Spaces:
Running
on
Zero
Running
on
Zero
module for storing data
Browse files- data_store.py +53 -0
data_store.py
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
from pathlib import Path
|
| 3 |
+
import uuid
|
| 4 |
+
import json
|
| 5 |
+
from huggingface_hub import HfApi, HfFileSystem
|
| 6 |
+
|
| 7 |
+
DATASET_REPO = "EarthSpeciesProject/naturelm-audio-space-logs"
|
| 8 |
+
SPLIT = "test"
|
| 9 |
+
TESTING = os.getenv("TESTING", "0") == "1"
|
| 10 |
+
api = HfApi()
|
| 11 |
+
# Upload audio
|
| 12 |
+
# check if file exists
|
| 13 |
+
hf_fs = HfFileSystem()
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
def upload_data(audio: str | Path, user_text: str, model_response: str):
|
| 17 |
+
data_id = str(uuid.uuid4())
|
| 18 |
+
if TESTING:
|
| 19 |
+
data_id = "test-" + data_id
|
| 20 |
+
# Audio path in repo
|
| 21 |
+
suffix = Path(audio).suffix
|
| 22 |
+
audio_p = f"{SPLIT}/audio/" + data_id + suffix
|
| 23 |
+
|
| 24 |
+
api.upload_file(
|
| 25 |
+
path_or_fileobj=str(audio),
|
| 26 |
+
path_in_repo=audio_p,
|
| 27 |
+
repo_id=DATASET_REPO,
|
| 28 |
+
repo_type="dataset",
|
| 29 |
+
)
|
| 30 |
+
|
| 31 |
+
text = {
|
| 32 |
+
"user_message": user_text,
|
| 33 |
+
"model_response": model_response,
|
| 34 |
+
"file_name": audio_p,
|
| 35 |
+
"original_file_name": os.path.basename(audio),
|
| 36 |
+
"id": data_id,
|
| 37 |
+
}
|
| 38 |
+
|
| 39 |
+
# Append to a jsonl file in the repo
|
| 40 |
+
# APPEND DOESNT WORK, have to open first
|
| 41 |
+
if hf_fs.exists(f"datasets/{DATASET_REPO}/{SPLIT}/metadata.jsonl"):
|
| 42 |
+
with hf_fs.open(f"datasets/{DATASET_REPO}/{SPLIT}/metadata.jsonl", "r") as f:
|
| 43 |
+
lines = f.readlines()
|
| 44 |
+
lines.append(json.dumps(text) + "\n")
|
| 45 |
+
with hf_fs.open(f"datasets/{DATASET_REPO}/{SPLIT}/metadata.jsonl", "w") as f:
|
| 46 |
+
f.writelines(lines)
|
| 47 |
+
else:
|
| 48 |
+
with hf_fs.open(f"datasets/{DATASET_REPO}/{SPLIT}/metadata.jsonl", "w") as f:
|
| 49 |
+
f.write(json.dumps(text) + "\n")
|
| 50 |
+
|
| 51 |
+
# Write a separate file instead
|
| 52 |
+
# with hf_fs.open(f"datasets/{DATASET_REPO}/{data_id}.json", "w") as f:
|
| 53 |
+
# json.dump(text, f)
|