Llama-3-1-8B-Uzbek / app /dataset.py
futzone
json files is done
311ac7e
import json
import os
source_folder = "datasets"
all_data = []
for filename in os.listdir(source_folder):
try:
if filename.endswith(".json"):
path = os.path.join(source_folder, filename)
with open(path, "r", encoding="utf-8") as f:
data = json.load(f)
for item in data:
try:
all_data.append({
"instruction": "Foydalanuvchining savoliga javob bering.",
"input": item["input"],
"output": item["output"]
})
except KeyError as e:
print(f"❌ {filename} faylida xatolik: {e}")
except json.JSONDecodeError as e:
print(f"❌ {filename} faylini o'qishda xatolik: {e}")
with open(os.path.join(source_folder, "train.json"), "w", encoding="utf-8") as f:
json.dump(all_data, f, ensure_ascii=False, indent=4)
print(f"βœ… {len(all_data)} ta yozuv muvaffaqiyatli train.json fayliga saqlandi.")