shanto268 commited on
Commit
ff01de7
·
verified ·
1 Parent(s): 9166863

Add PlanarCapacitance-flagship checkpoints, sha256 list and model card

Browse files
README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ library_name: pytorch
4
+ tags: [superconducting-qubits, capacitance, electrostatics, gds, graph-neural-network, squadds]
5
+ ---
6
+
7
+ # Planar capacitance: flagship models (SQuADDS)
8
+
9
+ GDS layout -> full Maxwell capacitance matrix of superconducting-circuit components. Networks trained on all 45,771 on-stack Q3D layouts (no held-out score of their own: the reference rows are the same recipe trained on the TRAIN split; B5 is one fresh check of 39 new layouts of one generated family, 3-4 conductors). For benchmarks use the paper model (`SQuADDS/PlanarCapacitance-paper`).
10
+
11
+ ```python
12
+ from squadds.ml.capacitance import CapacitancePredictor, load_model
13
+ res = CapacitancePredictor(load_model("flagship", tier="accurate")).predict("chip.gds")
14
+ res.names, res.matrix_fF, res.to_sim_results()
15
+ ```
16
+
17
+ | tier | folder | what |
18
+ |---|---|---|
19
+ | **Accurate** (default) | `accurate/` | Two thin-film method-of-moments solves (corrected prior with substrate far field, 1.5 and 3 panels per gap), Richardson-extrapolated, corrected by a 3-seed ensemble of the paper model's networks (89,858 parameters per seed, bounded residual) with a learned row-sum closure: the complete Maxwell matrix, self terms included. Networks retrained on all 45,771 on-stack Q3D layouts (2026-10-20, on the corrected prior with substrate far field). |
20
+ | Fast | `fast/` | A learned surrogate (1.02M parameters, trained on the prior of every on-stack layout plus 432,697 generated layouts labelled by the prior alone) predicts the two uncorrected-prior solves at 1.5 and 3 panels per gap; networks retrained on all 45,771 on-stack layouts on that prior correct them. No field solve: the time is the geometric embedding. |
21
+
22
+ > **Warning (Fast).** Use Fast only for quick screening of the component families it was trained on (interdigital, paddle, concentric, star, multipad star, Xmon, TransmonPocket, shielded concentric, two-net, random, procedurally generated and synthetic campaign layouts). On new designs of a trained family (B5) it was worse than the physics alone, 3.175 % (MAE 0.583 / RMSE 0.824 fF) against 1.765 % (0.275 / 0.376 fF), and in the paper it was worse than the physics alone on a held-out family too, 3.11 % (0.533 / 0.633 fF) against 1.55 % (0.166 / 0.205 fF). Use Accurate for anything that matters.
23
+
24
+ ## Accurate
25
+
26
+ Reference recipe: the paper's model (same networks and prior stack at inference; its networks were trained on 35,354 TRAIN layouts on the uncorrected prior; retrained on the corrected prior, as here, they score validation 0.218 % (0.0774 / 0.0888 fF)).
27
+
28
+ | reference | median % (MAE / RMSE) |
29
+ |---|---|
30
+ | paper recipe, validation (4,608 layouts) | 0.229 % (MAE 0.0674 fF / RMSE 0.0792 fF) |
31
+ | paper recipe, test in family (4,009) | 0.177 % (MAE 0.0203 fF / RMSE 0.025 fF) |
32
+ | paper recipe, test zero-shot macro (1,878) | 0.587 % (MAE 1.092 fF / RMSE 1.152 fF) |
33
+ | paper recipe, held-out family (4,071) | 0.408 % (MAE 0.049 fF / RMSE 0.059 fF) |
34
+ | B5 fresh check (39 layouts) | 1.05 % (MAE 0.132 fF / RMSE 0.188 fF) |
35
+ | B5: physics alone C^R = R(1.5, 3) | 1.765 % (MAE 0.275 fF / RMSE 0.376 fF) |
36
+
37
+ Time: x86 cluster node, 2 CPUs: 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). HF Space (2 vCPU): 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). SQuADDS on a Mac: N<=4: 1.37 s (29 layouts); 5<=N<=9: 4.73 s (3 layouts); 10<=N<=20: 13.6 s (3 layouts); N>20: 12.5 s (1 layouts) (median wall time per layout, cold caches, 2 threads, Apple M5 Pro, 18 CPUs, Darwin arm64).
38
+
39
+ ## Fast
40
+
41
+ Reference recipe: the paper's fast model (learned prior surrogate + the networks on its prior, trained on 35,354 TRAIN layouts).
42
+
43
+ | reference | median % (MAE / RMSE) |
44
+ |---|---|
45
+ | paper recipe, validation (4,608 layouts) | 0.846 % (MAE 0.491 fF / RMSE 0.549 fF) |
46
+ | paper recipe, test in family (4,009) | 0.63 % (MAE 0.113 fF / RMSE 0.1414 fF) |
47
+ | paper recipe, test zero-shot macro (1,878) | 3.576 % (MAE 4.608 fF / RMSE 5.058 fF) |
48
+ | paper recipe, held-out family (4,071) | 3.113 % (MAE 0.533 fF / RMSE 0.633 fF) |
49
+ | held-out family: physics alone | 1.554 % (MAE 0.166 fF / RMSE 0.205 fF) |
50
+ | this surrogate recipe, validation | 0.722 % (MAE 0.387 fF / RMSE 0.436 fF) |
51
+ | B5 fresh check (39 layouts) | 3.175 % (MAE 0.583 fF / RMSE 0.824 fF) |
52
+ | B5: physics alone C^R = R(1.5, 3) (worse than the physics: Fast is behind it here) | 1.765 % (MAE 0.275 fF / RMSE 0.376 fF) |
53
+
54
+ Time: x86 cluster node, 2 CPUs: 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. HF Space (2 vCPU): 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). SQuADDS on a Mac: N<=4: 0.0551 s (29 layouts); 5<=N<=9: 0.133 s (3 layouts); 10<=N<=20: 0.141 s (3 layouts); N>20: 0.724 s (1 layouts) (median wall time per layout, cold caches, 2 threads, Apple M5 Pro, 18 CPUs, Darwin arm64).
55
+
56
+ ## Files
57
+
58
+ | file | sha256 |
59
+ |---|---|
60
+ | `accurate/acc_v2m_sf_s0.pt` | `cc69c5818f5028c26177159b88239116520abbb290743d6a6c091657e8dbf1c5` |
61
+ | `accurate/acc_v2m_sf_s1.pt` | `2f88606dbec5c599550d088213a4c37af3b3ad5134ebad2e60519fb534553a57` |
62
+ | `accurate/acc_v2m_sf_s2.pt` | `d13bafc6b2eb7cadcd97a387d791c0210976db1a9d618f9acc1b614e3701daed` |
63
+ | `fast/acc_v1_s0.pt` | `9dff20ade26619e7713c2d145dada65e4554a623780290673cfe9452cf9b64ba` |
64
+ | `fast/acc_v1_s1.pt` | `f1ad2ca065dd650ccabcad73a3c6ad2ac1042f4f68bb828893bbdecc47411ffe` |
65
+ | `fast/acc_v1_s2.pt` | `8cbada211724df5cc05a7b4fdd232607de047b248a43993d3844bd68ff8060c4` |
66
+ | `fast/sall_all_s0.pt` | `132d6f389eda798d31bd5654a96e24cce18fa750134a9fb2a9725eafed339dac` |
67
+
68
+ Units and conventions: fF; the full N x N Maxwell matrix of every conductor the layout has (nets are never merged), ground
69
+ first (row/column 0), diagonal = self, off-diagonal = -mutual. Served for Si (eps_r 11.45), 350 µm substrates, zero-thickness
70
+ metal; ground on GDS layer 1/0, conductors on 1/10 (configurable). Other substrates raise unless
71
+ `substrate_rescale="quasi_static"` is passed (flagged in the result, unvalidated).
72
+
73
+ Metric: layout-mean median relative error over mutual entries against Q3D (0.1 fF floor), with MAE / RMSE in fF.
74
+
75
+ Integrity: SQuADDS pins a revision of this repository and checks the sha256 of every file against the manifest shipped in the
76
+ package (`squadds/ml/capacitance/manifest.json`); see `SHA256SUMS`.
77
+
78
+ Prior stack (Accurate): `v2m_sf_fc_dipfill_margin_slabfar` (corrected prior with substrate far field, compiled exact solver). Physics priors computed on macOS differ from the HPC x86 ones by up to ~2.4e-3 relative.
SHA256SUMS ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ cc69c5818f5028c26177159b88239116520abbb290743d6a6c091657e8dbf1c5 accurate/acc_v2m_sf_s0.pt
2
+ 2f88606dbec5c599550d088213a4c37af3b3ad5134ebad2e60519fb534553a57 accurate/acc_v2m_sf_s1.pt
3
+ d13bafc6b2eb7cadcd97a387d791c0210976db1a9d618f9acc1b614e3701daed accurate/acc_v2m_sf_s2.pt
4
+ 9dff20ade26619e7713c2d145dada65e4554a623780290673cfe9452cf9b64ba fast/acc_v1_s0.pt
5
+ f1ad2ca065dd650ccabcad73a3c6ad2ac1042f4f68bb828893bbdecc47411ffe fast/acc_v1_s1.pt
6
+ 8cbada211724df5cc05a7b4fdd232607de047b248a43993d3844bd68ff8060c4 fast/acc_v1_s2.pt
7
+ 132d6f389eda798d31bd5654a96e24cce18fa750134a9fb2a9725eafed339dac fast/sall_all_s0.pt
accurate/acc_v2m_sf_s0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc69c5818f5028c26177159b88239116520abbb290743d6a6c091657e8dbf1c5
3
+ size 8410606
accurate/acc_v2m_sf_s1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f88606dbec5c599550d088213a4c37af3b3ad5134ebad2e60519fb534553a57
3
+ size 8410606
accurate/acc_v2m_sf_s2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d13bafc6b2eb7cadcd97a387d791c0210976db1a9d618f9acc1b614e3701daed
3
+ size 8410606
fast/acc_v1_s0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9dff20ade26619e7713c2d145dada65e4554a623780290673cfe9452cf9b64ba
3
+ size 8410094
fast/acc_v1_s1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1ad2ca065dd650ccabcad73a3c6ad2ac1042f4f68bb828893bbdecc47411ffe
3
+ size 8410094
fast/acc_v1_s2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cbada211724df5cc05a7b4fdd232607de047b248a43993d3844bd68ff8060c4
3
+ size 8410094
fast/sall_all_s0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:132d6f389eda798d31bd5654a96e24cce18fa750134a9fb2a9725eafed339dac
3
+ size 4172218