0ce879aa44
- Implemented HTML pages for datasets, models, training, testing, and results. - Created API endpoints for managing repositories, results, tests, and training sessions. - Added functionality for streaming training progress via Server-Sent Events (SSE). - Introduced a Dockerfile for the ML runner with necessary dependencies. - Developed an SDK for user code execution within the runner container. - Enhanced CSS styles for improved UI layout and navigation. - Established a layout template for consistent HTML structure across pages. - Added JavaScript for dynamic interactions on the models page. - Implemented WebSocket handling for real-time communication with kiosk devices and controllers. - Implemented model registration and management API at /api/models - Added Gitea proxy API for repository interactions at /api/repos - Created results API for listing and comparing training results at /api/results - Developed training management API for enqueueing and retrieving training jobs at /api/trainings - Introduced SSE endpoint for live training progress updates - Added HTML pages for models, datasets, and training management - Created a Dockerfile for the ML runner with necessary dependencies - Developed SDK for user code execution within the runner container - Enhanced CSS styles for improved UI/UX - Implemented WebSocket communication for real-time device and controller interactions in the kiosk system
81 lines
2.2 KiB
Python
81 lines
2.2 KiB
Python
"""meb_ml — SDK importabile dal codice utente dentro il container runner.
|
|
|
|
API:
|
|
from meb_ml import emit_metric, emit_series, emit_matrix, emit_log, save_artifact
|
|
|
|
emit_metric(iter=10, loss=0.23)
|
|
emit_series("roc_curve", x=fpr, y=tpr, kind="line")
|
|
emit_matrix("confusion", labels=[...], values=[[...],[...]])
|
|
emit_log("info", "epoch done")
|
|
|
|
Scrive righe JSON su stdout; il parent (ml-service) le inoltra su Redis/Influx.
|
|
Per risultati finali scrivere `out/metrics.json` con:
|
|
{"metrics": {...}, "plots": {"loss_curve": {"x": [...], "y": [...]}, ...}}
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
import os
|
|
import sys
|
|
from pathlib import Path
|
|
from typing import Any, Iterable, Sequence
|
|
|
|
|
|
def _print(obj: dict) -> None:
|
|
sys.stdout.write(json.dumps(obj, default=float) + "\n")
|
|
sys.stdout.flush()
|
|
|
|
|
|
def emit_metric(**fields: Any) -> None:
|
|
_print({"type": "metric", **fields})
|
|
|
|
|
|
def emit_series(name: str, x: Sequence, y: Sequence, kind: str = "line") -> None:
|
|
_print({
|
|
"type": "series",
|
|
"name": name,
|
|
"kind": kind,
|
|
"x": list(x),
|
|
"y": list(y),
|
|
})
|
|
|
|
|
|
def emit_matrix(name: str, labels: Sequence, values: Sequence[Sequence]) -> None:
|
|
_print({
|
|
"type": "matrix",
|
|
"name": name,
|
|
"labels": list(labels),
|
|
"values": [list(row) for row in values],
|
|
})
|
|
|
|
|
|
def emit_log(level: str, message: str) -> None:
|
|
_print({"type": "log", "level": level, "message": message})
|
|
|
|
|
|
def save_artifact(path: str) -> str:
|
|
"""Copia `path` nella cartella artefatti (MEB_ARTIFACTS_DIR). Ritorna la dest."""
|
|
dest_dir = Path(os.environ.get("MEB_ARTIFACTS_DIR", "/workdir/out"))
|
|
dest_dir.mkdir(parents=True, exist_ok=True)
|
|
src = Path(path)
|
|
dest = dest_dir / src.name
|
|
dest.write_bytes(src.read_bytes())
|
|
return str(dest)
|
|
|
|
|
|
def dataset_path() -> str:
|
|
return os.environ["MEB_DATASET_PATH"]
|
|
|
|
|
|
def artifacts_dir() -> str:
|
|
return os.environ.get("MEB_ARTIFACTS_DIR", "/workdir/out")
|
|
|
|
|
|
def read_test_input() -> dict:
|
|
"""Legge un singolo JSON da stdin (per script di test)."""
|
|
return json.loads(sys.stdin.readline())
|
|
|
|
|
|
def write_test_output(outputs: dict) -> None:
|
|
_print({"type": "result", "outputs": outputs})
|