danvics/haiku.rag
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

No zip

Browse source
This commit is contained in:
Yiorgis Gozadinos 2026-01-26 16:26:08 +02:00
parent 9b166969e2
commit 369b3a4bf7
No known key found for this signature in database
2 changed files with 12 additions and 28 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
docs/benchmarks.md
View file

@ -32,7 +32,7 @@ Available datasets:
| Dataset | Size | | Dataset | Size |
|---------|------| |---------|------|
| `repliqa` | ~18MB | | `repliqa` | ~30MB |
| `hotpotqa` | ~331MB | | `hotpotqa` | ~331MB |
| `wix` | ~511MB | | `wix` | ~511MB |
| `open_rag_bench` | ~14GB | | `open_rag_bench` | ~14GB |

38
evaluations/evaluations/benchmark.py
View file

@ -1,7 +1,5 @@
import asyncio import asyncio
import shutil import shutil
import tempfile
import zipfile
from collections.abc import Mapping from collections.abc import Mapping
from pathlib import Path from pathlib import Path
from typing import Any, cast from typing import Any, cast
@ -9,7 +7,7 @@ from typing import Any, cast
import logfire import logfire
import typer import typer
from dotenv import find_dotenv, load_dotenv from dotenv import find_dotenv, load_dotenv
from huggingface_hub import HfApi, hf_hub_download from huggingface_hub import HfApi, snapshot_download
from pydantic_evals import Case, Dataset as EvalDataset from pydantic_evals import Case, Dataset as EvalDataset
from pydantic_evals.evaluators import LLMJudge from pydantic_evals.evaluators import LLMJudge
from pydantic_evals.reporting import ReportCaseFailure from pydantic_evals.reporting import ReportCaseFailure
@ -482,13 +480,12 @@ def download(
continue continue
console.print(f"[blue]Downloading {spec.key}...[/blue]") console.print(f"[blue]Downloading {spec.key}...[/blue]")
zip_filename = f"{spec.db_filename}.zip"
try: try:
zip_path = hf_hub_download( downloaded_path = snapshot_download(
repo_id=HF_REPO_ID, repo_id=HF_REPO_ID,
filename=zip_filename,
repo_type="dataset", repo_type="dataset",
allow_patterns=f"{spec.db_filename}/*",
) )
except Exception as e: except Exception as e:
console.print(f"[red]Failed to download {spec.key}: {e}[/red]") console.print(f"[red]Failed to download {spec.key}: {e}[/red]")
@ -498,10 +495,9 @@ def download(
if db.exists(): if db.exists():
shutil.rmtree(db) shutil.rmtree(db)
# Extract to the parent directory # Copy from cache to target location
db.parent.mkdir(parents=True, exist_ok=True) db.parent.mkdir(parents=True, exist_ok=True)
with zipfile.ZipFile(zip_path, "r") as zf: shutil.copytree(Path(downloaded_path) / spec.db_filename, db)
zf.extractall(db.parent)
console.print(f"[green]Downloaded {spec.key} to {db}[/green]") console.print(f"[green]Downloaded {spec.key} to {db}[/green]")
@ -531,24 +527,12 @@ def upload(
continue continue
console.print(f"[blue]Uploading {spec.key}...[/blue]") console.print(f"[blue]Uploading {spec.key}...[/blue]")
zip_filename = f"{spec.db_filename}.zip" api.upload_folder(
folder_path=str(db),
with tempfile.TemporaryDirectory() as tmpdir: path_in_repo=spec.db_filename,
zip_path = Path(tmpdir) / zip_filename repo_id=HF_REPO_ID,
console.print("[dim]Creating zip archive...[/dim]") repo_type="dataset",
with zipfile.ZipFile(zip_path, "w", zipfile.ZIP_DEFLATED) as zf: )
for file in db.rglob("*"):
if file.is_file():
arcname = f"{spec.db_filename}/{file.relative_to(db)}"
zf.write(file, arcname)
console.print("[dim]Uploading to HuggingFace...[/dim]")
api.upload_file(
path_or_fileobj=str(zip_path),
path_in_repo=zip_filename,
repo_id=HF_REPO_ID,
repo_type="dataset",
)
console.print(f"[green]Uploaded {spec.key} to {HF_REPO_ID}[/green]") console.print(f"[green]Uploaded {spec.key} to {HF_REPO_ID}[/green]")