Create app.py

app.py

@@ -0,0 +1,49 @@
+from transformers import AutoConfig, AutoModelForTokenClassification, AutoTokenizer
+import numpy as np
+import torch
+import gradio as gr
+import requests
+
+test_model = AutoModelForTokenClassification.from_pretrained("PRAli22/arabert_arabic_ner")
+TOKENIZER =  AutoTokenizer.from_pretrained("PRAli22/arabert_arabic_ner")
+
+label_map = {'B-LOC': 0, 'O': 1, 'B-PERS': 2, 'I-PERS': 3, 'B-ORG': 4, 'I-LOC': 5, 'I-ORG': 6, 'B-MISC': 7, 'I-MISC': 8}
+inv_label_map = {0: 'B-LOC', 1: 'O', 2: 'B-PERS', 3: 'I-PERS', 4: 'B-ORG', 5: 'I-LOC', 6: 'I-ORG', 7: 'B-MISC', 8: 'I-MISC'}
+
+
+def predict_sent(sentences):
+    input_ids = TOKENIZER.encode(sentences, return_tensors='pt')
+
+    with torch.no_grad():
+        test_model.to('cpu')
+        output = test_model(input_ids)
+    label_indices = np.argmax(output[0].to('cpu').numpy(), axis=2)
+
+    tokens = TOKENIZER.convert_ids_to_tokens(input_ids.to('cpu').numpy()[0])
+
+    new_tokens, new_labels = [], []
+    for token, label_idx in zip(tokens, label_indices[0]):
+        if token.startswith("##"):
+            new_tokens[-1] = new_tokens[-1] + token[2:]
+        else:
+            new_labels.append(inv_label_map[label_idx])
+            new_tokens.append(token)
+
+    output_string = "\n".join(["{}\t{}".format(label, token) for token, label in zip(new_tokens, new_labels)])
+
+    return output_string
+
+css_code='body{background-image:url("https://media.istockphoto.com/id/1256252051/vector/people-using-online-translation-app.jpg?s=612x612&w=0&k=20&c=aa6ykHXnSwqKu31fFR6r6Y1bYMS5FMAU9yHqwwylA94=");}'
+
+demo = gr.Interface(
+    fn=predict_sent,
+    inputs=
+        gr.Textbox(label="sentence", placeholder=" Enter the sentence "),
+        
+    
+    outputs=[gr.Textbox(label="labels")],
+    title="Arabic Named Entity Recognition",
+    description= "This is Arabic Named Entity Recognition System, it takes a sentence  in Arabic as inputs and returns it's labels",
+    css = css_code
+)
+demo.launch()