o

inference.py

@@ -3,7 +3,7 @@ import json
 from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, AutoModel, AutoConfig
 from peft import LoraConfig, get_peft_model, PeftModel
 from modeling_bilstm_crf import BERT_BiLSTM_CRF
-from seq2seq_inference import infer_t5_prompt
+from seq2seq_inference import infer_t5_bart
 from huggingface_hub import hf_hub_download
 
 # Define supported models and their adapter IDs
@@ -61,7 +61,6 @@ def infer_araberta(text):
     else:
         tokenizer, model = cached_models["Araberta"]
 
-
     device = next(model.parameters()).device
 
     inputs = tokenizer(text, return_tensors='pt', truncation=True, padding='max_length', max_length=128)
@@ -73,15 +72,18 @@ def infer_araberta(text):
         predicted_ids = outputs['logits'][0].cpu().tolist()
 
     tokens = tokenizer.convert_ids_to_tokens(input_ids[0].cpu())
-    # predicted_labels = [model.config.id2label.get(p, 'O') for p in predicted_ids]
     predicted_labels = [model.id2label.get(p, 'O') for p in predicted_ids]
 
-
     clean_tokens = [t for t in tokens if t not in tokenizer.all_special_tokens]
     clean_labels = [l for t, l in zip(tokens, predicted_labels) if t not in tokenizer.all_special_tokens]
 
+    # ✅ New: map short to full sentiment
+    sentiment_map = {
+        "POS": "positive",
+        "NEG": "negative",
+        "NEU": "neutral"
+    }
 
-    # Group by aspect span
     aspects = []
     current_tokens = []
     current_sentiment = None
@@ -91,7 +93,7 @@ def infer_araberta(text):
             if current_tokens:
                 aspects.append({
                     "aspect": " ".join(current_tokens).replace("##", ""),
-                    "sentiment": current_sentiment
+                    "sentiment": sentiment_map.get(current_sentiment, current_sentiment)
                 })
             current_tokens = [token]
             current_sentiment = label.split("-")[1]
@@ -101,7 +103,7 @@ def infer_araberta(text):
             if current_tokens:
                 aspects.append({
                     "aspect": " ".join(current_tokens).replace("##", ""),
-                    "sentiment": current_sentiment
+                    "sentiment": sentiment_map.get(current_sentiment, current_sentiment)
                 })
                 current_tokens = []
                 current_sentiment = None
@@ -109,7 +111,7 @@ def infer_araberta(text):
     if current_tokens:
         aspects.append({
             "aspect": " ".join(current_tokens).replace("##", ""),
-            "sentiment": current_sentiment
+            "sentiment": sentiment_map.get(current_sentiment, current_sentiment)
         })
 
     token_predictions = [
@@ -125,6 +127,7 @@ def infer_araberta(text):
 
 
 
+
 def load_model(model_key):
     if model_key in cached_models:
         return cached_models[model_key]
@@ -148,7 +151,7 @@ def predict_absa(text, model_choice):
 
     if model_choice in ['mT5', 'mBART']:
         tokenizer, model = load_model(model_choice)
-        decoded = infer_t5_prompt(text, tokenizer, model)
+        decoded = infer_t5_bart(text, tokenizer, model)
 
     elif model_choice == 'Araberta':