Update app.py

app.py

@@ -1,24 +1,26 @@
 import torch
-from utils import label_full_decoder
+# from utils import label_full_decoder
 import sys
-import dataset
-import engine
-from model import BERTBaseUncased
-from tokenizer import tokenizer
-import config
-
+# import dataset
+# import engine
+# from model import BERTBaseUncased
+# from tokenizer import tokenizer
+# import config
+from transformers import pipeline, AutoTokenizer, AutoModel
 import gradio as gr
 
 DEVICE = config.device
 
+classifier = pipeline("sentiment-analysis",model="thak123/bert-emoji-latvian-twitter-classifier", tokenizer = "FFZG-cleopatra/bert-emoji-latvian-twitter")
+
 # MODEL = BERTBaseUncased()
 # MODEL.load_state_dict(torch.load(config.MODEL_PATH, map_location=torch.device(DEVICE)))
 # MODEL.eval()
 
 
 
-T = tokenizer.TweetTokenizer(
-    preserve_handles=True, preserve_hashes=True, preserve_case=False, preserve_url=False)
+# T = tokenizer.TweetTokenizer(
+#     preserve_handles=True, preserve_hashes=True, preserve_case=False, preserve_url=False)
 
 def preprocess(text):
     tokens = T.tokenize(text)
@@ -43,29 +45,32 @@ def preprocess(text):
 
 def sentence_prediction(sentence):
     sentence = preprocess(sentence)
-    model_path = config.MODEL_PATH
+    # model_path = config.MODEL_PATH
 
-    test_dataset = dataset.BERTDataset(
-        review=[sentence],
-        target=[0]
-    )
+    # test_dataset = dataset.BERTDataset(
+    #     review=[sentence],
+    #     target=[0]
+    # )
 
-    test_data_loader = torch.utils.data.DataLoader(
-        test_dataset,
-        batch_size=config.VALID_BATCH_SIZE,
-        num_workers=3
-    )
+    # test_data_loader = torch.utils.data.DataLoader(
+    #     test_dataset,
+    #     batch_size=config.VALID_BATCH_SIZE,
+    #     num_workers=3
+    # )
 
     # device = config.device
 
-    model = BERTBaseUncased()
-    # model.load_state_dict(torch.load(
-    #     model_path, map_location=torch.device(device)))
-    model.to(device)
+    # model = BERTBaseUncased()
+    # # model.load_state_dict(torch.load(
+    # #     model_path, map_location=torch.device(device)))
+    # model.to(device)
 
-    outputs, [] = engine.predict_fn(test_data_loader, MODEL, device)
+    # outputs, [] = engine.predict_fn(test_data_loader, MODEL, device)
+
+    outputs =  sentiment(input_text)
+    
     print(outputs)
-    return {"label":outputs[0]}
+    return outputs #{"label":outputs[0]}
 
 if __name__ == "__main__":