repo stringclasses 454
values | file_path stringlengths 5 201 | extension stringclasses 1
value | content stringlengths 8 509k | num_lines int64 3 16.9k | size_bytes int64 8 511k |
|---|---|---|---|---|---|
cs249r_book | book/cli/utils/__init__.py | .py | """
Utility functions and helpers for the MLSysBook CLI.
Contains shared utilities for console output, validation, and other
common functionality.
"""
| 7 | 152 |
cs249r_book | book/cli/checks/math_multiplier_style.py | .py | """Multiplier and ``\\times`` style checks for QMD prose.
This checker is intentionally example-heavy because most failures are visual
typography mistakes that look plausible in source.
It catches:
* ``body_multiplier_suffix``:
``speedup_str = fmt(speedup, suffix="×")`` or ``suffix="x"``.
Fix by using the typed ... | 333 | 12,860 |
cs249r_book | book/cli/checks/lego_prose_literals.py | .py | #!/usr/bin/env python3
"""Flag hardcoded numeric literals in LEGO walkthrough prose.
Policy (computation-sensitive, not number-free prose)
-----------------------------------------------------
Flag literals when they appear in a *computational walkthrough*: a callout
notebook or worked example whose nearby LEGO cell e... | 457 | 15,467 |
cs249r_book | book/cli/checks/currency_style.py | .py | r"""Currency style checks for book currency notation.
Policy:
* Reader-facing prices use ``$``.
* Source prose and LEGO currency prefixes escape the dollar sign as ``\$``.
* ``USD`` appears only once, in the shared notation definition that says
dollar-denominated costs are U.S. dollars unless noted otherwise... | 329 | 12,401 |
cs249r_book | book/cli/checks/lego_dead_code.py | .py | #!/usr/bin/env python3
"""LEGO exported-variable liveness check.
This powers::
./book/binder check code --scope lego-dead-code
Binder owns the implementation; pre-commit reaches it through ``binder check``.
"""
from __future__ import annotations
import argparse
import re
import sys
from dataclasses import data... | 207 | 6,370 |
cs249r_book | book/cli/checks/__init__.py | .py | """
Binder-native check implementations.
Check logic that powers ``./book/binder check <group> --scope …`` lives here
as ordinary Python modules. ``book/cli/commands/validate.py`` imports from
this package and converts results to ``ValidationIssue`` records.
Temporary standalone shims may import from ``cli.checks`` d... | 13 | 483 |
cs249r_book | book/cli/checks/fmt_prose_contract.py | .py | #!/usr/bin/env python3
"""
fmt_prose_contract.py — enforce the OUTPUT-formatter ↔ prose glyph contract.
Static checker for invariant **I2** (ASSESSMENT.md §3 / fmt.md §7). For every
``{python} Class.var_str`` reference in prose it looks up the formatter that
produced ``var_str`` (by AST-parsing the chapter's cells) an... | 390 | 15,736 |
cs249r_book | book/cli/checks/binder_canonical.py | .py | """Binder-as-front-door invariant check for the pre-commit config.
This powers::
./book/binder check cli --scope binder-canonical
The rule it enforces, in one sentence: **every pre-commit hook that targets
book content must dispatch through ``./book/binder``, not call a raw script.**
Why this exists
-----------... | 143 | 5,313 |
cs249r_book | book/cli/checks/percent_prose.py | .py | #!/usr/bin/env python3
"""Percent-in-prose: detect the `%` symbol where body prose should spell out 'percent'.
This is the symmetric inverse of ``percent_tables.py``. House style (MIT Press
``AU_QUERY_RESPONSES.md`` Category H, aligned with Chicago Manual of Style
§9.18) splits on *sentence vs. data*, not on context t... | 145 | 5,689 |
cs249r_book | book/cli/checks/fmt_semantic_suffix.py | .py | #!/usr/bin/env python3
"""
fmt_semantic_suffix check — value-kind must be a typed formatter, not a
free-text ``suffix=`` on the generic ``fmt()``/``fmt_int()``.
This is the regression gate for the typed-formatter migration. The semantic
kind of a value (percent, multiplier, percentage-points, count-scale) belongs
in t... | 223 | 7,256 |
cs249r_book | book/cli/checks/percent_tables.py | .py | #!/usr/bin/env python3
"""Percent-symbol-in-tables: single source of truth for the check and the fixer.
House style inverts between prose and tables:
- In PROSE and CAPTIONS, "percent" is spelled out (MIT Press §10.2); the
`%` symbol is banned there.
- In TABLES, cells are dense tabular data and the conventio... | 217 | 7,433 |
cs249r_book | book/cli/checks/mitpress_terms.py | .py | #!/usr/bin/env python3
"""MIT Press canonical spelling dictionary (§10.7) — detection + fix.
The MIT Press editorial standard §10.7 fixes the canonical spelling of a long
list of terms (Webster's 11th first spelling). This module enforces the
**unambiguous** subset — terms with a single correct form regardless of
gram... | 179 | 6,951 |
cs249r_book | book/cli/checks/markdown_list_spacing.py | .py | """Flag bold lead-in paragraphs that are immediately followed by lists.
Pandoc treats a paragraph followed directly by ``- item`` as one paragraph,
not as a separate list block. In rendered output this can collapse into text
like ``**Step 2**: ... - item``. This check catches the high-signal house
style case: a non-li... | 106 | 2,920 |
cs249r_book | book/cli/checks/prose_integrity.py | .py | """Prose-integrity detectors: sentence starts, hand-typed attributions, italics.
Added 2026-08-14 after a tone-audit pass introduced nine banned section
meta-openers, two of which left a sentence starting with a lowercase word
("...energy budget. this section introduces..."). Nothing in the gate set
noticed. These thr... | 205 | 8,327 |
cs249r_book | book/cli/checks/bib_lint.py | .py | #!/usr/bin/env python3
"""BibTeX linter, validator, and formatter for the MLSysBook project.
Enforces the canonical schema and formatting rules documented in
the project prose style guide §5 Bibliography Hygiene.
Usage:
python3 book/tools/bib_lint.py <file.bib> [--check|--fix|--report]
python3 book/tools/bib_... | 1,138 | 43,679 |
cs249r_book | book/cli/checks/cli_contract.py | .py | """Public command contract checks for the Binder CLI.
This powers::
./book/binder check cli --scope contract
The check is intentionally small and example-heavy. It runs read-only commands
that define the public Binder surface and fails if help text, migration hints,
or exit codes drift.
Examples of what it catc... | 375 | 11,985 |
cs249r_book | book/cli/checks/lego_units.py | .py | #!/usr/bin/env python3
"""LEGO unit discipline linter for QMD cells (warnings + blocking errors)."""
from __future__ import annotations
import argparse
import json
import os
import re
import subprocess
import sys
from dataclasses import dataclass
from pathlib import Path
@dataclass(frozen=True)
class LintIssue:
... | 279 | 10,849 |
cs249r_book | book/cli/checks/margin_geometry.py | .py | """Binder-native rendered-PDF margin geometry checks.
This module is the shared implementation behind ``binder layout overlaps`` and
post-build PDF validation. It intentionally depends only on rendered PDF
geometry from PyMuPDF, not on source anchors or LaTeX log heuristics.
"""
from __future__ import annotations
fr... | 391 | 12,089 |
cs249r_book | book/cli/checks/lego_prose_units.py | .py | #!/usr/bin/env python3
"""Flag redundant unit/currency tokens after closed ``{python} *_str`` prose refs.
Policy
------
Only **closed** exports (domain formatters, ``fmt_qty``, ``fmt_percent`` with
``style='prose'|'symbol'``, ``fmt_count(scale=...)``, ``fmt_multiple``, or
``*_unit_str`` names fed by open ``fmt()`` on ... | 320 | 13,002 |
cs249r_book | book/cli/checks/math_canonical.py | .py | #!/usr/bin/env python3
"""
Math canonical check — fmt-family and suffix discipline for LEGO cells.
Enforces the canonical math-rendering convention (see the project math rules):
1. Every ``*_str`` / ``*_math`` / ``*_eq`` / ``*_frac`` assignment in a
``{python}`` cell must use ``mlsysim.fmt`` helpers or ``Markd... | 917 | 35,589 |
cs249r_book | book/cli/commands/layout.py | .py | """
``binder layout`` — PDF page-layout diagnostics.
Subcommands:
check — Scan a built PDF for pages with excessive bottom whitespace
in the main body column, and guess the likely cause (the
block at the top of the next page that probably forced the
break).
The plan... | 3,742 | 146,342 |
cs249r_book | book/cli/commands/bib.py | .py | """
``binder bib`` — Bibliography management.
Subcommands:
list — Show all .bib files with entry counts
mechanical — Apply safe §5 field-level fixes (pre-commit first step)
clean — Remove unused entries from .bib files
update — Run betterbib sync-preserve on .bib files (fetch metadata, keep ci... | 547 | 21,123 |
cs249r_book | book/cli/commands/reset.py | .py | """Reset generated/local CLI state to a clean baseline."""
from __future__ import annotations
import argparse
from typing import List
from rich.console import Console
from rich.markup import escape as _rich_escape
from rich.panel import Panel
from rich.table import Table
console = Console()
class ResetCommand:
... | 73 | 2,629 |
cs249r_book | book/cli/commands/_registry_checks.py | .py | """Registry migration gates for `binder check registry`."""
from __future__ import annotations
import importlib.util
import subprocess
import sys
from dataclasses import dataclass
from pathlib import Path
@dataclass(frozen=True)
class RegistryIssue:
code: str
message: str
file: str = "(registry)"
li... | 239 | 8,559 |
cs249r_book | book/cli/commands/headings.py | .py | """
``binder headings`` — Heading-case enforcement for MIT Press style.
Enforces the H1/H2 headline-case + H3+ sentence-case policy documented in
the project prose style guide §10.3.1. Preserves ten documented exceptions:
acronyms, hyphenated-acronym compounds, digit-letter models (70B, 3D),
single-letter labels (Arch... | 569 | 24,059 |
cs249r_book | book/cli/commands/render.py | .py | """
``binder render`` — Render generated figures to a browsable gallery.
Subcommands:
plots — Render all matplotlib/Python figures from QMD files to PNG.
Outputs to ``_output/plots/<chapter>/<fig-label>.png``.
Future subcommands:
diagrams — Render TikZ diagrams (requires lualatex)
all ... | 339 | 11,602 |
cs249r_book | book/cli/commands/audit.py | .py | """Chapter-level build audits (`binder audit chapter-pdf|chapter-html`)."""
from __future__ import annotations
import argparse
import importlib.util
import sys
from pathlib import Path
from rich.console import Console
console = Console()
def _repo_root() -> Path:
return Path(__file__).resolve().parents[3]
d... | 73 | 2,580 |
cs249r_book | book/cli/commands/maintenance.py | .py | """
Maintenance commands for MLSysBook CLI.
Handles setup, switch, hello, about, and other maintenance operations.
"""
import argparse
import hashlib
import json
import os
import re
import subprocess
import shutil
import time
from collections import defaultdict
from datetime import datetime
from pathlib import Path
f... | 1,108 | 49,068 |
cs249r_book | book/cli/commands/clean.py | .py | """
Clean command implementation for MLSysBook CLI.
Handles cleaning build artifacts and restoring configurations.
"""
import shutil
from rich.console import Console
from rich.markup import escape as _rich_escape
from rich.panel import Panel
from rich.table import Table
from cli.core.artifacts import clean_build_art... | 212 | 8,091 |
cs249r_book | book/cli/commands/newsletter.py | .py | """
Newsletter command implementation for MLSysBook CLI.
Manages newsletter drafts, publishing to Buttondown, and fetching
sent newsletters for the website archive.
Requires BUTTONDOWN_API_KEY environment variable for API operations.
"""
import json
import os
import re
import shutil
from datetime import datetime
fro... | 426 | 17,410 |
cs249r_book | book/cli/commands/_index_checks.py | .py | """Index check primitives for `binder check index` scopes."""
from __future__ import annotations
import re
from dataclasses import dataclass
from pathlib import Path
INDEX_RE = re.compile(r"\\index\{([^{}]*(?:\{[^{}]*\}[^{}]*)?)\}")
SEEREF_RE = re.compile(r"^([^|]+)\|see(?:also)?\{([^}]+)\}$")
GENERIC_BARE = {
... | 195 | 7,058 |
cs249r_book | book/cli/commands/info.py | .py | """
``binder info`` — Book statistics and figure extraction.
Subcommands:
stats — Count figures, tables, equations, listings, text lines, words
figures — Extract figure list with labels, captions, and alt-text
Use --with-pdf to merge LaTeX figure numbers and page numbers
from ... | 1,129 | 44,557 |
cs249r_book | book/cli/commands/preview.py | .py | """
Preview command implementation for MLSysBook CLI.
Handles starting development servers with live reload for interactive development.
"""
import subprocess
import signal
import sys
from pathlib import Path
from typing import List, Optional
from rich.console import Console
console = Console()
class PreviewComman... | 118 | 4,009 |
cs249r_book | book/cli/commands/__init__.py | .py | """
Command implementations for the MLSysBook CLI.
Each command is implemented as a separate module for better maintainability
and testing.
"""
| 7 | 145 |
cs249r_book | book/cli/commands/_pdf_checks.py | .py | """PDF post-build verification used by `binder build pdf` and `binder check pdf`.
Scans rendered PDF text (via ``pdftotext``) for defects Quarto/LuaLaTeX can
emit without failing the render: unresolved cross-refs (``?@sec-foo``), literal
unrendered cross-refs (``@sec-foo``), undefined LaTeX references
(``Figure ??``),... | 804 | 28,698 |
cs249r_book | book/cli/commands/debug.py | .py | """
Debug command implementation for MLSysBook CLI.
Two-phase build debugger that isolates build failures:
Phase 1: Scan chapters one-by-one to find which chapter(s) fail
Phase 2: Binary search within a failing chapter to find the exact section
Usage via binder:
./binder debug pdf --vol1 # ... | 615 | 22,618 |
cs249r_book | book/cli/commands/reference_check.py | .py | """
Native reference check: validate .bib entries against academic DBs (hallucinator).
Used by `binder check references --scope hallucinator`. Requires: pip install hallucinator bibtexparser
(or install optional extra: pip install -e ".[reference-check]").
Optional env: OPENALEX_KEY, S2_API_KEY. Note: Semantic Schola... | 428 | 16,817 |
cs249r_book | book/cli/commands/_epub_checks.py | .py | """EPUB check primitives used by the `binder check epub` command group.
This module holds the pure-Python logic for the two `check epub` scopes
so that `validate.py` can call it as ordinary Python (not via a
subprocess-to-a-script, which is what the older delegated-script
pattern used). All checks here return a list o... | 791 | 30,328 |
cs249r_book | book/cli/commands/release.py | .py | """Release orchestration for Binder.
`binder release` is the high-level release gate: it runs the checks that a
human release audit currently remembers by hand, classifies build/layout
outcomes, and writes one structured report for follow-up fixes.
"""
from __future__ import annotations
import argparse
import json
i... | 637 | 21,431 |
cs249r_book | book/cli/commands/build.py | .py | """
Build command implementation for MLSysBook CLI.
Handles building chapters and full books in different formats (HTML, PDF, EPUB).
"""
import os
import platform
import subprocess
import signal
import sys
from pathlib import Path
from typing import Dict, List, Optional, Tuple
from rich.console import Console
from ri... | 1,903 | 86,445 |
cs249r_book | book/cli/commands/doctor.py | .py | """
Health check command for MLSysBook CLI.
Performs comprehensive system health checks to ensure everything is working properly.
"""
import subprocess
import sys
from pathlib import Path
from typing import List, Tuple, Dict, Any
from rich.console import Console
from rich.table import Table
from rich.panel import Pan... | 608 | 23,503 |
cs249r_book | book/cli/commands/formatting.py | .py | """
Format commands for MLSysBook CLI.
Auto-formatters for QMD content: blank lines, Python code blocks,
list spacing, div spacing, and table formatting.
Usage:
binder format blanks — Collapse extra blank lines
binder format python — Format Python code blocks (Black; display 70, LEGO 150)
binder forma... | 572 | 22,732 |
cs249r_book | interviews/vault/topic_schema.py | .py | """Pydantic schema for the canonical topic taxonomy.
Validates topics.json: unique IDs, kebab-case format, valid areas,
prerequisite existence, and DAG acyclicity.
Usage:
python3 topic_schema.py # Validate topics.json
python3 topic_schema.py --stats # Print topology stats
python3 top... | 261 | 8,081 |
cs249r_book | interviews/vault/vault.py | .py | #!/usr/bin/env python3
"""vault.py — Unified CLI for the StaffML interview question vault.
Usage:
python3 vault.py validate # Schema check
python3 vault.py stats # Full statistics
python3 vault.py gaps # 3D coverage cube analysis
python3 vault.py dedup # Multi-stage dedup (exa... | 2,823 | 108,968 |
cs249r_book | interviews/vault/schema.py | .py | """Pydantic schema for the StaffML question corpus (schema v1.0).
Enum values are imported from :mod:`vault.schema.enums` — the single source
of truth. Do not redefine them here. See ``schema/question_schema.yaml`` for
the canonical LinkML schema.
This module validates dict records (e.g. loaded from corpus.json or fr... | 370 | 12,091 |
cs249r_book | interviews/vault/scripts/iterate_coverage_loop.py | .py | #!/usr/bin/env python3
"""Iterative coverage loop: analyze → generate → render → judge → apply.
The loop keeps tightening corpus balance by re-analyzing after every
generation pass. It stops automatically when the corpus reaches a
steady state — no big gaps remain, hallucination rate spikes, or
budget is exhausted. Th... | 334 | 13,485 |
cs249r_book | interviews/vault/scripts/generate.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". Coverage-survey-driven generation (find empty
# cells, fill emptiest first, stop when saturated). `vault generate` does
# per-cell generation but NOT the auto-balance loop. Ada... | 682 | 30,263 |
cs249r_book | interviews/vault/scripts/audit_question_backfill_balance.py | .py | #!/usr/bin/env python3
"""Audit StaffML `question` field coverage and balance after backfill."""
from __future__ import annotations
import argparse
import json
import re
import statistics
from collections import Counter, defaultdict
from pathlib import Path
import yaml
VAULT_DIR = Path(__file__).resolve().parent.par... | 183 | 5,973 |
cs249r_book | interviews/vault/scripts/reclassify_zone_bloom_mismatch.py | .py | #!/usr/bin/env python3
"""Reclassify questions whose (zone, bloom_level) pair violates the matrix.
The ZONE_BLOOM_AFFINITY matrix in interviews/vault/schema/enums.py is the
canonical "this zone admits these Bloom verbs" rule. Items that violate
the matrix have a self-contradicting classification: zone says one thing,
... | 122 | 4,205 |
cs249r_book | interviews/vault/scripts/audit_applicability_matrix.py | .py | #!/usr/bin/env python3
"""Audit StaffML topic-track applicability sources.
Compares:
- schema enum topics,
- taxonomy_data.yaml topic-track applicability,
- paper app_matrix.tex topic labels,
- observed YAML topic-track pairs.
This report is advisory. It helps decide whether a sparse/invalid-looking cell
should be ge... | 197 | 7,165 |
cs249r_book | interviews/vault/scripts/compare_semantic_passes.py | .py | #!/usr/bin/env python3
"""Compare two semantic audit passes and summarize agreement."""
from __future__ import annotations
import argparse
import collections
import json
from pathlib import Path
from typing import Any
VAULT_DIR = Path(__file__).resolve().parents[1]
RESULTS_DIR = VAULT_DIR / "audit" / "semantic-revie... | 109 | 4,232 |
cs249r_book | interviews/vault/scripts/format_yaml_questions.py | .py | #!/usr/bin/env python3
"""Canonical formatter for StaffML question YAML files."""
from __future__ import annotations
import argparse
import difflib
import sys
from pathlib import Path
from typing import Any
import yaml
VAULT_DIR = Path(__file__).resolve().parents[1]
REPO_ROOT = VAULT_DIR.parents[1]
QUESTIONS_DIR = ... | 245 | 7,371 |
cs249r_book | interviews/vault/scripts/preprocess.py | .py | #!/usr/bin/env python3
"""Preprocess QMD textbook chapters into clean prose for taxonomy extraction.
Strips code blocks, LaTeX, TikZ, figures, tables, and Quarto markup.
Keeps section headers, paragraph prose, and lists.
Usage:
python3 preprocess.py # Process all chapters
python3 prep... | 197 | 6,516 |
cs249r_book | interviews/vault/scripts/summarize_semantic_audit.py | .py | #!/usr/bin/env python3
"""Summarize semantic audit JSONL files into a release report."""
from __future__ import annotations
import argparse
import collections
import json
from pathlib import Path
from typing import Any
VAULT_DIR = Path(__file__).resolve().parents[1]
RESULTS_DIR = VAULT_DIR / "audit" / "semantic-revi... | 111 | 3,454 |
cs249r_book | interviews/vault/scripts/gemini_backfill_question.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". The corpus walk targets the post-v1.0 YAML layout,
# but the prompt / batching / threadpool pattern is reused by the upcoming
# audit_corpus_batched.py (CORPUS_HARDENING_PLAN.m... | 338 | 13,164 |
cs249r_book | interviews/vault/scripts/repair_registry.py | .py | #!/usr/bin/env python3
"""Append-only repair of id-registry.yaml.
The registry is an append-only audit log of every ID ever assigned. Two
prior events broke its integrity:
1. Commit ``8a5c3ff3c`` (2026-04-25) renamed 4,754 cohort-tagged IDs
(e.g. ``cloud-fill-04027``, ``tinyml-cell-13251``) to clean
``<track>-N... | 118 | 4,570 |
cs249r_book | interviews/vault/scripts/gemini_cli_generate_questions.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". `vault generate` (vault-cli) is the modern
# entrypoint but does NOT batch — it calls per-question. This script's batching
# (12 cells / call, balanced track×area×zone×level ro... | 781 | 35,326 |
cs249r_book | interviews/vault/scripts/vault_fill.py | .py | #!/usr/bin/env python3
"""
Vault Fill — Parallel question generation for every empty cube cell.
Architecture:
1. Analyze the 3D coverage cube
2. Generate one shell command per deficit cell
3. Each command writes to a temp JSON file (no file conflicts)
4. Run ALL commands in parallel (via xargs or background processes)... | 334 | 11,641 |
cs249r_book | interviews/vault/scripts/semantic_audit_questions.py | .py | #!/usr/bin/env python3
"""Parallel semantic audit runner for published StaffML questions.
This consumes JSONL queues created by prepare_semantic_review_queue.py and
appends one structured finding per question. It is resumable: existing qids in
the output file are skipped on later runs. The runner batches a few questio... | 300 | 10,862 |
cs249r_book | interviews/vault/scripts/vault_invariants.py | .py | #!/usr/bin/env python3
"""Vault Invariant Checks — structural guardrails for StaffML data integrity.
This script enforces invariants ACROSS the three data files (corpus.json,
taxonomy.json, chains.json) that per-question schema validation cannot catch.
Run it as a pre-commit gate or CI check whenever vault data change... | 588 | 24,843 |
cs249r_book | interviews/vault/scripts/migrate_to_v1_0.py | .py | """One-time migration: corpus.json -> per-question YAML at schema_version 1.0.
This script replaces the `split_corpus.py` migration that introduced the
bugs fixed here. It writes one YAML file per question into a flat-by-track
layout:
questions/<track>/<id>.yaml
Every axis (track, level, zone, topic, competency_... | 405 | 13,198 |
cs249r_book | interviews/vault/scripts/repair_chains.py | .py | #!/usr/bin/env python3
"""Repair chain integrity issues in the StaffML question bank.
Two kinds of chain damage accrue over time:
1. **Orphan singletons**: chains where the only remaining member is a
single question. Sibling questions were deleted/renamed/archived,
leaving a chain-of-one. By definition a chain ... | 176 | 6,661 |
cs249r_book | interviews/vault/scripts/audit_visual_questions.py | .py | #!/usr/bin/env python3
"""Audit and plan StaffML visual-question coverage."""
from __future__ import annotations
import json
from collections import Counter
from pathlib import Path
from typing import Any
import yaml
VAULT_DIR = Path(__file__).resolve().parent.parent
ROOT_DIR = VAULT_DIR.parents[1]
QUESTIONS_DIR = ... | 177 | 6,650 |
cs249r_book | interviews/vault/scripts/run_semantic_audit_tracks.py | .py | #!/usr/bin/env python3
"""Launch semantic audit jobs per track with separate output files."""
from __future__ import annotations
import argparse
import os
import shlex
import subprocess
import sys
from pathlib import Path
VAULT_DIR = Path(__file__).resolve().parents[1]
QUEUE_DIR = VAULT_DIR / "audit" / "semantic-rev... | 73 | 2,428 |
cs249r_book | interviews/vault/scripts/prepare_semantic_review_queue.py | .py | #!/usr/bin/env python3
"""Build a semantic review queue for published StaffML questions."""
from __future__ import annotations
import argparse
import json
from pathlib import Path
from typing import Any
import yaml
VAULT_DIR = Path(__file__).resolve().parents[1]
REPO_ROOT = VAULT_DIR.parents[1]
QUESTIONS_DIR = VAUL... | 108 | 4,166 |
cs249r_book | interviews/vault/scripts/build_semantic_fix_queue.py | .py | #!/usr/bin/env python3
"""Build a prioritized fix queue from semantic audit JSONL results."""
from __future__ import annotations
import argparse
import collections
import json
from pathlib import Path
from typing import Any
VAULT_DIR = Path(__file__).resolve().parents[1]
RESULTS_DIR = VAULT_DIR / "audit" / "semantic... | 104 | 3,437 |
cs249r_book | interviews/vault/scripts/gemini_fix_errors.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". The HARDWARE_REFERENCE constant (V100/A100/H100/B200/T4
# specs as ground-truth context for the judge) is exactly what
# audit_corpus_batched.py needs in Phase 5 (CORPUS_HARDEN... | 236 | 8,612 |
cs249r_book | interviews/vault/scripts/generate_gaps.py | .py | #!/usr/bin/env python3
"""
Gap-fill generator for underfilled StaffML corpus cells.
Reads corpus.json, identifies all (track, level, competency_area) cells with
fewer than 3 questions, then generates the missing questions using either:
- Gemini 2.5 Flash via the `gemini` CLI (default)
- Claude Opus 4.6 via the Ant... | 643 | 23,505 |
cs249r_book | interviews/vault/scripts/vault_loop.py | .py | #!/usr/bin/env python3
"""
Vault Loop — Overnight autonomous generation for balanced question coverage.
Unlike vault_fill.py (which fills deficit cells once), this script:
1. Fills deficit cells (coverage gaps)
2. Balances over-represented cells by boosting under-represented ones
3. Loops until the cube is both FULL a... | 465 | 16,252 |
cs249r_book | interviews/vault/scripts/gpt_backfill_question.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". OpenAI variant of gemini_backfill_question.py.
# Useful when the Gemini quota is exhausted or for cross-provider comparison.
# Verify the OpenAI SDK version + the chat-completi... | 417 | 15,981 |
cs249r_book | interviews/vault/scripts/plan_gap_improvements.py | .py | #!/usr/bin/env python3
"""Build StaffML gap-analysis and generation-planning artifacts.
This script implements the release-oriented improvement plan:
- canonical v1 coverage cube from YAML source,
- repair backlog for metadata/content drift,
- 50-question pilot pack,
- validation gates for generated items,
- scaled 2... | 1,058 | 41,039 |
cs249r_book | interviews/vault/scripts/scorecard.py | .py | #!/usr/bin/env python3
"""
StaffML Quality Scorecard — single-command corpus health check.
Usage:
python3 staffml/vault/scripts/scorecard.py
python3 staffml/vault/scripts/scorecard.py --compare path/to/previous.json
"""
import json
import argparse
from collections import Counter, defaultdict
from datetime imp... | 242 | 8,535 |
cs249r_book | interviews/vault/scripts/generate_hard_questions.py | .py | #!/usr/bin/env python3
"""Generate ~500 hard (L4-L6+) questions to fill level distribution gaps.
Uses Gemini API (gemini-3.1-pro-preview) to generate questions across all 35
knowledge areas, targeting deficit levels. Each batch includes examples from
the existing corpus and valid concept tags for that KA.
Usage:
... | 560 | 24,684 |
cs249r_book | interviews/vault/scripts/audit_resources_migration.py | .py | #!/usr/bin/env python3
"""
Phase 0 audit for the deep_dive → resources migration.
Walks every question YAML under interviews/vault/questions/, extracts the
details.deep_dive field (when present), and emits counts that determine
the migration's blast radius:
- total questions
- questions with deep_dive present / a... | 183 | 7,280 |
cs249r_book | interviews/vault/scripts/fix_yaml_hygiene.py | .py | #!/usr/bin/env python3
"""Apply conservative mechanical hygiene fixes to question YAML files."""
from __future__ import annotations
import argparse
import re
from pathlib import Path
from typing import Any
import yaml
VAULT_DIR = Path(__file__).resolve().parents[1]
QUESTIONS_DIR = VAULT_DIR / "questions"
CODE_SPAN... | 205 | 12,249 |
cs249r_book | interviews/vault/scripts/validate_generation_gates.py | .py | #!/usr/bin/env python3
"""Validate generated StaffML question candidates against release gates.
The script is intentionally local and deterministic. It checks the gates that
do not require remote model calls: schema, question shape, duplication signals,
topic-track applicability, zone-level affinity, chain references,... | 245 | 9,684 |
cs249r_book | interviews/vault/scripts/migrate_to_hierarchical_layout.py | .py | #!/usr/bin/env python3
"""One-shot migration: flat <track>/<id>.yaml -> hierarchical <track>/<area>/<id>.yaml.
Reads each YAML's `track` and `competency_area` fields, then moves the file
to the corresponding hierarchical path. Idempotent — running on an
already-migrated tree is a no-op.
Validates pre-flight that ever... | 116 | 3,863 |
cs249r_book | interviews/vault/scripts/fix_competency_areas.py | .py | #!/usr/bin/env python3
"""Phase 0 cleanup: remap malformed `competency_area` values to canonical.
The schema's `competency_area` field is documented as "one of 13 canonical
areas" but is enforced as a free-form string at the LinkML layer. This
left a hole that Gemini-generated questions slipped through: when asked
to ... | 216 | 8,729 |
cs249r_book | interviews/vault/scripts/deep_verify.py | .py | #!/usr/bin/env python3
# STATUS (2026-05-03): preserved as a reference pattern — see vault/scripts/DEPRECATED.md
# §"Preserved for adaptation". Unique among the audit scripts: Claude Opus +
# extended thinking, asks the model to SHOW ITS WORK on every napkin-math claim.
# Useful as a tiebreaker when audit_corpus_batche... | 306 | 12,509 |
cs249r_book | interviews/vault/scripts/portfolio_balance_loop.py | .py | #!/usr/bin/env python3
"""Plan iterative StaffML portfolio-balancing passes."""
from __future__ import annotations
import argparse
import json
import sys
from collections import Counter
from pathlib import Path
from typing import Any
import yaml
VAULT_DIR = Path(__file__).resolve().parent.parent
ROOT_DIR = VAULT_DI... | 237 | 9,332 |
cs249r_book | interviews/vault/scripts/verify_math.py | .py | #!/usr/bin/env python3
"""Gemini-powered math verification pass for StaffML corpus.
Sends chunks of questions to gemini-3.1-pro-preview for independent
math verification. Each call checks ~25 questions. With 250 calls/day
quota, this covers ~6,250 questions per day.
Usage:
python3 scripts/verify_math.py ... | 231 | 8,006 |
cs249r_book | interviews/vault/scripts/generate_batch.py | .py | #!/usr/bin/env python3
"""Fast parallel batch generator — fills ALL topic×track×zone gaps.
Reads pre-computed jobs from /tmp/staffml_jobs.json and spawns
gemini CLI calls in parallel, writing results to a temp file.
Usage:
python3 generate_batch.py --workers 40 --output /tmp/batch_all.json
"""
import argparse
im... | 152 | 5,730 |
cs249r_book | interviews/vault/scripts/render_visuals.py | .py | #!/usr/bin/env python3
"""Render question visuals to ship-ready SVG.
The schema the website cares about is minimal::
visual:
kind: svg # always svg — that's what the website ships
path: <id>.svg # the static asset
alt: <text>
caption: <text>
# Build metadata ... | 276 | 10,450 |
cs249r_book | interviews/vault/scripts/validate_questions.py | .py | #!/usr/bin/env python3
"""Parallel Gemini validation of corpus questions.
Validates math correctness, factual accuracy, and question quality
using gemini-3.1-pro-preview across parallel batches.
Usage:
python3 validate_questions.py # Validate all 4,779 questions
python3 validate_questions.p... | 363 | 14,738 |
cs249r_book | interviews/vault/scripts/audit_yaml_corpus.py | .py | #!/usr/bin/env python3
"""Read-only deterministic audit for StaffML question YAML files.
This script intentionally ignores historical LLM audit artifacts. It validates
the current YAML corpus against the local schema and authoring conventions, then
writes a JSONL issue log plus a short Markdown summary.
"""
from __fu... | 459 | 16,595 |
cs249r_book | interviews/vault/scripts/promote_validated.py | .py | #!/usr/bin/env python3
"""Promote LLM-judge-PASS draft YAMLs to status:published.
Reads the latest llm_judge summary.json files, collects every question
ID that received a PASS verdict and is still a status:draft, and flips
its lifecycle fields to the canonical published shape:
status: draft -> publis... | 131 | 4,356 |
cs249r_book | interviews/vault/scripts/analyze_coverage_gaps.py | .py | #!/usr/bin/env python3
"""Coverage-gap analyzer for the StaffML question corpus.
Surfaces where the corpus is thin so generation can be aimed precisely
rather than scattered. Output is a structured report that downstream
batched generation (`gemini_cli_generate_questions.py`) can consume.
The dimensions analyzed:
... | 461 | 19,642 |
cs249r_book | interviews/vault/scripts/rename_legacy_ids.py | .py | #!/usr/bin/env python3
"""Rename cohort-tagged legacy IDs to clean <track>-NNNN form.
The 2026-04-21 ID_SCHEMES.md retired the cohort-tagged form (cloud-fill-*,
cloud-cell-*, etc.) but kept legacy IDs unchanged because rewriting them
would have broken ~3,100 chain references and external bookmarks.
This script applie... | 243 | 9,396 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1978.py | .py | import os
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(5,4))
ax.bar(['TCM Hit', 'LPDDR5x Spill'], [1, 100], color=['#d4edda', '#fdebd0'], edgecolor=['#3d9e5a', '#c87b2a'])
ax.set_ylabel('Access Latency (ns, Approx)')
ax.set_title('Impact of TCM Miss on NPU')
plt.savefig(os.environ['VISUAL_OUT_PATH'],... | 7 | 355 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1965.py | .py | import os
import matplotlib.pyplot as plt
tiers = ['UFS 4.0', 'LPDDR5X', 'NPU SRAM']
bw = [4, 60, 1000]
plt.figure(figsize=(6, 3))
plt.barh(tiers, bw, color=['#c87b2a', '#4a90c4', '#3d9e5a'])
plt.xlabel('Bandwidth (GB/s, log scale)')
plt.xscale('log')
plt.title('Mobile Memory Tier Bandwidths')
plt.savefig(os.environ.... | 12 | 389 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1890.py | .py | import os
import numpy as np
import matplotlib.pyplot as plt
n = np.arange(0, 10)
rho = 0.75
prob = (1 - rho) * (rho**n)
plt.figure(figsize=(6,4))
plt.bar(n[:5], prob[:5], color='#cfe2f3', label='NPU')
plt.bar(n[5:], prob[5:], color='#fdebd0', label='CPU Fallback')
plt.xlabel('Tasks in System')
plt.ylabel('Probability... | 15 | 447 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1914.py | .py | import os
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(6,2))
ax.broken_barh([(0,2)], (0,1), facecolors='#cfe2f3', label='ISP Write')
ax.broken_barh([(2,2)], (0,1), facecolors='#d4edda', label='GPU Convert')
ax.broken_barh([(4,3)], (0,1), facecolors='#fdebd0', label='NPU Infer')
ax.set_xlim(0, 8)
ax.s... | 12 | 501 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1909.py | .py | import os
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(6,2))
ax.broken_barh([(0,13), (20,13)], (0, 5), facecolors='#cfe2f3', label='Sleep')
ax.broken_barh([(13,2), (33,2)], (0, 200), facecolors='#fdebd0', label='Transient')
ax.broken_barh([(15,5), (35,5)], (0, 450), facecolors='#d4edda', label='Activ... | 11 | 510 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1979.py | .py | import os
import matplotlib.pyplot as plt
import numpy as np
fig, ax = plt.subplots(figsize=(5,2))
ax.barh(['Sequential'], [2], color='#d4edda', edgecolor='#3d9e5a', label='ISP')
ax.barh(['Sequential'], [3], left=[2], color='#cfe2f3', edgecolor='#4a90c4', label='NPU')
ax.legend()
ax.set_xlabel('Latency (ms)')
plt.savef... | 9 | 403 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1912.py | .py | import os
import matplotlib.pyplot as plt
fig, ax = plt.subplots(figsize=(5,3))
ax.bar(['FP16', 'INT4'], [40, 10], color='#cfe2f3', edgecolor='#4a90c4')
ax.axhline(16, color='red', linestyle='--', label='SRAM Limit')
ax.set_ylabel('Model Size (MB)')
ax.legend()
out = os.environ.get('VISUAL_OUT_PATH', 'out.svg')
plt.sav... | 9 | 364 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1887.py | .py | import os
import matplotlib.pyplot as plt
ctx = [1000, 2000, 3814, 5000]
mem = [128, 256, 500, 640]
plt.figure(figsize=(6,4))
plt.bar([str(c) for c in ctx], mem, color='#cfe2f3', edgecolor='#4a90c4')
plt.axhline(500, color='red', linestyle='--', label='500MB Limit')
plt.xlabel('Sequence Length')
plt.ylabel('Memory (MB... | 13 | 448 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1985.py | .py | import os
import matplotlib.pyplot as plt
import numpy as np
fig, ax = plt.subplots(figsize=(6,2))
ax.barh(['Memory'], [1.5], color='#d4edda', edgecolor='#3d9e5a', label='Weights (1.5GB)')
ax.barh(['Memory'], [2.5], left=[1.5], color='#cfe2f3', edgecolor='#4a90c4', label='KV Cache (2.5GB)')
ax.set_xlim(0, 4.5)
ax.axvli... | 11 | 533 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1895.py | .py | import os
import matplotlib.pyplot as plt
import numpy as np
t = np.linspace(0, 200, 400)
active = np.where(t % 50 < 5, 1, 0)
plt.plot(t, active, color='#3d9e5a', drawstyle='steps-pre')
plt.xlabel('Time (ms)')
plt.yticks([0, 1], ['Sleep', 'Active'])
plt.ylim(-0.2, 1.2)
out = os.environ.get('VISUAL_OUT_PATH', 'out.svg')... | 11 | 372 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1872.py | .py | import matplotlib.pyplot as plt
import numpy as np
import os
lambda_rate = np.linspace(0.1, 2.4, 50)
mu = 1 / 0.4 # 2.5 req/s
# M/D/1 wait time: rho = lambda/mu. W_q = (rho * service_time) / (2 * (1 - rho))
rho = lambda_rate / mu
wait_time = (rho * 0.4) / (2 * (1 - rho))
fig, ax = plt.subplots(figsize=(6, 4))
ax.plot... | 22 | 782 |
cs249r_book | interviews/vault/visuals/mobile/mobile-1901.py | .py | import os
import matplotlib.pyplot as plt
import numpy as np
t = np.linspace(0, 10, 200)
power = np.where(t % 1 < 0.1, 50, 2)
plt.plot(t, power, color='#c87b2a', drawstyle='steps-pre')
plt.xlabel('Time')
plt.ylabel('Power (mW)')
plt.ylim(0, 60)
out = os.environ.get('VISUAL_OUT_PATH', 'out.svg')
plt.savefig(out, format=... | 11 | 347 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.