fix: fix dataset path calculation

This commit is contained in:
Christian Medina
2026-06-30 19:01:39 -04:00
parent 14a302dc5e
commit bbdbe91871

View File

@@ -61,10 +61,10 @@ def train(config_path):
from datasets import load_dataset
import os
# Get dataset paths relative to script location
script_dir = Path(__file__).parent.parent
train_path = str(script_dir / ".." / ".." / "training" / "data" / "train.jsonl")
test_path = str(script_dir / ".." / ".." / "training" / "data" / "test.jsonl")
# Get dataset paths - dataset is in training/data/ relative to repo root
repo_root = Path(__file__).parent.parent.parent
train_path = str(repo_root / "training" / "data" / "train.jsonl")
test_path = str(repo_root / "training" / "data" / "test.jsonl")
dataset = load_dataset(
"json",