Instructions to use ssg1/doc-compass with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ssg1/doc-compass with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="ssg1/doc-compass")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ssg1/doc-compass", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 978 Bytes
967b9df 1b1ee73 967b9df 709e76a 967b9df 141ef99 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 | """Where local data lives. Everything under data/ except label and id files stays out of git."""
from pathlib import Path
# absolute(), not resolve(): the Hugging Face cache stores files as symlinks, and following
# them would point ROOT at the cache's blob folder instead of the folder holding checkpoints/.
ROOT = Path(__file__).absolute().parents[1]
PUBLIC = ROOT / "data" / "public" # downloaded and derived public data
MANUAL = ROOT / "data" / "manual" # our pool and labels
MODELS = ROOT / "checkpoints"
MEDIQ_POSTS = PUBLIC / "mediq_posts.csv"
PATIENT_COMMENTS = PUBLIC / "patient_comments.csv"
POOL_IDS = MANUAL / "pool.ids.csv" # committed: which posts we label, and who labels them
POOL = MANUAL / "pool.csv" # local only: the same posts with their text
DRAFT_LABELS = MANUAL / "draft.labels.csv" # first-pass labels shown in the labelling tool
EMERGENCY_CASES = ROOT / "data" / "synthetic" / "emergency_cases.csv" # written by us to test the emergency check
|