Update app.py

app.py

@@ -4,7 +4,6 @@ import shap
 import numpy as np
 import scipy as sp
 import torch
-import tensorflow as tf
 import transformers
 from transformers import pipeline
 from transformers import RobertaTokenizer, RobertaModel
@@ -21,7 +20,7 @@ csv.field_size_limit(sys.maxsize)
 device = "cuda:0" if torch.cuda.is_available() else "cpu"
 
 tokenizer = AutoTokenizer.from_pretrained("vikvenk/ADR_Detection")
-model = AutoModelForSequenceClassification.from_pretrained("vikvenk/ADR_Detection", from_tf=True).to(device)
+model = AutoModelForSequenceClassification.from_pretrained("vikvenk/ADR_Detection").to(device)
 # build a pipeline object to do predictions
 pred = transformers.pipeline("text-classification", model=model, 
                              tokenizer=tokenizer, return_all_scores=True)
@@ -50,8 +49,8 @@ ner_pipe = pipeline("ner", model=ner_model, tokenizer=ner_tokenizer, aggregation
 def adr_predict(x):
     encoded_input = tokenizer(x, return_tensors='pt')
     output = model(**encoded_input)
-    scores = output[0][0].detach().numpy()
-    scores = tf.nn.softmax(scores)
+    scores = output[0][0].detach()
+    scores = torch.nn.functional.softmax(scores)
    
     shap_values = explainer([str(x).lower()])
     # # Find the index of the class you want as the default reference (e.g., 'label_1')