from transformers import T5Tokenizer, T5ForConditionalGeneration

tokenizer = T5Tokenizer.from_pretrained("cedricgaudron/scanner-tickets")
model = T5ForConditionalGeneration.from_pretrained("cedricgaudron/scanner-tickets")

texte = """CARREFOUR
TOTAL TTC : 24,75€
Date : 12/06/2024
SIRET : 123 456 789 00012
TVA : FR 12 345678912"""

input_ids = tokenizer("Extrais les données suivantes en format JSON :\n" + texte, return_tensors="pt").input_ids
output = model.generate(input_ids, max_length=128)
print(tokenizer.decode(output[0], skip_special_tokens=True))