Spaces:
Sleeping
Sleeping
Tracy André
commited on
Commit
·
d988d52
1
Parent(s):
6a2a2f3
updated
Browse files- data_loader.py +35 -1
data_loader.py
CHANGED
|
@@ -2,9 +2,12 @@
|
|
| 2 |
Module de chargement des données depuis Hugging Face
|
| 3 |
"""
|
| 4 |
import os
|
|
|
|
| 5 |
import pandas as pd
|
| 6 |
from datasets import load_dataset
|
|
|
|
| 7 |
from huggingface_hub import HfApi, hf_hub_download
|
|
|
|
| 8 |
from config import HF_TOKEN, DATASET_ID, REQUIRED_COLUMNS, MESSAGES
|
| 9 |
|
| 10 |
|
|
@@ -30,7 +33,7 @@ class DataLoader:
|
|
| 30 |
DATASET_ID,
|
| 31 |
split="train",
|
| 32 |
token=HF_TOKEN,
|
| 33 |
-
trust_remote_code
|
| 34 |
)
|
| 35 |
print(f"📊 Dataset chargé: {len(dataset)} exemples")
|
| 36 |
|
|
@@ -59,8 +62,39 @@ class DataLoader:
|
|
| 59 |
self.df = None
|
| 60 |
|
| 61 |
except Exception as e:
|
|
|
|
| 62 |
print(f"❌ Erreur lors du chargement depuis Hugging Face: {str(e)}")
|
| 63 |
print(f"❌ Type d'erreur: {type(e).__name__}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
|
| 65 |
# 2) Fallback: récupérer directement les fichiers du repo
|
| 66 |
try:
|
|
|
|
| 2 |
Module de chargement des données depuis Hugging Face
|
| 3 |
"""
|
| 4 |
import os
|
| 5 |
+
import traceback
|
| 6 |
import pandas as pd
|
| 7 |
from datasets import load_dataset
|
| 8 |
+
import datasets as hf_datasets
|
| 9 |
from huggingface_hub import HfApi, hf_hub_download
|
| 10 |
+
import huggingface_hub as hf_hub
|
| 11 |
from config import HF_TOKEN, DATASET_ID, REQUIRED_COLUMNS, MESSAGES
|
| 12 |
|
| 13 |
|
|
|
|
| 33 |
DATASET_ID,
|
| 34 |
split="train",
|
| 35 |
token=HF_TOKEN,
|
| 36 |
+
# trust_remote_code n'est plus supporté; retiré pour éviter le warning
|
| 37 |
)
|
| 38 |
print(f"📊 Dataset chargé: {len(dataset)} exemples")
|
| 39 |
|
|
|
|
| 62 |
self.df = None
|
| 63 |
|
| 64 |
except Exception as e:
|
| 65 |
+
print("\n===== 🔎 Détails de l'erreur Hugging Face =====")
|
| 66 |
print(f"❌ Erreur lors du chargement depuis Hugging Face: {str(e)}")
|
| 67 |
print(f"❌ Type d'erreur: {type(e).__name__}")
|
| 68 |
+
# Détails de l'exception
|
| 69 |
+
try:
|
| 70 |
+
print(f"❖ repr(e): {repr(e)}")
|
| 71 |
+
if getattr(e, '__cause__', None) is not None:
|
| 72 |
+
print(f"❖ Cause: {repr(e.__cause__)}")
|
| 73 |
+
if getattr(e, '__context__', None) is not None:
|
| 74 |
+
print(f"❖ Contexte: {repr(e.__context__)}")
|
| 75 |
+
if getattr(e, 'args', None):
|
| 76 |
+
print(f"❖ Args: {e.args}")
|
| 77 |
+
except Exception:
|
| 78 |
+
pass
|
| 79 |
+
# Versions des libs pour diagnostic
|
| 80 |
+
try:
|
| 81 |
+
print(f"❖ datasets version: {getattr(hf_datasets, '__version__', 'unknown')}")
|
| 82 |
+
print(f"❖ huggingface_hub version: {getattr(hf_hub, '__version__', 'unknown')}")
|
| 83 |
+
except Exception:
|
| 84 |
+
pass
|
| 85 |
+
# Environnement réseau de base
|
| 86 |
+
try:
|
| 87 |
+
proxies = {k: v for k, v in os.environ.items() if k.lower().endswith('proxy')}
|
| 88 |
+
print(f"❖ Proxies détectés: {proxies if proxies else 'aucun'}")
|
| 89 |
+
except Exception:
|
| 90 |
+
pass
|
| 91 |
+
# Trace complète
|
| 92 |
+
try:
|
| 93 |
+
print("❖ Traceback complet:")
|
| 94 |
+
print(traceback.format_exc())
|
| 95 |
+
except Exception:
|
| 96 |
+
pass
|
| 97 |
+
print("===== 🔎 Fin des détails =====\n")
|
| 98 |
|
| 99 |
# 2) Fallback: récupérer directement les fichiers du repo
|
| 100 |
try:
|