fixed sarcasm

backend/services.py

@@ -216,7 +216,7 @@ class SentimentAnalyzer:
             # "sa_no_aoa_in_neutral": NewArabicPreprocessorBalanced(model_name='UBC-NLP/MARBERT'),
             # "sa_cnnbert": CNNMarbertArabicPreprocessor(model_name='UBC-NLP/MARBERT'),
             # "sa_sarcasm": SarcasmArabicPreprocessor(model_name='UBC-NLP/MARBERT'),
-            # "sar_trial10": SarcasmArabicPreprocessor(model_name='UBC-NLP/MARBERT'),
+            "sar_trial10": SarcasmArabicPreprocessor(model_name='UBC-NLP/MARBERT'),
             # "sa_no_AOA": NewArabicPreprocessorBalanced(model_name='UBC-NLP/MARBERT'),
         }
 
@@ -224,7 +224,7 @@ class SentimentAnalyzer:
             "sa_trial5_1": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sa_trial5_1",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_trial5_1")],
             # "sa_no_aoa_in_neutral": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sa_no_aoa_in_neutral",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_no_aoa_in_neutral")],
             # "sa_cnnbert": [CNNTextClassificationPipeline("{}/train_{}/best_model".format("sa_cnnbert",i), device=-1, return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_cnnbert")],
-            # "sa_sarcasm": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sa_sarcasm",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_sarcasm")],
+            "sa_sarcasm": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sa_sarcasm",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_sarcasm")],
             # "sar_trial10": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sar_trial10",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sar_trial10")],
             # "sa_no_AOA": [pipeline("sentiment-analysis", model="{}/train_{}/best_model".format("sa_no_AOA",i), device=-1,return_all_scores =True) for i in tqdm(range(0,5), desc=f"Loading pipeline for model: sa_no_AOA")],
         }