Update app.py

app.py

@@ -1,12 +1,44 @@
 import streamlit as st
 from transformers import pipeline
 
-pipe = pipeline(model="RuudVelo/dutch_news_classifier_bert_finetuned")
-text = st.text_area('Please type/copy/paste the Dutch article')
+#pipe = pipeline(model="RuudVelo/dutch_news_classifier_bert_finetuned")
+#text = st.text_area('Please type/copy/paste the Dutch article')
+
+#labels = ['Binnenland' 'Buitenland' 'Cultuur & Media' 'Economie' 'Koningshuis'
+# 'Opmerkelijk' 'Politiek' 'Regionaal nieuws' 'Tech']
+
+#if text:
+#   out = pipe(text)
+#   st.json(out)
+   
+   
+   # load tokenizer and model, create trainer
+  #model_name = "RuudVelo/dutch_news_classifier_bert_finetuned"
+  #tokenizer = AutoTokenizer.from_pretrained(model_name)
+  #model = AutoModelForSequenceClassification.from_pretrained(model_name)
+  #trainer = Trainer(model=model)  
+  #print(filename, type(filename))
+  #print(filename.name)
+  
+from transformers import BertForSequenceClassification, BertTokenizer
+
+model = BertForSequenceClassification.from_pretrained("RuudVelo/dutch_news_classifier_bert_finetuned")
+#from transformers import BertTokenizer
 
-labels = ['Binnenland' 'Buitenland' 'Cultuur & Media' 'Economie' 'Koningshuis'
- 'Opmerkelijk' 'Politiek' 'Regionaal nieuws' 'Tech']
+tokenizer = BertTokenizer.from_pretrained("RuudVelo/dutch_news_classifier_bert_finetuned")
+
+#text = ["this is one sentence", "this is another sentence"]
+text = st.text_area('Please type/copy/paste the Dutch article')
 
 if text:
-   out = pipe(text)
-   st.json(out)
+   encoding = tokenizer(text, return_tensors="pt")
+   outputs = model(**encoding)
+   predictions = outputs.logits.argmax(-1)
+   #out = pipe(text)
+   st.json(predictions)
+   
+#encoding = tokenizer(text, return_tensors="pt")
+
+# forward pass
+#outputs = model(**encoding)
+#predictions = outputs.logits.argmax(-1)