update app.py

app.py

@@ -79,14 +79,14 @@ LABEL = [
 veracity_tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
 bert_model = BertForSequenceClassification.from_pretrained("bert-base-uncased", num_labels=4, problem_type="single_label_classification")
 veracity_checkpoint_path = os.getcwd() + "/averitec/pretrained_models/bert_veracity.ckpt"
-veracity_model = SequenceClassificationModule.load_from_checkpoint(veracity_checkpoint_path,tokenizer=veracity_tokenizer, model=bert_model)
+veracity_model = SequenceClassificationModule.load_from_checkpoint(veracity_checkpoint_path,tokenizer=veracity_tokenizer, model=bert_model).to('cuda')
 # veracity_model = SequenceClassificationModule.load_from_checkpoint(veracity_checkpoint_path,tokenizer=veracity_tokenizer, model=bert_model).to(device)
 
 # Justification
 justification_tokenizer = BartTokenizer.from_pretrained('facebook/bart-large', add_prefix_space=True)
 bart_model = BartForConditionalGeneration.from_pretrained("facebook/bart-large")
 best_checkpoint = os.getcwd()+ '/averitec/pretrained_models/bart_justifications_verdict-epoch=13-val_loss=2.03-val_meteor=0.28.ckpt'
-justification_model = JustificationGenerationModule.load_from_checkpoint(best_checkpoint, tokenizer=justification_tokenizer, model=bart_model)
+justification_model = JustificationGenerationModule.load_from_checkpoint(best_checkpoint, tokenizer=justification_tokenizer, model=bart_model).to('cuda')
 # justification_model = JustificationGenerationModule.load_from_checkpoint(best_checkpoint, tokenizer=justification_tokenizer, model=bart_model).to(device)