Update app.py

app.py

@@ -96,7 +96,7 @@ model = nn.Sequential(
     nn.Dropout(0.05)  # Dropout
 )
 
-model.load_state_dict(torch.load("large.pth", map_location=torch.device('cpu')))
+model.load_state_dict(torch.load("large-tuned.pth", map_location=torch.device('cpu')))
 model.eval()
 
 
@@ -107,7 +107,7 @@ def inf(inpt):
     embedding = sentencemodel.encode(TEXT, convert_to_tensor=True)
     INPUT = torch.cat((probs, embedding))
     output = F.softmax(model(INPUT.view(1, -1)), dim=1)
-    if not output[0][1] > 0.68:
+    if not output[0][1] > 0.4:
         return "Not toxic " + str(output[0][0])
     else:
         return "Toxic! " + str(output[0][1])