text
stringlengths
185
73.3k
repo
stringlengths
7
100
path
stringlengths
4
146
language
stringclasses
7 values
hash
stringlengths
16
16
score
float64
7
8.5
stars
int64
0
237k
""" Discord channel adapter for RagLeap Core. Receives Discord messages via webhook, answers them using the core RAG pipeline (core.chat.ask), and sends the response back. Single-tenant: one bot, one document set, configured entirely via .env. """ import os import logging import requests from core.chat import ask fr...
antonyrag/ragleap-core
channels/discord/router.py
.py
8b3742ae6f302ccf
7.45
7
""" Telegram channel adapter for RagLeap Core. Receives Telegram messages via webhook, answers them using the core RAG pipeline (core.chat.ask), and sends the response back. Single-tenant: one bot, one document set, configured entirely via .env. """ import os import hmac import hashlib import logging import requests ...
antonyrag/ragleap-core
channels/telegram/router.py
.py
71bf37e0c20ea5ac
7.45
7
""" Speech-to-text and text-to-speech for the RagLeap Core voice channel. v1 supports OpenAI Whisper (STT) and OpenAI TTS only — Deepgram and ElevenLabs are documented as good-first-issue additions, not yet implemented. """ import io import logging import requests logger = logging.getLogger(__name__) OPENAI_STT_URL ...
antonyrag/ragleap-core
channels/voice/audio.py
.py
1852e1902a0e462d
7.45
7
""" WhatsApp channel adapter for RagLeap Core. Receives WhatsApp messages via Twilio or Gupshup webhooks, answers them using the core RAG pipeline (core.chat.ask), and sends the response back. Single-tenant: one bot, one document set, configured entirely via .env. No workspace concept, no Manager AI, no lead capture —...
antonyrag/ragleap-core
channels/whatsapp/router.py
.py
f40977f77a26c3db
7.45
7
""" Chat Pipeline for RagLeap Core Wires together: query language detection -> embedding the query -> hybrid (dense + sparse, graph-boosted) retrieval -> answer generation (blocking or streaming). """ import logging from typing import Iterator, Optional from core.embedding import EmbeddingService from core.retrieval i...
antonyrag/ragleap-core
core/chat.py
.py
a90f05957dcda201
7.45
7
""" Text Chunking Service for RagLeap Core Adapted from RagLeap's production chunking logic — Django dependency removed, now uses plain constructor defaults instead of settings.py. """ import re from typing import List import logging logger = logging.getLogger(__name__) DEFAULT_CHUNK_SIZE = 512 DEFAULT_CHUNK_OVERLAP ...
antonyrag/ragleap-core
core/chunker.py
.py
990bc73d2eb15946
7.45
7
""" Embedding Service for RagLeap Core Uses Google Gemini embeddings (gemini-embedding-001, 3072 dimensions) — the same embedding technology used in RagLeap's production platform. Bring-your-own-key only: this service NEVER falls back to a shared or system-provided key. You must supply your own GEMINI_API_KEY. """ imp...
antonyrag/ragleap-core
core/embedding.py
.py
1f560093ccc8a918
7.45
7
""" Owner-configurable default role per channel, plus a deterministic intent-based override for a safe subset of customer-facing roles. Design choice, stated plainly: automatic intent detection is deliberately scoped to roles already marked customer-facing in their own DEFAULT_ROLES channels config (support, sales, ma...
antonyrag/ragleap-core
core/employees/channel_roles.py
.py
33def66bb62c5116
7.45
7
""" Real outcome signal for outcome-weighted memory: tracks the last role-based reply per (channel, chat_id) so an owner can send a quick feedback command afterward (thumbs up/down, "helpful"/"not helpful") that gets attributed to the correct employee_memory entries via core.employees.learning.record_role_memory_outcom...
antonyrag/ragleap-core
core/employees/feedback.py
.py
c2c149d683797b76
7.45
7
""" Single-tenant learned-memory store for AI Employees. Same design as production's skill_context.py write path (dedupe by content hash, importance-weighted, semantic + tag-fallback retrieval) but backed by its own employee_memory table instead of the multi-tenant MemoryEntry system. """ import hashlib import json imp...
antonyrag/ragleap-core
core/employees/memory.py
.py
ceece1c34cd3183b
7.45
7
"""AI Employee role CRUD — single-tenant port of production's AIEmployeeRole model.""" import json import logging from datetime import datetime from typing import Dict, List, Optional from core.employees._db import get_connection from core.employees.defaults import DEFAULT_ROLES logger = logging.getLogger(__name__) ...
antonyrag/ragleap-core
core/employees/roles.py
.py
fe13f1d6355edae5
7.45
7
"""Read path: get_role_skills / get_role_personality / get_capability_summary.""" import logging from core.employees import memory, roles from core.employees.defaults import ROLE_SKILL_TAGS, DEFAULT_ROLES from core.employees._db import get_connection logger = logging.getLogger(__name__) def get_role_skills(role: st...
antonyrag/ragleap-core
core/employees/skills.py
.py
38f78737935fd0cd
7.45
7
""" Answer Generation Service for RagLeap Core Bring-your-own-key only — no system key, no fallback provided by RagLeap itself. You can configure your OWN fallback chain across providers you have keys for (see LLM_FALLBACK_PROVIDERS below). Supports multiple LLM providers via LLM_PROVIDER env var (default: gemini). Na...
antonyrag/ragleap-core
core/generation.py
.py
f10fd7595857d3e0
7.45
7
""" Knowledge Graph Service for RagLeap Core. Neo4j-backed entity extraction and graph traversal, layered on top of the existing pgvector retrieval. Single-tenant: no workspace scoping needed, all graph nodes belong to this one deployment. Ported from RagLeap's production retrieval/graph_service.py — same entity extr...
antonyrag/ragleap-core
core/graph.py
.py
71835ed4895363ba
7.45
7
""" Document Ingestion Pipeline for RagLeap Core Wires together: chunking -> embedding -> database storage -> knowledge graph -> language detection. """ import os import logging import uuid from core.chunker import TextChunker from core.embedding import EmbeddingService from core.graph import graph_service from core.l...
antonyrag/ragleap-core
core/ingest.py
.py
8d66456da212981c
7.45
7
""" Base classes for RagLeap Core's data source integrations. Ported from production's api/addon_models.py + api/addon_services.py — Django model fields replaced with a plain dataclass, Fernet encryption kept (same library, no Django dependency), workspace scoping removed (single-tenant). """ import os import logging f...
antonyrag/ragleap-core
core/integrations/base.py
.py
6d88134bdf4f0555
7.45
7
""" CSV connector for RagLeap Core integrations. Content is stored directly in Postgres (data_sources.csv_content) rather than on disk — the app container has no persistent volume mount, so a file saved to the container filesystem would be lost on the next `docker compose up --build`. """ import csv import io import lo...
antonyrag/ragleap-core
core/integrations/csv_connector.py
.py
76f80c20bd58995e
7.45
7
""" Data source CRUD and sync orchestration for RagLeap Core integrations. Talks directly to Postgres (plain psycopg2, matching core/ingest.py and core/retrieval.py's pattern) rather than through an ORM. """ import os import json import logging import uuid from typing import Dict, Any, List, Optional from core.integra...
antonyrag/ragleap-core
core/integrations/service.py
.py
8291040082429579
7.45
7
""" Document text extraction for RagLeap Core. Supports .txt, .pdf, .docx — extracts plain text so core.ingest.ingest_document can chunk/embed/store it exactly the same way regardless of source format. """ import io import logging logger = logging.getLogger(__name__) SUPPORTED_EXTENSIONS = {".txt", ".pdf", ".docx"} ...
antonyrag/ragleap-core
core/parsers.py
.py
36fd96a35163c670
7.45
7
""" Vector Retrieval Service for RagLeap Core Adapted from RagLeap's production pgvector cosine-distance search — rewritten as plain SQL (psycopg2), no Django ORM, no multi-tenancy. Supports dense (vector), sparse (Postgres full-text), and hybrid (Reciprocal Rank Fusion of both) retrieval, optionally boosted by the kno...
antonyrag/ragleap-core
core/retrieval.py
.py
8cf6aaf07be4f8b6
7.45
7
""" Example: Ingest a document and ask a question about it. Prerequisites: - RagLeap Core running locally (docker compose up --build -d) - A .txt, .pdf, or .docx file to test with Usage: python examples/01_ingest_and_query.py path/to/document.txt "Your question here" """ import sys import requests API_URL = ...
antonyrag/ragleap-core
examples/01_ingest_and_query.py
.py
a676a2855d621057
7.45
7
""" Example: Test a channel adapter's RAG-answering logic directly, without needing real WhatsApp/Telegram/Discord credentials configured. Useful for verifying document Q&A works correctly before wiring up real webhook credentials for any channel. Usage: python examples/02_test_channel_directly.py <channel> <ques...
antonyrag/ragleap-core
examples/02_test_channel_directly.py
.py
7bac006ef5de7099
7.95
7
""" Audit logging for ragleap-graph (v0.6.6+), backed by Postgres. Fully optional. If GraphIndex is constructed without an AuditConfig (or with one that has no database_url set), no audit logging occurs and no psycopg2 import is even attempted - this module has zero effect unless explicitly configured. Graceful-degra...
antonyrag/ragleap-core
packages/ragleap-graph/src/ragleap_graph/_audit.py
.py
30dd11f287fa1f9d
7.45
7
#!/usr/bin/env python3 """ Seqcore Usage Examples - Real-World Scenarios Author: Dr. Pritam Kumar Panda @ Stanford University Run these examples to see seqcore in action with real biological data. """ import seqcore as sc import numpy as np def example_1_sequence_analysis(): """Example 1: Analyze DNA sequences ...
pritampanda15/Seqcore
examples/usage_examples.py
.py
2458f1d21b8fd369
7.5
9
"""Sequence alignment and pattern matching operations. Provides pairwise alignment, pattern searching, and motif finding. """ from __future__ import annotations import re from dataclasses import dataclass from typing import TYPE_CHECKING, Any import numpy as np if TYPE_CHECKING: from seqcore.core.arrays import...
pritampanda15/Seqcore
seqcore/alignment/__init__.py
.py
e466bd4017bc9bf1
7.5
9
"""Core biological sequence array implementations. Provides efficient, memory-optimized storage for biological sequences with optional GPU acceleration. Optimized for high-performance batch operations using NumPy vectorization. """ from __future__ import annotations from abc import ABC from collections.abc import I...
pritampanda15/Seqcore
seqcore/core/arrays.py
.py
8d08e1dcc7e950c6
7.5
9
"""Device management for GPU acceleration and performance controls. Provides a unified interface for CPU/GPU device selection and memory management. """ from __future__ import annotations import time import warnings from contextlib import contextmanager from typing import Any # Global state _current_device = "cpu" ...
pritampanda15/Seqcore
seqcore/core/device.py
.py
114c9d8232e79f1d
7.5
9
"""K-mer extraction and analysis operations. Provides efficient k-mer counting, extraction, and spectrum analysis. """ from __future__ import annotations from collections import Counter from typing import TYPE_CHECKING import numpy as np if TYPE_CHECKING: from seqcore.core.arrays import BioArray def extract_...
pritampanda15/Seqcore
seqcore/core/kmers.py
.py
b1d2e862c8590bf7
7.5
9
"""Structure array for 3D molecular structures. Provides efficient storage and operations for protein/nucleic acid structures. """ from __future__ import annotations from dataclasses import dataclass from typing import TYPE_CHECKING import numpy as np if TYPE_CHECKING: import pandas as pd @dataclass class At...
pritampanda15/Seqcore
seqcore/core/structure.py
.py
095b8c9cd9870a5e
7.5
9
"""Phylogenetic tree construction and analysis. Provides distance-based tree building methods and tree operations. """ from __future__ import annotations from dataclasses import dataclass, field from typing import TYPE_CHECKING, Any import numpy as np if TYPE_CHECKING: from seqcore.core.arrays import BioArray ...
pritampanda15/Seqcore
seqcore/phylogenetics/__init__.py
.py
5d9b39bfc997d609
7.5
9
"""Structural biology operations. Provides operations for protein/nucleic acid 3D structures including distance calculations, contact analysis, RMSD, and more. """ from __future__ import annotations from dataclasses import dataclass from typing import TYPE_CHECKING, Any import numpy as np if TYPE_CHECKING: fro...
pritampanda15/Seqcore
seqcore/structure/__init__.py
.py
9c3fd564990f44be
7.5
9
"""Tests for core seqcore functionality.""" import numpy as np import pytest class TestDNAArray: """Tests for DNAArray class.""" def test_single_sequence(self): """Test creating DNAArray from single sequence.""" from seqcore.core.arrays import DNAArray dna = DNAArray("ACGT") ...
pritampanda15/Seqcore
tests/test_core.py
.py
13987a9aa0e3b71e
7
9
"""Comprehensive tests for all supported file formats. Author: Dr. Pritam Kumar Panda @ Stanford University """ import os import numpy as np import pytest # Get the test data directory TEST_DATA_DIR = os.path.join(os.path.dirname(__file__), "data") class TestSequenceFormats: """Tests for sequence file formats...
pritampanda15/Seqcore
tests/test_file_formats.py
.py
7382ed4b496f637a
7
9
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Tests for the mechanical documentation checks. Each check is exercised against a synthetic tree rather than the repository, so a test pins the rule instead of the state of today's docs. The repository is covered by one end-to-end case: the hook...
fractalyze/zorch
tools/testing/lint_docs_test.py
.py
e9407f2824254fe1
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Byte-oriented Fiat-Shamir transcript: the `ByteTranscript` seam and a Merlin-over-hash duplex (`ByteHashTranscript`) parameterized by a `ByteHash`. This is the HOST-side, byte-oriented sibling of the device-resident algebraic `DuplexTranscript`...
fractalyze/zorch
zorch/byte_transcript.py
.py
ee47016e1342aa2a
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Shared Fiat-Shamir challenge-field policy.""" from __future__ import annotations from dataclasses import dataclass from typing import Any import frx.numpy as fnp from frx import Array from zk_dtypes import efinfo from zorch.transcript import...
fractalyze/zorch
zorch/challenge.py
.py
272563fa50925d77
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Reed-Solomon over a binary field's additive-NTT domain; implements FoldableCode. A binary field has no power-of-two multiplicative subgroup (the unit group's order is odd) and `x == -x` in characteristic 2, so `ReedSolomon`'s `(x, -x)`-conjugat...
fractalyze/zorch
zorch/coding/additive_reed_solomon.py
.py
fb8bb110d708f10e
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The FoldableCode seam: a LinearCode whose codewords fold round-by-round. BaseFold-style IOPPs (FRI, BaseFold) need more than linearity from their code. Layer `level` (length `block_len >> level`; layer 0 is the fresh codeword) pairs entries `(j...
fractalyze/zorch
zorch/coding/foldable_code.py
.py
830f0051ad0cbaa4
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The LinearCode seam every encoding builds on. A linear code maps a length-`message_len` message to a length-`block_len` codeword over a single field dtype (`block_len > message_len`; the rate is `message_len / block_len`). `encode` acts on the ...
fractalyze/zorch
zorch/coding/linear_code.py
.py
76023ce9aec8124d
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The TensorCode seam: a LinearCode whose codeword coordinates are *point evaluations* of the committed message's multilinear extension. Ligerito recurses by reading each proximity right-hand side `<G[s], w> = encode(w)[s]` as an eval-claim `ŵ(p_...
fractalyze/zorch
zorch/coding/tensor_code.py
.py
f2ec61347ba74ac1
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Reed-Solomon encode and the FRI fold: checked against independent oracles. The oracle never reuses the encoder. The NTT evaluation domain is recovered straight from `lax.ntt` of an impulse (NTT(e_1)_j = w^j), then the codeword is compared to a ...
fractalyze/zorch
zorch/coding/testing/reed_solomon_test.py
.py
57d2c19c52998bbc
7.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The TensorCode seam, checked against an independent multilinear oracle. The oracle shares no code with `eval_point`: it converts the message coefficients to the hypercube-evaluation basis (`mle_coeffs_to_evals`) and evaluates that MLE (`eval_ml...
fractalyze/zorch
zorch/coding/testing/tensor_code_test.py
.py
d71c13578bc54046
8.06
12
"""Module-lattice commitments beside the Merkle tree — Ajtai, and BDLOP for hiding. The lattice counterpart of `merkle`: scheme-agnostic, with no domain separator, no transcript, and no challenge — those belong to the consumer's commitment scheme. What lives here is the algebra and the opening predicate: - **Ajtai**:...
fractalyze/zorch
zorch/commit/ajtai.py
.py
0d4266783a9ddc8c
7.56
12
"""Layer-by-layer k-ary Merkle commitment — scheme-agnostic, on Sponge + Compression. `commit` hashes each matrix row to a leaf digest (Sponge), then folds sibling groups per layer (Compression, whose `arity` sets the tree's) down to a single root, returning `(raw_root, digest_layers)` (leaf digests first, root last)....
fractalyze/zorch
zorch/commit/merkle.py
.py
ea8b5621b1cfabab
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """SP1 single-matrix commitment (SMCS) over zorch's Merkle blocks. Semantically SP1's ``CudaTcsProver::commit_tensors`` for one matrix of power-of-two height: hash each row to a leaf, fold sibling pairs to a Merkle root (zorch's ``MerkleTree`` ove...
fractalyze/zorch
zorch/commit/smcs.py
.py
6f222ddd6879a6d7
7.56
12
"""Query-strided binary Merkle commitment — scheme-agnostic, on Sponge + Compression. A plain Merkle tree pairs *adjacent* leaves (0-1, 2-3, …), so authenticating a leaf opens one row and one sibling path. Some PCS query phases instead open a whole *coset* of ``rows_per_query`` rows at once, and those rows sit a fixed...
fractalyze/zorch
zorch/commit/strided_merkle.py
.py
f24173ca0bfb134b
7.56
12
"""Ajtai/BDLOP commit — roundtrip, homomorphism, and the opening predicate. Structural correctness without goldens, like `merkle_test`: a valid opening re-commits to the committed value, the additive homomorphism (the folding prerequisite) is checked as algebra, and an over-bound or substituted opening is rejected by ...
fractalyze/zorch
zorch/commit/testing/ajtai_test.py
.py
c77ec31a1ad24ce4
8.06
12
"""koalabear-16 Merkle stack fixture — TEST only. The Poseidon2(koalabear-16) -> Sponge -> Compression -> MerkleTree wiring used by merkle_test. Lives under commit/ rather than next to the permutation fixture in `zorch/testkit/koalabear16.py`, so that fixture stays free of a back-dependency on the commit layer. """ f...
fractalyze/zorch
zorch/commit/testing/koalabear16.py
.py
8388d0b19717b6c9
7.06
12
"""MerkleTree.commit — structure, digest coherence, and open->verify roundtrip. Correctness here is structural and does not need a golden vector: an independent reconstruction of the root from each leaf digest plus its sibling path (using the same compressor) must equal the committed root. Plonky3 merkle-root golden v...
fractalyze/zorch
zorch/commit/testing/merkle_test.py
.py
a584fd1e05938a0d
7.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """SMCS.commit — Merkle root + SP1 domain separator. The raw Merkle root and the domain-separated commitment are pinned to SP1's koalabear16 parameterization (the powers-of-two internal diagonal, distinct from the Plonky3 instance zorch's merkle_t...
fractalyze/zorch
zorch/commit/testing/smcs_test.py
.py
073b1193b7a8cd57
8.06
12
"""StridedMerkleTree.commit — plain-tree equivalence, strided structure, and an opened-rows -> root reconstruction roundtrip. Correctness is structural, no golden vector: rebuilding the root from one query's opened rows (re-hashed and folded through the strided levels) plus its sibling path must equal the committed ro...
fractalyze/zorch
zorch/commit/testing/strided_merkle_test.py
.py
cfb2e3038d8be3a0
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Proof-of-work grind search: the lowest counter whose candidate passes. One engine for every transcript's grind. Each `lax.while_loop` step tests a `window`-wide counter batch IN PARALLEL through the caller-supplied predicate and keeps the lowes...
fractalyze/zorch
zorch/grind.py
.py
cdf7a8e5c444cf77
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The LNP challenge space `C ⊆ S_κ^{σ₋₁}` over an injected byte stream. LNP proofs (eprint 2022/284, §2.7) draw challenges from C = { c ∈ S_κ : σ₋₁(c) = c, ²ᵏ√‖σ₋₁(cᵏ)·cᵏ‖₁ ≤ η }, polynomials of `Z[X]/(X^d + 1)` with coefficients in `[-κ, ...
fractalyze/zorch
zorch/lnp/challenge.py
.py
9edd44578d7736c1
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Π_many — ZK opening of an ABDLOP commitment with N linear relations. The first protocol layer of the LNP framework (eprint 2022/284, Fig. 4): prove knowledge of `(s1, s2)` opening `t_A = A1·s1 + A2·s2` — with the message implicitly `m = t_B − B...
fractalyze/zorch
zorch/lnp/opening.py
.py
7cef636ac606ff57
7.56
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """LNP challenge space — structure, the η gate, and the byte-stream contract. Structural correctness without goldens, like `ajtai_test`: every property the challenge space definition states (σ₋₁-invariance, the ℓ∞ bound, the operator-norm gate, in...
fractalyze/zorch
zorch/lnp/testing/challenge_test.py
.py
e5ea4592831d13b5
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Π_eval — completeness, the constant-coefficient claim, and its soundness. Structural correctness without goldens, like `opening_test`. The statement this layer proves is weaker than the one below it, and the tests are built around exactly that ...
fractalyze/zorch
zorch/lnp/testing/eval_test.py
.py
f3de609303b04cf5
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The exact ℓ2 bound (Fig. 10, eq. 53) — the two evaluations it builds, and the end-to-end proof they buy, at `E = I` and at a general affine image. The weight here is on the *algebra*, for the reason it is one layer down: a prove/verify round-tr...
fractalyze/zorch
zorch/lnp/testing/exact_test.py
.py
a13cf7811fa7bdfd
7.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """The LNP test parameter point, and the fixtures every protocol suite builds on it. One definition, because the numbers below are *derived* — from Lemma 2.14-1, from the Figure-3 challenge point, from a rejection budget — and a suite that copies ...
fractalyze/zorch
zorch/lnp/testing/lnp_fixture.py
.py
b568ca7c17acad4d
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Π_many opening — completeness, soundness rejections, and the FS chain. Structural correctness without goldens, like `ajtai_test`: an honest prover's proof verifies (the budgeted repeat loop included — at the test's repetition rates most proofs ...
fractalyze/zorch
zorch/lnp/testing/opening_test.py
.py
e435631069982f85
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Π_eval^(2) (Fig. 8) — completeness, soundness, the wire contract, and the lift layout the whole protocol is indexed against. Two honest statements are built here, and they are built differently on purpose. A *relation* vanishes as a ring elemen...
fractalyze/zorch
zorch/lnp/testing/quadratic_eval_test.py
.py
75c7defcf0f94471
8.06
12
# Copyright 2026 The Zorch Authors. SPDX-License-Identifier: Apache-2.0 """Π^(2) (Fig. 6) and Π_many^(2) (Fig. 7) — completeness, soundness, and the wire contract. The honest statement is built the only way a quadratic one can be without solving for a witness: pick `R2` and `r1` freely, then *define* `r0 := −(sᵀR2s + ...
fractalyze/zorch
zorch/lnp/testing/quadratic_test.py
.py
f32571ddcfeb5d80
8.06
12
import numpy as np from typing import Tuple, List, Union, Any import bittensor from numpy import ndarray, dtype, floating, complexfloating U32_MAX = 4294967295 U16_MAX = 65535 def normalize_max_weight(x: np.ndarray, limit: float = 0.1) -> np.ndarray: r"""Normalizes the numpy array x so that sum(x) = 1 and the ma...
leadpoet/leadpoet
Leadpoet/base/utils/weight_utils.py
.py
c5cfa9ef6ab646ca
7.54
11
# Copyright © 2025 Leadpoet import typing import bittensor as bt class LeadRequest(bt.Synapse): num_leads: int business_desc: str = "" industry: typing.Optional[str] = "" region: typing.Optional[str] = "" leads: typing.Optional[typing.List[dict]] = None def deserialize(self) -> typing....
leadpoet/leadpoet
Leadpoet/protocol.py
.py
136414f4f128d273
7.54
11
"""Compatibility helpers for Bittensor extrinsic responses.""" from __future__ import annotations from contextlib import contextmanager from dataclasses import dataclass from types import SimpleNamespace from typing import Any, Iterator @dataclass(frozen=True) class ExtrinsicOutcome: """Normalized result from a...
leadpoet/leadpoet
Leadpoet/utils/bittensor_sdk.py
.py
c76345a0572b25a9
7.54
11
""" Shared hashing utilities for the lead fulfillment commit-reveal scheme. Lives in Leadpoet/utils/ so both miners and the gateway import the same ``hash_lead()`` function — single source of truth for hash agreement. """ import json import hashlib HASH_SCHEMA_VERSION = 1 _JSON_NATIVE_TYPES = (str, int, float, bool...
leadpoet/leadpoet
Leadpoet/utils/hashing.py
.py
d8b6a77ee7ac74b0
7.54
11
""" Source Provenance Validation Module Validates lead sources against regulatory requirements: - Checks against restricted data broker denylist - Validates domain age (≥7 days) - Verifies URL reachability - Categorizes source types for compliance tracking """ import aiohttp import bittensor as bt import hashlib impo...
leadpoet/leadpoet
Leadpoet/utils/source_provenance.py
.py
8838ce3c911cb359
7.54
11
import os import time import requests import jwt as pyjwt from datetime import datetime, timedelta from typing import Optional import logging import bittensor as bt logger = logging.getLogger(__name__) class TokenManager: """ Manages JWT token lifecycle for Supabase authentication. Automatically refreshe...
leadpoet/leadpoet
Leadpoet/utils/token_manager.py
.py
9d9c2bbaea059286
7.54
11
""" Consensus calculation for distributed validator ranking. Implements weighted consensus where each validator's ranking is weighted by their trust value: S_lead = Σ(S_v * V_v) for all validators """ import bittensor as bt from typing import List, Dict, Tuple from collections import defaultdict from Leadpoet.utils.ut...
leadpoet/leadpoet
Leadpoet/validator/consensus.py
.py
8a7cac561f3f29dc
7.54
11
""" TEE Attestation Endpoint This endpoint returns the TEE attestation document, which provides cryptographic proof that the gateway is running the canonical code from GitHub. Anyone (miners, validators, auditors) can call this endpoint to verify: 1. The attestation signature (proves it came from AWS Nitro hardware) ...
leadpoet/leadpoet
gateway/api/attest.py
.py
a05eadda4babd33e
8.04
11
#!/usr/bin/env python3 """Remap comment[proteomics data acquisition method] to PRIDE:0000659 policy. Rules (see issue #31 / design 2026-08-10): - Wrong/foreign AC → map by recognized method to recommended PRIDE NT+AC - Valid known AC with casing/order drift → normalize to NT=<lowercase>;AC=<accession> - Plain free-tex...
bigbio/sdrf-annotated-datasets
scripts/remap_acquisition_method.py
.py
4e1f8c142a203ef8
7.5
9
import numpy as np import torch import torchaudio import yaml from utils.commons.dwt import separate_signal_dwt from utils.commons.shc import subharmonics_correction from .base import BaseSVC class Svc(BaseSVC): def __init__( self, svc_model_path=None, svc_config_path=None...
castlechoi/VibE-SVC2
inference/infer_vibesvcII.py
.py
7aba89972f8e5c6f
7.62
16
import numpy as np class F0Predictor(object): def compute_f0(self, wav, p_len): """ input: wav:[signal_length] p_len:int output: f0:[signal_length//hop_length] """ pass def compute_f0_uv(self, wav, p_len): """ input: wav:[signal_length] ...
castlechoi/VibE-SVC2
modules/F0Predictor/F0Predictor.py
.py
48a0a45109ed4692
7.62
16
from typing import Optional, Union try: from typing import Literal except Exception: from typing_extensions import Literal import numpy as np import torch import torchcrepe from torch import nn from torch.nn import functional as F #from:https://github.com/fishaudio/fish-diffusion def repeat_expand( conte...
castlechoi/VibE-SVC2
modules/F0Predictor/crepe.py
.py
ea2517c28a4f25a0
7.62
16
import sys from functools import reduce import librosa import numpy as np import torch from torch.nn.modules.module import _addindent from .constants import * # noqa: F403 def cycle(iterable): while True: for item in iterable: yield item def summary(model, file=sys.stdout): def repr(m...
castlechoi/VibE-SVC2
modules/F0Predictor/rmvpe/utils.py
.py
f1c5b8b755781020
7.62
16
import math import torch from torch import nn from torch.nn import functional as F import modules.commons as commons from modules.DSConv import weight_norm_modules from modules.modules import LayerNorm class FFT(nn.Module): def __init__(self, hidden_channels, filter_channels, n_heads, n_layers=1, kernel_size=1, p...
castlechoi/VibE-SVC2
modules/attentions.py
.py
66a0af869c21a412
7.62
16
import torch def feature_loss(fmap_r, fmap_g): loss = 0 for dr, dg in zip(fmap_r, fmap_g): for rl, gl in zip(dr, dg): rl = rl.float().detach() gl = gl.float() loss += torch.mean(torch.abs(rl - gl)) return loss * 2 def discriminator_loss(disc_real_outputs, disc_generated_outputs): los...
castlechoi/VibE-SVC2
modules/losses.py
.py
947d96e530127f2c
7.62
16
#!/usr/bin/env python3 """Validate the component inventory and its coverage of tracked files.""" from __future__ import annotations import argparse import json import subprocess import sys from pathlib import Path ROOT = Path(__file__).resolve().parents[1] MANIFEST_PATH = ROOT / "components" / "manifest.json" sys.p...
conorbronsdon/agent-context-os
scripts/component-manifests.py
.py
6e68ba51ff2c054b
7.6
15
"""Loading and provenance helpers for the Minari-derived ``.npz`` datasets baseline orchestrators train on, shared across baseline orchestrators. """ from __future__ import annotations import hashlib import json from pathlib import Path import subprocess import numpy as np DEFAULT_DATASET_KEYS = ( "observations"...
JaimeParker/rl-garden
baselines/core/dataset.py
.py
bb0201fab7d37596
7.48
8
"""Reader for ``baselines/baselines.yaml``. The manifest is the single source of truth for what's registered as a runnable baseline (git submodule under ``3rd_party/`` + a dedicated venv + an ``baselines/<name>/`` orchestrator) versus a read-only reference clone. ``scripts/install_baseline.sh`` reads it through this m...
JaimeParker/rl-garden
baselines/core/manifest.py
.py
7326260b0c23241c
7.48
8
"""ACRLPD: Q-chunking's action-chunked RLPD (``3rd_party/qc/agents/acrlpd.py``). Extends rl-garden's existing ``RLPD`` (itself a thin ``SAC`` extension -- ensemble critics, LayerNorm, high UTD, ``PriorDataReplayMixin``'s ratio-mixed offline/online sampling from step 0) with action chunking: ``horizon_length`` consecut...
JaimeParker/rl-garden
rl_garden/algorithms/acrlpd.py
.py
6dc29390d3ba0080
7.48
8
"""Base algorithm: seeding / logger / device / checkpoint I/O. Minimal counterpart to SB3's ``BaseAlgorithm``. We deliberately keep this thin so GPU-parallel specifics live in ``OffPolicyAlgorithm``. """ from __future__ import annotations from abc import ABC, abstractmethod from collections import defaultdict from pa...
JaimeParker/rl-garden
rl_garden/algorithms/base_algorithm.py
.py
9624ad6bb101ee1f
7.48
8
"""Cal-QL algorithm layer. Cal-QL extends CQL by lower-bounding OOD Q-values with Monte Carlo returns from the replay sample. The rest of the SAC/REDQ/CQL update path is inherited from ``CQL``. """ from __future__ import annotations import warnings from typing import Any, Literal, Optional import torch import torch....
JaimeParker/rl-garden
rl_garden/algorithms/calql.py
.py
3652763830c0160d
7.48
8
"""DAgger (Ross et al. 2011): interactive imitation learning via a beta-mixed scripted-expert rollout that always labels the aggregated dataset with the expert's action, alternated with BC-style retraining. Composes two existing, unmodified pieces rather than introducing new machinery: ``BC`` (``algorithms/bc.py``, po...
JaimeParker/rl-garden
rl_garden/algorithms/dagger.py
.py
1f915d0d5a9e8c85
7.48
8
"""EDAC: SAC-N (offline SAC, large critic ensemble) plus a gradient- diversity penalty across the ensemble. Ported from ``3rd_party/CORL/algorithms/offline/edac.py`` (arXiv 2110.01548). ``OfflineSAC`` (``rl_garden/algorithms/offline_sac.py``) already implements everything EDAC needs except the diversity term: full cri...
JaimeParker/rl-garden
rl_garden/algorithms/edac.py
.py
d0f60406c9c08abb
7.48
8
"""Generic offline-to-online (off2on) transition machinery. ``Off2OnReplayMixin`` owns every offline->online transition mechanic that is generic across off2on algorithm families (Cal-QL, IQL, ...): offline/online replay bookkeeping (``switch_to_online_mode``, mixed-batch sampling with a fixed or adaptive ``offline_dat...
JaimeParker/rl-garden
rl_garden/algorithms/off2on.py
.py
14c612a100e5b88c
7.48
8
"""Original Cal-QL paper's own offline-to-online design, built on the shared shell. ``Off2OnCalQL`` adds no behavior on top of ``_CalQLRolloutTrainingShell`` beyond a construction-time preset: no warmup by default (unlike ``WSRL``), offline data retained and mixed throughout online fine-tuning by default, and the CQL/...
JaimeParker/rl-garden
rl_garden/algorithms/off2on_calql.py
.py
28c3c7b55a8b8282
7.48
8
from __future__ import annotations from typing import Any from core.storage.base import BaseVectorStorage from core.utils import logger class VectorWriteBatch: """Coalesce vector mutations and apply the final state in one write. Merge logic can update the same entity several times while processing a do...
Preciso-GR/preciso-graphrag
core/storage/vector_write_batch.py
.py
6ccac03a77722c48
7.42
6
from __future__ import annotations from config import SUMMARY_MARKER # Reason surfaced on the merge return tuple (and mirrored into the pending-summary # queue record) when compression is needed. Preciso never compresses descriptions # itself — see preciso_mcp/tools/pending_summaries_tool.py, which is the only path ...
Preciso-GR/preciso-graphrag
core/summary.py
.py
4735ad3d9dd70bbc
7.42
6
""" reconciler.py Reconciles multiple subagent extraction outputs into a single unified extraction dict ready for ingestion. Called by ingest_with_reconciliation MCP tool. """ import re import time def _normalize(name: str) -> str: """ Normalize entity name for comparison. Uppercase, remove punctuation...
Preciso-GR/preciso-graphrag
ingest/reconciler.py
.py
c4b83abd10ce4194
7.42
6
from __future__ import annotations from collections.abc import Mapping, Sequence from config import GRAPH_FIELD_SEP def namespace_source_id( source_id: str, document_id: str, chunk_part_ids: Mapping[str, Sequence[str]] | None = None, ) -> str: """Prefix each chunk id referenced by `source_id` with `...
Preciso-GR/preciso-graphrag
ingest/transformer.py
.py
e209ba6691020776
7.42
6
""" reconcile_tool.py MCP tool: ingest_with_reconciliation Called by orchestrator agent after all subagents finish. Reads multiple extraction files, reconciles them into one unified extraction, writes unified file to disk, then runs the ingestion pipeline. """ import json import re import time from pathlib import Pa...
Preciso-GR/preciso-graphrag
preciso_mcp/tools/reconcile_tool.py
.py
ddcfef1806f07f25
7.42
6
"""Deterministic offline stubs shared by the pytest suite. Mirrors the stubs used by test/summary_merge_manual.py and test/marker_leak_manual.py so no Ollama/network is ever required. """ from __future__ import annotations import json from config import GRAPH_FIELD_SEP, SOURCE_IDS_LIMIT_METHOD_KEEP, SUMMARY_MARKER ...
Preciso-GR/preciso-graphrag
tests/_stubs.py
.py
f4505ff9f7d6964a
7.92
6
"""Domain heuristics moved from core/query.py into config.py (Tier 4): comparison-query detection reads its phrase list from global_config, and the rag_response persona is configurable with a domain-neutral default.""" from __future__ import annotations from config import DEFAULT_YOY_SIGNAL_PHRASES, PROMPTS, build_gl...
Preciso-GR/preciso-graphrag
tests/test_query_heuristics.py
.py
84e01463992eaa75
7.92
6
"""Binary sensor platform for xComfort integration with Home Assistant.""" import logging from homeassistant.components.binary_sensor import ( BinarySensorDeviceClass, BinarySensorEntity, ) from homeassistant.config_entries import ConfigEntry from homeassistant.core import HomeAssistant from homeassistant.hel...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/binary_sensor.py
.py
12a95ba6f37de0b4
7.63
17
"""Support for xComfort Bridge cover shades.""" import logging from homeassistant.components.cover import ( ATTR_POSITION, CoverDeviceClass, CoverEntity, CoverEntityFeature, ) from homeassistant.config_entries import ConfigEntry from homeassistant.core import HomeAssistant from homeassistant.helpers.e...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/cover.py
.py
f1f2c108196af379
7.63
17
"""Helpers for Home Assistant device registry metadata.""" from __future__ import annotations from typing import TYPE_CHECKING from homeassistant.helpers.device_registry import DeviceInfo from .const import DOMAIN from .xcomfort.constants import ComponentTypes if TYPE_CHECKING: from .hub import XComfortHub ...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/device_info.py
.py
b8b90a2bde246e33
7.63
17
"""Helpers for safe Rx subscriptions on Home Assistant entities.""" from __future__ import annotations from collections import defaultdict import logging from typing import Any from homeassistant.helpers.entity import Entity _LOGGER = logging.getLogger(__name__) _DROP_LOG_COUNTS = {1, 2, 3, 5, 10, 25, 50, 100} d...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/entity_lifecycle.py
.py
388624b4761e0d2f
7.63
17
"""Support for xComfort buttons.""" import asyncio import logging from homeassistant.components.event import EventDeviceClass, EventEntity from homeassistant.config_entries import ConfigEntry from homeassistant.core import HomeAssistant, callback from homeassistant.helpers.device_registry import DeviceInfo from homea...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/event.py
.py
1dcb5c47390b3029
7.63
17
"""Class used to communicate with xComfort bridge.""" from __future__ import annotations import asyncio import logging from homeassistant.config_entries import ConfigEntry from homeassistant.core import HomeAssistant from .const import DOMAIN from .xcomfort.bridge import Bridge _LOGGER = logging.getLogger(__name__...
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/hub.py
.py
dc9d713e1604fa78
7.63
17
"""Support for xComfort lights.""" from functools import cached_property import logging from math import ceil from homeassistant.components.light import ATTR_BRIGHTNESS, ColorMode, LightEntity from homeassistant.config_entries import ConfigEntry from homeassistant.core import HomeAssistant from homeassistant.helpers....
javydekoning/ha-xcomfort-bridge
custom_components/xcomfort_bridge/light.py
.py
e580b5a9ba31cd36
7.63
17