Skip to main content

Serigy

Serigy

Classificador local de fake news em PT-BR. Carrega pesos do Hub e faz predict. Também expõe primitivos System One: Escolha, Indice, Verdadeiro, Prioridade.

Marca: assets/brand/ · spec.

pip install serigy
from serigy import SerigyModel

model = SerigyModel("wendellperbar/serigy-jev")
out = model.predict(
    "Senado aprova lei que PROÍBE jogo de videogame para menores de 18 anos."
)
print(out["predicao"], out["score_final"], out["confianca_calibrada"])
serigy --texto "texto da notícia..."

Se o sensor for gated, configure HF_TOKEN.

Saída (predict = Prioridade)

Campo Significado
predicao Claramente Falso / Incerto/Misto / Claramente Verdadeiro
classe_ordinal 0, 1 ou 2
score_final veracidade 0.0…1.0
confianca_calibrada perto de 1 = mais concentrado
probabilidades massa em cada classe

Critérios usam type: "prioridade" (alias de input: ordinal_score).

System One (system_one)

from serigy import SerigyModel, Escolha, Indice, Verdadeiro, Prioridade

model = SerigyModel("wendellperbar/serigy-jev")
result = model.system_one(
    "texto a avaliar…",
    {
        "rota": Escolha(
            instructions="Qual ação?",
            criteria={"abaixar": None, "fazer_nada": None, "pular": None},
        ),
        "nivel": Indice(
            instructions="Quão grave?",
            criteria=["Baixa", "Média", "Alta"],
        ),
        "eh_fato": Verdadeiro(instructions="É verdadeira?"),
    },
)
print(result.answers["rota"].escolha, result.answers["eh_fato"].verdadeiro)
# JSON público: result.as_public_dict()
Jev Serigy O que responde
Choice escolha Qual opção (conjunto fechado)?
Score indice Em que ponto do espectro (K∈[2,10])?
Noul verdadeiro A proposição é verdadeira? (P(sim))
— prioridade Em qual de 3 patamares? (predict é atalho)

CLI System One: serigy --json '{"state":"…","questions":{…}}'.

Spec pública: .cursor/specs/sdk-primitivos/spec.md.

Pesos FakeNews: huggingface.co/wendellperbar/serigy-jev

Golpes BR (System One)

Heads multi-task para mensagens de golpe/legítimas em PT-BR (staging hf/golpes/ → Hub wendellperbar/serigy-golpes-br):

from serigy import SerigyModel, Escolha, Verdadeiro

model = SerigyModel("wendellperbar/serigy-golpes-br")
result = model.system_one(
    "Confirme seu CPF neste link para liberar o PIX…",
    {
        "acao": Escolha(
            instructions="Como tratar?",
            criteria={"bloquear": None, "revisar": None, "entregar": None},
        ),
        "eh_golpe": Verdadeiro(instructions="É golpe?"),
    },
)

Sync / upload: hf/README.md · card hf/golpes/README.md.

Extra: benchmark T-Rex (monorepo)

Comparativo Serigy × Jev × Laya no Chrome T-Rex Runner (serigy_trex):
src/serigy_trex/README.md · docs/RELATORIO_TREX.md · spec.

MIT — Copyright © 2026 Wendell Barreto.

Metadata

Release files for serigy 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for serigy 0.2.0
File Size Uploaded Provenance Actions
serigy-0.2.0.tar.gz 81.5 kB With provenance Details for serigy-0.2.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for serigy 0.2.0
File Interpreter ABI Platform Actions
serigy-0.2.0-py3-none-any.whl Python 3 none any Details for serigy-0.2.0-py3-none-any.whl

Total release size: 140.1 kB

Release files / serigy-0.2.0.tar.gz

Download URL serigy-0.2.0.tar.gz
Size 81.5 kB
Tags Source
SHA-256 checksum
How to use checksums
34215dc35e60121e111a91f0754638b08c0b737199e0a8d5b5c8cd36bfe5b1f4
BLAKE2b-256 checksum
How to use checksums
a77e380927aeb013cf7c1a6f2d7b03eb01c23f92b5c20e5ed9c9b1b526be3aa7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Release files / serigy-0.2.0-py3-none-any.whl

Download URL serigy-0.2.0-py3-none-any.whl
Size 58.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
dc2e7e91ac4520b92d03aaf047a0a65f4a074b7b0d5dec8e1d08619f94b2eccd
BLAKE2b-256 checksum
How to use checksums
60f55b10be9f26c5857e30e4498e340e1b751c7541dd9c920aeab29e65150243
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

Security for serigy 0.2.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for serigy-0.2.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 10, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page