Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .github/workflows/test.yml
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@ jobs:
- uses: actions/setup-python@v5
with:
python-version: "3.12"
- uses: astral-sh/setup-uv@v5
- run: pip install pytest pytest-cov numpy scipy matplotlib
- run: make test

Expand Down
18 changes: 13 additions & 5 deletions scripts/julia-daemon.sh
Original file line number Diff line number Diff line change
Expand Up @@ -82,13 +82,20 @@ runner_path() {
runner=$(cd "$(dirname "$0")" && pwd)/$(basename "$0")
}

install_shim() {
validate_shim_path() {
runner_path
shim_path="${1:-$HOME/.local/bin/julia}"
case "$shim_path" in /*) ;; *) shim_path="$PWD/$shim_path";; esac
if [ -e "$shim_path" ] && [ "$shim_path" -ef "$runner" ]; then
validated_shim_path="${1:-$HOME/.local/bin/julia}"
case "$validated_shim_path" in
/*) ;;
*) validated_shim_path="$PWD/$validated_shim_path";;
esac
if [ -e "$validated_shim_path" ] && [ "$validated_shim_path" -ef "$runner" ]; then
fail "refusing to replace the daemon runner itself"
fi
}

install_shim() {
shim_path="$1"
resolve_real_julia
if [ -e "$shim_path" ]; then
first_line=$(IFS= read -r line <"$shim_path" && printf '%s' "$line")
Expand Down Expand Up @@ -388,8 +395,9 @@ case "$cmd" in
;;
install-shim)
[ $# -le 1 ] || fail "install-shim accepts at most one path"
validate_shim_path "${1:-}"
install_daemonmode
install_shim "${1:-}"
install_shim "$validated_shim_path"
;;
alias)
[ $# -eq 0 ] || fail "alias takes no arguments"
Expand Down
68 changes: 68 additions & 0 deletions tracks/qmc/solutions/GeGeWu-121/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,68 @@
# 格格巫 — Challenge 121 final submission

## Team

| | |
|---|---|
| **Team name** | 格格巫 |
| **Members** | Bei Qiao (乔北), Zongyue Liu (刘宗岳), Kexiang Mao (毛柯翔) |
| **Contribution policy** | Joint team result; contributions are not partitioned by member. |

## Challenge

| Row | |
|---|---|
| **Challenge** | Construct physically interacting fermion models whose determinantal-QMC weights are provably nonnegative at arbitrary auxiliary-field depth, and delimit failed extensions with exact certificates. |
| **Catalog issue** | Addresses [#121 — Sign-problem-free hunter](https://github.com/QuantumBFS/quantum.harness/issues/121), released by Lei Wang. |
| **Track** | `qmc`, following the challenge issue. |

## Final reviewer entry point

Challenge 121 is maintained in a dedicated public repository, separate from
Challenge 15:

- [English final report at immutable commit `3a20980`](https://github.com/Joe-Nor/gegewu-challenge-121-sign-free-hunter/blob/3a20980b352afb95cd05a3f5f321ca4585d81abc/FINAL_SUBMISSION_REPORT.md)
- [Static HTML report](https://github.com/Joe-Nor/gegewu-challenge-121-sign-free-hunter/blob/3a20980b352afb95cd05a3f5f321ca4585d81abc/FINAL_SUBMISSION_REPORT.html)
- [Clean-checkout reproduction guide](https://github.com/Joe-Nor/gegewu-challenge-121-sign-free-hunter/blob/3a20980b352afb95cd05a3f5f321ca4585d81abc/REPRODUCIBILITY.md)
- [Machine-readable final manifest](https://github.com/Joe-Nor/gegewu-challenge-121-sign-free-hunter/blob/3a20980b352afb95cd05a3f5f321ca4585d81abc/experiments/FINAL_SUBMISSION_V1.json)
- [Source repository](https://github.com/Joe-Nor/gegewu-challenge-121-sign-free-hunter)

The report derives the interacting Hamiltonian, Trotter/HS construction,
determinant weight, and configuration-wise sign theorem for three primary
models: the 2D totally-nonnegative tensor lift, exterior-representation lift,
and conjugate-tensor lift. It also records DQMC/exact-Trotter/ED checks of
particle number and energy at several chemical potentials, plus the exact
PGL(2,7) local HS theorem and its conventional-hopping no-go boundary.

## Reproduction

From a clean checkout of the dedicated repository:

```bash
python3 -m venv .venv
.venv/bin/python -m pip install --upgrade pip
.venv/bin/python -m pip install -r requirements.txt
export PYTHONPATH="$PWD/src"
export OPENBLAS_NUM_THREADS=1
export OMP_NUM_THREADS=1
.venv/bin/python validate_final_submission.py --focused-tests
```

The final local acceptance run reported:

```text
103 passed, 21 subtests passed
status: pass
2D-TN: 90,003 stored small-system determinant evaluations,
336,000 stored 4×4 local ratios
exterior: 132,003 stored chain determinants + 3,000 independent fields
conjugate tensor: 220,005 stored chain determinants + 5,000 independent fields
negative evaluated weights: 0
zero evaluated weights: 0
PGL closure: 336 states, 50 fields, 16,800 transitions
```

The finite numerical checks validate the implementations and estimators.
Arbitrary-size and arbitrary-depth sign claims come from the analytic proofs,
not from these sample counts. Candidate-registry states and novelty boundaries
are stated explicitly in the English report.
Loading