_pocket.gds": "TransmonPocket", "xmon.gds": "Xmon"}
|
| 328 |
+
EXAMPLE_NOTES = {"lattice_1q.gds": "≈20 s on QCap", "lattice_5q.gds": "≈20 s on QCap"}
|
| 329 |
LAYER_HELP = ("Set the role of each layer by picking it from the list or clicking its row. Use `ground` for the ground plane (one net), "
|
| 330 |
"`conductor` for metal islands (each connected body is one net, C1, C2, …), `ground+conductor` when both are on one layer "
|
| 331 |
"(we then find the ground) and `ignore` for markers, junctions and labels. By default 1/0 is ground and 1/10 is conductor.")
|
|
|
|
| 334 |
def about_md(reg) -> str:
|
| 335 |
h = reg["flagship_hub"]
|
| 336 |
|
| 337 |
+
return f"""The models handle planar layouts only, i.e., one metal layer of zero thickness on a dielectric substrate. Each model computes a thin-film physics prior (QCap-Fast learns it instead) and corrects it with networks trained on Q3D simulations. This is a private preview while the paper is in preparation.
|
| 338 |
|
| 339 |
|
| 340 |
|
|
|
|
| 345 |
```python
|
| 346 |
from gradio_client import Client, handle_file
|
| 347 |
c = Client("{h['space']}", token=HF_TOKEN, oauth_token=HF_TOKEN) # token while the Space is private, oauth_token signs you in (your own layouts)
|
| 348 |
+
r = c.predict(handle_file("layout.gds"), "1/0", "1/10", 0, "Si (cryogenic, training)", 11.45, 350, "{FS.name(reg, h['default_arm'])}",
|
| 349 |
api_name="/predict_gds")
|
| 350 |
r["names"], r["matrix_fF"]
|
| 351 |
```
|
| 352 |
|
| 353 |
+
The arguments are, in order, the GDS file, the ground layers, the conductor layers, the crop margin in gaps (0 keeps the training geometry), the substrate, eps_r (for a custom substrate), the thickness in µm, the model ({', '.join(names(reg))}, in any case) and the ground pieces ("" picks them automatically when ground and conductors share a layer).
|
| 354 |
|
| 355 |
{PRIVACY_MD}
|
| 356 |
|
|
|
|
| 510 |
if on:
|
| 511 |
return ""
|
| 512 |
return UI.callout("warn", "Off the training stack (Si, eps_r 11.45, 350 µm)", [
|
| 513 |
+
f"QCap and QCap-Lite solve the physics prior at your stack, and the networks correct it unchanged. We tested this on "
|
| 514 |
+
f"{A.TESTED_RANGE}. QCap-Fast only rescales its prediction, and it tells you so."])
|
| 515 |
|
| 516 |
|
| 517 |
def _api(fn, n_out):
|
|
|
|
| 523 |
reg = load_registry()
|
| 524 |
subs = list(reg.substrates()) + [CUSTOM]
|
| 525 |
ms = served(reg)
|
| 526 |
+
default = FS.name(reg, reg["flagship_hub"]["default_arm"])
|
| 527 |
ex = sorted(Path(examples_dir).glob("*.gds")) if examples_dir and Path(examples_dir).exists() else []
|
| 528 |
AC.register_examples(ex)
|
| 529 |
usage_stats()
|
| 530 |
+
label = {"accurate": "QCap (≈1 s, default)", "balanced": "QCap-Lite (≈0.5 s)", "fast": "QCap-Fast (≈0.1 s, trained families only)"}
|
| 531 |
with UI.blocks(title, delete_cache=(60, PURGE_AGE_S)) as demo:
|
| 532 |
gr.HTML(UI.header(reg["flagship_hub"]["title"], LEDE), container=False, padding=False, elem_classes="tl-flush")
|
| 533 |
with gr.Row():
|
|
|
|
| 567 |
gparts = GU.build()
|
| 568 |
gst = gparts[0]
|
| 569 |
with gr.Row(equal_height=False):
|
| 570 |
+
model = gr.Radio([(label.get(k, FS.name(reg, k)), FS.name(reg, k)) for k in ms], value=default, label="Model", interactive=True, elem_id="tl-tier",
|
| 571 |
scale=3, min_width=300)
|
| 572 |
substrate = gr.Dropdown(subs, value=subs[0], label="Substrate", scale=2, min_width=220)
|
| 573 |
custom_eps = gr.Number(value=11.45, label="eps_r", visible=False, scale=1, min_width=110)
|
|
|
|
| 576 |
with gr.Accordion("Models and their accuracy", open=False):
|
| 577 |
gr.Markdown(tiers_md(reg))
|
| 578 |
if "fast" in ms:
|
| 579 |
+
gr.Markdown("Use QCap-Fast only for quick screening of the component families it was trained on. On new designs it did "
|
| 580 |
+
"worse than the physics alone (see the table above), so use QCap for anything that matters.")
|
| 581 |
pred_btn = gr.Button("Predict", variant="primary")
|
| 582 |
status = gr.HTML(elem_classes="tl-flush")
|
| 583 |
with gr.Column(visible=False) as result_box, gr.Row():
|
|
|
|
| 600 |
a_c = gr.Textbox("1/10")
|
| 601 |
a_m = gr.Number(0)
|
| 602 |
a_gp = gr.Textbox("")
|
| 603 |
+
a_model = gr.Textbox(default) # API model argument: any case of QCap / QCap-Lite / QCap-Fast (a Radio takes exact choices only)
|
| 604 |
a_out = gr.JSON()
|
| 605 |
a_btn = gr.Button("api")
|
| 606 |
m_btn = gr.Button("models")
|
|
|
|
| 615 |
prepared_gds = gr.File()
|
| 616 |
pins = [gds, layers, crop, margin, margin_um, select, substrate, custom_eps, thickness, model, gst]
|
| 617 |
pouts = [mat, csv_f, json_f, report, res_state]
|
| 618 |
+
api_pins = [a_model if x is model else x for x in pins]
|
| 619 |
vin = [view, vst, gds, layers, crop, margin, margin_um, select, gst]
|
| 620 |
vout = [img, svg_btn, view_msg, vst]
|
| 621 |
+
ui_predict = functools.wraps(predict_fn)(UI.predict_ui(predict_fn, lambda r: expected(reg, FS.key(reg, r["tier"])), note=True))
|
| 622 |
ui_predict.__name__ = "predict_fn_ui"
|
| 623 |
uouts = pouts + [status, result_box, mat_note]
|
| 624 |
light = dict(concurrency_id="light", concurrency_limit=1) # previews and file handling: never queued behind a solve
|
|
|
|
| 641 |
for ev in (crop.change, margin.change, margin_um.change, select.change):
|
| 642 |
ev(preview_fn, vin, [view] + vout, trigger_mode="always_last", **hidden)
|
| 643 |
pred_btn.click(ui_predict, pins, uouts, api_visibility="private", **solve)
|
| 644 |
+
p_btn.click(_api(predict_fn, 5), api_pins, pouts, api_name="predict", **solve)
|
| 645 |
UI.mark_stale(res_state, status, mat, report, [gds.change, model.change, substrate.change, custom_eps.change, thickness.change,
|
| 646 |
select.change, crop.change, margin.change, margin_um.change, edit.then])
|
| 647 |
+
a_btn.click(_api(api_predict_gds, 1), [gds, a_g, a_c, a_m, substrate, custom_eps, thickness, a_model, a_gp], a_out, api_name="predict_gds", **solve)
|
| 648 |
m_btn.click(models_fn, None, m_out, api_name="models", **light)
|
| 649 |
return demo
|
geomgnn/flagship.py
CHANGED
|
@@ -1,5 +1,9 @@
|
|
| 1 |
"""Flagship Space tiers (deploy/registry.yaml: flagship_hub, flagship_tiers; deploy/GNN_SPACE.md "Flagship Space").
|
| 2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
The tiers run through the paper Space's code path (geomgnn.paper: scene -> embedding -> priors under the tier's prior stack, or the
|
| 4 |
learned prior surrogate -> networks), unchanged. `view(reg)` is the registry seen with paper_arms / paper_hub replaced by
|
| 5 |
flagship_tiers / flagship_hub, so every geomgnn.paper function serves the tiers as arms.
|
|
@@ -29,13 +33,35 @@ def served(reg) -> list[str]:
|
|
| 29 |
return [k for k in PA.live(v) if PA.available(k, v)]
|
| 30 |
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
def predict(prepared, tier, reg, **kw) -> dict:
|
| 33 |
-
"""Full N x N Maxwell matrix (fF; ground row/column 0) of a prepared layout with a flagship tier (geomgnn.paper.predict).
|
|
|
|
| 34 |
v = view(reg)
|
|
|
|
| 35 |
res = PA.predict(prepared, tier, v, **kw)
|
| 36 |
t = v["paper_arms"][tier]
|
| 37 |
res.pop("paper", None)
|
| 38 |
-
|
|
|
|
|
|
|
| 39 |
return res
|
| 40 |
|
| 41 |
|
|
|
|
| 1 |
"""Flagship Space tiers (deploy/registry.yaml: flagship_hub, flagship_tiers; deploy/GNN_SPACE.md "Flagship Space").
|
| 2 |
|
| 3 |
+
The tiers are the models QCap (default), QCap-Lite and QCap-Fast (registry paper_name; handoff/squadds_integration/RENAME_QCAP.md).
|
| 4 |
+
Their registry keys (accurate, balanced, fast) stay internal: model folders, checkpoints and the paper code path use them; every
|
| 5 |
+
result, label and API argument uses the model names (case-insensitive, "-" and "_" interchangeable).
|
| 6 |
+
|
| 7 |
The tiers run through the paper Space's code path (geomgnn.paper: scene -> embedding -> priors under the tier's prior stack, or the
|
| 8 |
learned prior surrogate -> networks), unchanged. `view(reg)` is the registry seen with paper_arms / paper_hub replaced by
|
| 9 |
flagship_tiers / flagship_hub, so every geomgnn.paper function serves the tiers as arms.
|
|
|
|
| 33 |
return [k for k in PA.live(v) if PA.available(k, v)]
|
| 34 |
|
| 35 |
|
| 36 |
+
def name(reg, tier) -> str:
|
| 37 |
+
"""Model name (QCap, QCap-Lite, QCap-Fast) of a tier key."""
|
| 38 |
+
return tiers(reg)[tier]["paper_name"]
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
def _norm(s) -> str:
|
| 42 |
+
return str(s).strip().lower().replace("_", "-")
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
def key(reg, model) -> str:
|
| 46 |
+
"""Tier key of a model name (case-insensitive, "-" and "_" interchangeable); ValueError naming the models otherwise."""
|
| 47 |
+
ts = tiers(reg)
|
| 48 |
+
k = next((k for k in ts if _norm(ts[k]["paper_name"]) == _norm(model)), None)
|
| 49 |
+
if k is None:
|
| 50 |
+
raise ValueError(f"Unknown model {model!r}. Choose one of {', '.join(ts[k]['paper_name'] for k in served(reg))}.")
|
| 51 |
+
return k
|
| 52 |
+
|
| 53 |
+
|
| 54 |
def predict(prepared, tier, reg, **kw) -> dict:
|
| 55 |
+
"""Full N x N Maxwell matrix (fF; ground row/column 0) of a prepared layout with a flagship tier (geomgnn.paper.predict).
|
| 56 |
+
tier: a tier key or a model name; the result names the model (model, arm, tier = QCap / QCap-Lite / QCap-Fast)."""
|
| 57 |
v = view(reg)
|
| 58 |
+
tier = tier if tier in v["paper_arms"] else key(reg, tier)
|
| 59 |
res = PA.predict(prepared, tier, v, **kw)
|
| 60 |
t = v["paper_arms"][tier]
|
| 61 |
res.pop("paper", None)
|
| 62 |
+
n = t["paper_name"]
|
| 63 |
+
res.update(model=n, arm=n, tier=n, tier_title=t["title"], reference=t.get("reference") or {}, own=t.get("own") or {}, b5=t.get("b5"),
|
| 64 |
+
b5_physics=t.get("b5_physics"))
|
| 65 |
return res
|
| 66 |
|
| 67 |
|
geomgnn/ui.py
CHANGED
|
@@ -246,7 +246,7 @@ def _offstack(res) -> str:
|
|
| 246 |
if sub.get("route") == A.FAST_ROUTE:
|
| 247 |
return (f"Substrate eps_r {e:g}, {t:g} µm. This model does not solve the physics prior at your stack. It rescales its "
|
| 248 |
f"training-stack prediction by (1 + eps_r) / (1 + 11.45) = {sub['eps_rescale_factor']:.4f} and does not model thickness. "
|
| 249 |
-
"Off the training stack we measured a median error of 6.4 % (MAE 1.7 fF, RMSE 2.4 fF), so use
|
| 250 |
head = (f"Substrate eps_r {e:g}, {t:g} µm. The physics prior was solved at this stack, and the networks (trained on Si, eps_r "
|
| 251 |
"11.45, 350 µm) are unchanged.")
|
| 252 |
if not sub.get("tested", True):
|
|
|
|
| 246 |
if sub.get("route") == A.FAST_ROUTE:
|
| 247 |
return (f"Substrate eps_r {e:g}, {t:g} µm. This model does not solve the physics prior at your stack. It rescales its "
|
| 248 |
f"training-stack prediction by (1 + eps_r) / (1 + 11.45) = {sub['eps_rescale_factor']:.4f} and does not model thickness. "
|
| 249 |
+
"Off the training stack we measured a median error of 6.4 % (MAE 1.7 fF, RMSE 2.4 fF), so use QCap for other substrates.")
|
| 250 |
head = (f"Substrate eps_r {e:g}, {t:g} µm. The physics prior was solved at this stack, and the networks (trained on Si, eps_r "
|
| 251 |
"11.45, 350 µm) are unchanged.")
|
| 252 |
if not sub.get("tested", True):
|
models/accurate/info.json
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"name": "accurate",
|
|
|
|
| 3 |
"kind": "gnn",
|
| 4 |
"checkpoints": [
|
| 5 |
"acc_v2m_sf_s0.pt",
|
|
|
|
| 1 |
{
|
| 2 |
"name": "accurate",
|
| 3 |
+
"model": "QCap",
|
| 4 |
"kind": "gnn",
|
| 5 |
"checkpoints": [
|
| 6 |
"acc_v2m_sf_s0.pt",
|
models/balanced/info.json
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"name": "balanced",
|
|
|
|
| 3 |
"kind": "gnn",
|
| 4 |
"checkpoints": [
|
| 5 |
"bal_v2m_sf_s0.pt",
|
|
|
|
| 1 |
{
|
| 2 |
"name": "balanced",
|
| 3 |
+
"model": "QCap-Lite",
|
| 4 |
"kind": "gnn",
|
| 5 |
"checkpoints": [
|
| 6 |
"bal_v2m_sf_s0.pt",
|
models/fast/info.json
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"name": "fast",
|
|
|
|
| 3 |
"kind": "gnn",
|
| 4 |
"checkpoints": [
|
| 5 |
"acc_v1_s0.pt",
|
|
|
|
| 1 |
{
|
| 2 |
"name": "fast",
|
| 3 |
+
"model": "QCap-Fast",
|
| 4 |
"kind": "gnn",
|
| 5 |
"checkpoints": [
|
| 6 |
"acc_v1_s0.pt",
|
registry.yaml
CHANGED
|
@@ -1393,6 +1393,15 @@ flagship_hub:
|
|
| 1393 |
deploy_20261114: # public-demo limits + short card (no paper model); Docker SYSTEM=spaces so HF OAuth works
|
| 1394 |
commit: 142609ed7021861aea2be8ee5f51594b70e5918b # + text in the author's voice, Fast in the selector, Model not Tier, footer (was 4d44ef6)
|
| 1395 |
live: deploy/tests/results/flagship_space_live_limits20261114.json # examples x tiers vs Mac <= 1.1e-3 rel; anon non-example refused; oauth_token call ran
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1396 |
space: shanto268/PlanarCapacitanceModel # the URL is kept; its earlier v3 tiers are archived (archive below)
|
| 1397 |
title: Planar Capacitance Model
|
| 1398 |
private: true # Spaces stay private (CLAUDE.md rule 11)
|
|
@@ -1408,6 +1417,16 @@ flagship_hub:
|
|
| 1408 |
SQuADDS/SQuADDS_PlanarCapacitance-heldout: c538385b3e8d48f4183b68d00e376a06762b0705
|
| 1409 |
SQuADDS/PlanarCapacitance-flagship: a89f30850b638da1f693042b7d7cf14f20b19098 # + balanced/bal_v2m_sf_s{0,1,2}.pt (same sha256 as the Space) and SHA256SUMS
|
| 1410 |
not_pushed: [SQuADDS/PlanarCapacitance-paper, shanto268/PlanarCapacitanceModel] # drafts in handoff/hf_cards/; author 2026-11-14: these stay private
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1411 |
training_layouts: 45771
|
| 1412 |
archive:
|
| 1413 |
space: shanto268/PlanarCapacitanceModel
|
|
@@ -1431,9 +1450,9 @@ flagship_hub:
|
|
| 1431 |
|
| 1432 |
flagship_tiers:
|
| 1433 |
accurate:
|
| 1434 |
-
title: "
|
| 1435 |
-
paper_name: "
|
| 1436 |
-
route: "
|
| 1437 |
kind: gnn
|
| 1438 |
arm: full_gnn
|
| 1439 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
@@ -1457,9 +1476,9 @@ flagship_tiers:
|
|
| 1457 |
time_space: "interdigital 0.47 s, xmon 0.91 s, 1-qubit lattice (N 15) 19.0 s, 16-conductor layout 9.9 s, 9-qubit lattice (N 67) 154 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1458 |
time_x86: "0.845 / 5.96 s in family, 1.75 / 18.1 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.99 s median (1.14 s on the 2026-10-20 re-run, Xeon E5-2640 v4)"
|
| 1459 |
balanced:
|
| 1460 |
-
title: "
|
| 1461 |
-
paper_name: "
|
| 1462 |
-
route: "
|
| 1463 |
kind: gnn
|
| 1464 |
arm: full_gnn
|
| 1465 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
@@ -1469,7 +1488,7 @@ flagship_tiers:
|
|
| 1469 |
- {name: bal_v2m_sf_s0, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s0/model.pt, sha256: aa742ab93b1e3ffcd86a753bda5cc4a68812acbe8140a6b0e311c11d1f36b4b2} # was bal_v2m_s0 50166723 (half-space far field)
|
| 1470 |
- {name: bal_v2m_sf_s1, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s1/model.pt, sha256: 486466473eba61c090f4e3fa7988bf880fc183570c5d02bd78f4c1356691f926} # was bal_v2m_s1 5d1f0ffb (half-space far field)
|
| 1471 |
- {name: bal_v2m_sf_s2, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s2/model.pt, sha256: 1b55ded04fd624585eeeccf2ab6e4a8e9354a9b96694131e1737dde43ad3ff58} # was bal_v2m_s2 dcd14c61 (half-space far field)
|
| 1472 |
-
summary: "The
|
| 1473 |
own: # data_local/farfield_adopt/insample.json, bal_v2m_sf ensemble
|
| 1474 |
insample: {set: "in-sample, all 45,771 training layouts (TRAIN + VAL + the paper's test split; it was trained on them)", median_pct: 0.300, mae_ff: 0.0872, rmse_ff: 0.0999}
|
| 1475 |
reference:
|
|
@@ -1480,9 +1499,9 @@ flagship_tiers:
|
|
| 1480 |
time_space: "interdigital 0.29 s, xmon 0.51 s, 1-qubit lattice (N 15) 7.7 s, 16-conductor layout 2.4 s, 9-qubit lattice (N 67) 54 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1481 |
time_x86: "0.490 / 2.31 s in family, 1.00 / 10.0 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.56 s median"
|
| 1482 |
fast:
|
| 1483 |
-
title: "Fast (~0.1 s): a learned prior replaces the solves (trained component families only)"
|
| 1484 |
-
paper_name: "Fast"
|
| 1485 |
-
route: "Fast"
|
| 1486 |
kind: gnn
|
| 1487 |
arm: full_gnn
|
| 1488 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
@@ -1495,7 +1514,7 @@ flagship_tiers:
|
|
| 1495 |
name: sall_all_s0 # surrogate seed 0 by declaration (PROTOCOL.md section 3, amendment A1)
|
| 1496 |
checkpoint: {name: sall_all_s0, path: data_local/flagship_v4/hpc/sur/sall_all_s0/model.pt, sha256: 132d6f389eda798d31bd5654a96e24cce18fa750134a9fb2a9725eafed339dac}
|
| 1497 |
summary: "A learned prior (1.02M parameters) predicts the two physics solves from the geometry, and the networks correct it. There is no field solve, so the time is the geometric embedding. The learned prior was trained on the physics prior of every Q3D design plus 432,697 generated layouts solved by the prior alone, and the networks on all 45,771 of our Q3D designs."
|
| 1498 |
-
warning: "Use Fast only for quick screening of the component families it was trained on (interdigital, paddle, concentric, star, multipad star, Xmon, TransmonPocket, shielded concentric and the generated layouts). On new designs of a trained family it was worse than the physics alone, 3.175 % (MAE 0.583 / RMSE 0.824 fF) against 1.765 % (0.275 / 0.376 fF). In the paper it was also worse than the physics alone on a component family held out of its training, 3.11 % (0.533 / 0.633 fF) against 1.55 % (0.166 / 0.205 fF). Use
|
| 1499 |
own: # data_local/flagship_v4/scores/tables.json pooled.fast_sall_all_s0_ens (agents/flagship_v4/RESULTS.md section 2)
|
| 1500 |
insample: {set: "in-sample, all 45,771 training layouts (TRAIN + VAL + the paper's test split; it was trained on them)", median_pct: 0.671, mae_ff: 0.353, rmse_ff: 0.395}
|
| 1501 |
reference:
|
|
@@ -1507,16 +1526,16 @@ flagship_tiers:
|
|
| 1507 |
heldout_family_physics: {median_pct: 1.554, mae_ff: 0.166, rmse_ff: 0.205}
|
| 1508 |
val_same_surrogate: {median_pct: 0.722, mae_ff: 0.387, rmse_ff: 0.436, note: "this release's surrogate recipe trained without the validation layouts (P6 SUR-ALL) + the paper's TRAIN-only networks, held out (agents/surcov_p6/RESULTS_P6.md)"}
|
| 1509 |
b5: {median_pct: 3.175, mae_ff: 0.583, rmse_ff: 0.824}
|
| 1510 |
-
b5_physics: {name: "physics alone C^R = R(1.5, 3) (worse than the physics: Fast is behind it here)", median_pct: 1.765, mae_ff: 0.275, rmse_ff: 0.376}
|
| 1511 |
time_space: "interdigital 0.036 s, xmon 0.11 s, 1-qubit lattice (N 15) 0.82 s, 16-conductor layout 0.49 s, 9-qubit lattice (N 67) 3.9 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1512 |
time_x86: "0.091 / 0.274 s in family, 0.30 / 2.45 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.24 s median"
|
| 1513 |
|
| 1514 |
squadds_ml_api: # unified SQuADDS ML inference API (handoff/squadds_ml_api/DESIGN.md, source deploy/squadds_ml_api/, build.py)
|
| 1515 |
space: SQuADDS/squadds-ml-inference-api # free cpu-basic (created 2026-04-10, before HF's org paywall; a rebuild stays free, checked 2026-11-14)
|
| 1516 |
-
deployed: 2026-11-
|
| 1517 |
-
commit:
|
| 1518 |
-
previous:
|
| 1519 |
rollback: legacy-olivia # tag on 01966faaa06432d425e06895f616db40e6cb858d (Olivia's single-model API); README-only test commit 7c985614
|
| 1520 |
-
models: {transmon_cross_hamiltonian_inverse: public,
|
| 1521 |
collections: {models: SQuADDS/squadds-models-6af8b7da67a22df62af374c6, datasets: SQuADDS/squadds-datasets-6a65152b857f8efa5620880a}
|
| 1522 |
gating: gated = HF token of a SQuADDS org member; needs Space secret HF_TOKEN (org read token) before the capacitance models go gated
|
|
|
|
| 1393 |
deploy_20261114: # public-demo limits + short card (no paper model); Docker SYSTEM=spaces so HF OAuth works
|
| 1394 |
commit: 142609ed7021861aea2be8ee5f51594b70e5918b # + text in the author's voice, Fast in the selector, Model not Tier, footer (was 4d44ef6)
|
| 1395 |
live: deploy/tests/results/flagship_space_live_limits20261114.json # examples x tiers vs Mac <= 1.1e-3 rel; anon non-example refused; oauth_token call ran
|
| 1396 |
+
deploy_20261115: # speed (phases A/B), outputs unchanged: concurrency groups, 2 threads, thickness floor 1 um, pinned deps (torch 2.14.1), background preload, slim bundle checkpoints, baked example results, prior-level cache
|
| 1397 |
+
commit: f76c3bf6cfcd7c499049d7cc3d3ef0258d65a3e1
|
| 1398 |
+
rollback: {commit: 142609ed7021861aea2be8ee5f51594b70e5918b, local_copy: data_local/hf_archive/PlanarCapacitanceModel_142609ed} # 102 files
|
| 1399 |
+
prior_level_cache_mb: 256 # TLSQ_PRIOR_LEVEL_CACHE_MB, set in the Space app.py
|
| 1400 |
+
bake: "18 entries (6 examples x 3 tiers), warm.py bake step 62.9 s, peak RSS 2328 MB; loaded at start (interdigital recomputes bit for bit)"
|
| 1401 |
+
gate: deploy/tests/results/flagship_space_gate_speed_20261115.json # + demo limits 11/11, examples 9/9, crop 8/8, substrate 3/3, flagship 6/6
|
| 1402 |
+
live: deploy/tests/results/flagship_space_live_speed_20261115.json # signed in (oauth_token): 12/12, examples bitwise = 2026-11-14 live, vs Mac <= 1.1e-3 rel (8.9e-3 fF)
|
| 1403 |
+
offstack: deploy/tests/results/flagship_space_live_offstack_speed_20261115.json # bitwise = 2026-11-06 live
|
| 1404 |
+
coldstart: {before: deploy/tests/results/flagship_space_coldstart_before_speed_20261115.json, after: deploy/tests/results/flagship_space_coldstart_speed_20261115.json}
|
| 1405 |
space: shanto268/PlanarCapacitanceModel # the URL is kept; its earlier v3 tiers are archived (archive below)
|
| 1406 |
title: Planar Capacitance Model
|
| 1407 |
private: true # Spaces stay private (CLAUDE.md rule 11)
|
|
|
|
| 1417 |
SQuADDS/SQuADDS_PlanarCapacitance-heldout: c538385b3e8d48f4183b68d00e376a06762b0705
|
| 1418 |
SQuADDS/PlanarCapacitance-flagship: a89f30850b638da1f693042b7d7cf14f20b19098 # + balanced/bal_v2m_sf_s{0,1,2}.pt (same sha256 as the Space) and SHA256SUMS
|
| 1419 |
not_pushed: [SQuADDS/PlanarCapacitance-paper, shanto268/PlanarCapacitanceModel] # drafts in handoff/hf_cards/; author 2026-11-14: these stay private
|
| 1420 |
+
model_repo: # rename 2026-11-16 (handoff/squadds_integration/RENAME_QCAP.md): flagship -> QCap; tier keys below stay internal
|
| 1421 |
+
id: SQuADDS/QCap # was SQuADDS/PlanarCapacitance-flagship (old URL redirects)
|
| 1422 |
+
folders_revision: 6dd46d33cd58788222dbf5478ca3550f2888f179 # accurate/ -> qcap/, balanced/ -> qcap-lite/, fast/ -> qcap-fast/; weights and sha256 unchanged
|
| 1423 |
+
folders: {accurate: qcap/, balanced: qcap-lite/, fast: qcap-fast/}
|
| 1424 |
+
names: {accurate: QCap, balanced: QCap-Lite, fast: QCap-Fast} # = flagship_tiers.*.paper_name; API ids qcap / qcap-fast (squadds_ml_api)
|
| 1425 |
+
hf_cards_20261116: # README-only, QCap names (handoff/hf_cards/QCap.md, _links.md, _tiers.md); private repos stay private
|
| 1426 |
+
SQuADDS/QCap: 25bdbb19ea3a17b2e86b08baea7b320e2d06d3a0
|
| 1427 |
+
SQuADDS/SQuADDS_PlanarCapacitance: e85ca683991bd3368f8c40d128ae2e7861760dbd
|
| 1428 |
+
SQuADDS/SQuADDS_PlanarCapacitance-Synthetic: d0058e32c863eee28a2060e87203c58527882320
|
| 1429 |
+
SQuADDS/README: 007e7d48481eb78894d4f1a0beb2e54764325c4e # org card Space (public)
|
| 1430 |
training_layouts: 45771
|
| 1431 |
archive:
|
| 1432 |
space: shanto268/PlanarCapacitanceModel
|
|
|
|
| 1450 |
|
| 1451 |
flagship_tiers:
|
| 1452 |
accurate:
|
| 1453 |
+
title: "QCap (~1 s): the physics prior from two coarse solves, extrapolated and corrected by the networks"
|
| 1454 |
+
paper_name: "QCap" # display name = the API name (case-insensitive; handoff/squadds_integration/RENAME_QCAP.md)
|
| 1455 |
+
route: "QCap"
|
| 1456 |
kind: gnn
|
| 1457 |
arm: full_gnn
|
| 1458 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
|
|
| 1476 |
time_space: "interdigital 0.47 s, xmon 0.91 s, 1-qubit lattice (N 15) 19.0 s, 16-conductor layout 9.9 s, 9-qubit lattice (N 67) 154 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1477 |
time_x86: "0.845 / 5.96 s in family, 1.75 / 18.1 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.99 s median (1.14 s on the 2026-10-20 re-run, Xeon E5-2640 v4)"
|
| 1478 |
balanced:
|
| 1479 |
+
title: "QCap-Lite (~0.5 s): the same recipe on two coarser solves"
|
| 1480 |
+
paper_name: "QCap-Lite"
|
| 1481 |
+
route: "QCap-Lite"
|
| 1482 |
kind: gnn
|
| 1483 |
arm: full_gnn
|
| 1484 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
|
|
| 1488 |
- {name: bal_v2m_sf_s0, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s0/model.pt, sha256: aa742ab93b1e3ffcd86a753bda5cc4a68812acbe8140a6b0e311c11d1f36b4b2} # was bal_v2m_s0 50166723 (half-space far field)
|
| 1489 |
- {name: bal_v2m_sf_s1, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s1/model.pt, sha256: 486466473eba61c090f4e3fa7988bf880fc183570c5d02bd78f4c1356691f926} # was bal_v2m_s1 5d1f0ffb (half-space far field)
|
| 1490 |
- {name: bal_v2m_sf_s2, path: data_local/flagship_v4/hpc/runs/bal_v2m_sf_s2/model.pt, sha256: 1b55ded04fd624585eeeccf2ab6e4a8e9354a9b96694131e1737dde43ad3ff58} # was bal_v2m_s2 dcd14c61 (half-space far field)
|
| 1491 |
+
summary: "The QCap recipe on two coarser meshes, about half the solve time, with networks retrained for that pair on all 45,771 of our Q3D designs. About 2x faster than QCap at 1.5-1.7x its error."
|
| 1492 |
own: # data_local/farfield_adopt/insample.json, bal_v2m_sf ensemble
|
| 1493 |
insample: {set: "in-sample, all 45,771 training layouts (TRAIN + VAL + the paper's test split; it was trained on them)", median_pct: 0.300, mae_ff: 0.0872, rmse_ff: 0.0999}
|
| 1494 |
reference:
|
|
|
|
| 1499 |
time_space: "interdigital 0.29 s, xmon 0.51 s, 1-qubit lattice (N 15) 7.7 s, 16-conductor layout 2.4 s, 9-qubit lattice (N 67) 54 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1500 |
time_x86: "0.490 / 2.31 s in family, 1.00 / 10.0 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.56 s median"
|
| 1501 |
fast:
|
| 1502 |
+
title: "QCap-Fast (~0.1 s): a learned prior replaces the solves (trained component families only)"
|
| 1503 |
+
paper_name: "QCap-Fast"
|
| 1504 |
+
route: "QCap-Fast"
|
| 1505 |
kind: gnn
|
| 1506 |
arm: full_gnn
|
| 1507 |
embedding: {far_k: 64, d_far_um: 100.0} # agents/embedspeed Part A (PROTOCOL.md amendment 2): far-sample decimation F64 of the pair vectors, the fastest arm tying the exact embedding on validation 4,608 and the 1q / 5q / 9q lattices; the paper Space keeps the exact embedding
|
|
|
|
| 1514 |
name: sall_all_s0 # surrogate seed 0 by declaration (PROTOCOL.md section 3, amendment A1)
|
| 1515 |
checkpoint: {name: sall_all_s0, path: data_local/flagship_v4/hpc/sur/sall_all_s0/model.pt, sha256: 132d6f389eda798d31bd5654a96e24cce18fa750134a9fb2a9725eafed339dac}
|
| 1516 |
summary: "A learned prior (1.02M parameters) predicts the two physics solves from the geometry, and the networks correct it. There is no field solve, so the time is the geometric embedding. The learned prior was trained on the physics prior of every Q3D design plus 432,697 generated layouts solved by the prior alone, and the networks on all 45,771 of our Q3D designs."
|
| 1517 |
+
warning: "Use QCap-Fast only for quick screening of the component families it was trained on (interdigital, paddle, concentric, star, multipad star, Xmon, TransmonPocket, shielded concentric and the generated layouts). On new designs of a trained family it was worse than the physics alone, 3.175 % (MAE 0.583 / RMSE 0.824 fF) against 1.765 % (0.275 / 0.376 fF). In the paper it was also worse than the physics alone on a component family held out of its training, 3.11 % (0.533 / 0.633 fF) against 1.55 % (0.166 / 0.205 fF). Use QCap for anything that matters."
|
| 1518 |
own: # data_local/flagship_v4/scores/tables.json pooled.fast_sall_all_s0_ens (agents/flagship_v4/RESULTS.md section 2)
|
| 1519 |
insample: {set: "in-sample, all 45,771 training layouts (TRAIN + VAL + the paper's test split; it was trained on them)", median_pct: 0.671, mae_ff: 0.353, rmse_ff: 0.395}
|
| 1520 |
reference:
|
|
|
|
| 1526 |
heldout_family_physics: {median_pct: 1.554, mae_ff: 0.166, rmse_ff: 0.205}
|
| 1527 |
val_same_surrogate: {median_pct: 0.722, mae_ff: 0.387, rmse_ff: 0.436, note: "this release's surrogate recipe trained without the validation layouts (P6 SUR-ALL) + the paper's TRAIN-only networks, held out (agents/surcov_p6/RESULTS_P6.md)"}
|
| 1528 |
b5: {median_pct: 3.175, mae_ff: 0.583, rmse_ff: 0.824}
|
| 1529 |
+
b5_physics: {name: "physics alone C^R = R(1.5, 3) (worse than the physics: QCap-Fast is behind it here)", median_pct: 1.765, mae_ff: 0.275, rmse_ff: 0.376}
|
| 1530 |
time_space: "interdigital 0.036 s, xmon 0.11 s, 1-qubit lattice (N 15) 0.82 s, 16-conductor layout 0.49 s, 9-qubit lattice (N 67) 3.9 s (server time, one request, Docker build 2026-10-20; a repeated request is answered from the cache in ~0.01 s)"
|
| 1531 |
time_x86: "0.091 / 0.274 s in family, 0.30 / 2.45 s zero-shot (median / p90; x86 cluster, 2 CPUs); B5 0.24 s median"
|
| 1532 |
|
| 1533 |
squadds_ml_api: # unified SQuADDS ML inference API (handoff/squadds_ml_api/DESIGN.md, source deploy/squadds_ml_api/, build.py)
|
| 1534 |
space: SQuADDS/squadds-ml-inference-api # free cpu-basic (created 2026-04-10, before HF's org paywall; a rebuild stays free, checked 2026-11-14)
|
| 1535 |
+
deployed: 2026-11-15
|
| 1536 |
+
commit: 8a0986e534f64165d2bbfe849412b05814f2f358 # squadds ec4414d (lean/fast prior, bitwise), thread env 2 + torch threads, check_thickness; warm.py uses physics_prior (0dc991fe failed to build on the old name); 56/56 local, 11/11 legacy oracle live, capacitance models 404
|
| 1537 |
+
previous: a4f1ca1bfd6f0087baa0ab493d3ae04438671321 # 2026-11-14 deploy (a97aa742 + later commits), squadds 04ebd59
|
| 1538 |
rollback: legacy-olivia # tag on 01966faaa06432d425e06895f616db40e6cb858d (Olivia's single-model API); README-only test commit 7c985614
|
| 1539 |
+
models: {transmon_cross_hamiltonian_inverse: public, qcap: hidden, qcap-fast: hidden} # ids renamed 2026-11-16 (were planar_capacitance_flagship / _fast)
|
| 1540 |
collections: {models: SQuADDS/squadds-models-6af8b7da67a22df62af374c6, datasets: SQuADDS/squadds-datasets-6a65152b857f8efa5620880a}
|
| 1541 |
gating: gated = HF token of a SQuADDS org member; needs Space secret HF_TOKEN (org read token) before the capacitance models go gated
|