Update app.py

app.py

@@ -15,42 +15,12 @@ import requests
 import json
 
 import os
-import openai
-
-
 
-class OpenAI_API:
-    def __init__(self):
-        self.openai_api_key = ''
-        
-    def single_request(self, address_text):
-        
-        openai.api_type = "azure"
-        openai.api_base = "https://damlaopenai.openai.azure.com/"
-        openai.api_version = "2022-12-01"
-        openai.api_key = os.getenv("API_KEY")
-        
-        response = openai.Completion.create(
-          engine="Davinci-003",
-          prompt=address_text,
-          temperature=0.9,
-          max_tokens=256,
-          top_p=1.0,
-          n=1,
-          logprobs=0,
-          echo=False,
-          stop=None,
-          frequency_penalty=0,
-          presence_penalty=0,
-          best_of=1)
-
-        return response
 
 ########################
 
-openai.api_key = os.getenv('API_KEY')
 reader = Reader(["tr"])
-
+API_TOKEN = os.getenv("HF_TOKEN")
 
 def get_parsed_address(input_img):
 
@@ -98,16 +68,75 @@ def text_dict(input):
     write_db(eval_result)
 
     return (
-        str(eval_result['city']),
-        str(eval_result['distinct']),
-        str(eval_result['neighbourhood']),
-        str(eval_result['street']),
-        str(eval_result['address']),
+        str(eval_result['il']),
+        str(eval_result['ilce']),
+        str(eval_result['mahalle']),
+        str(eval_result['sokak']),
+        str(eval_result['Apartman/site']),
         str(eval_result['tel']),
-        str(eval_result['name_surname']),
+        str(eval_result['isim']) + str(eval_result['soyad']),
         str(eval_result['no']),
     )
-        
+
+
+
+
+import json
+import requests
+headers = {"Authorization": f"Bearer {API_TOKEN}"}
+API_URL = "https://api-inference.huggingface.co/models/deprem-ml/deprem-ner"
+
+
+
+def query(payload):
+    data = json.dumps(payload)
+    response = requests.request("POST", API_URL, headers=headers, data=data)
+    return json.loads(response.content.decode("utf-8"))
+
+
+def infer(ocr_input)
+    return query({"inputs": ocr_input})
+
+
+"""
+ORNEK NER OUTPUT
+[
+  {
+    "entity_group": "mahalle",
+    "score": 0.8160411715507507,
+    "word": "Akevler mahallesi",
+    "start": 14,
+    "end": 31
+  },
+  {
+    "entity_group": "sokak",
+    "score": 0.940501868724823,
+    "word": "Rüzgar sokak",
+    "start": 32,
+    "end": 44
+  },
+  {
+    "entity_group": "Apartman/Site",
+    "score": 0.8081040978431702,
+    "word": "Tuncay apartmanı",
+    "start": 45,
+    "end": 61
+  },
+  {
+    "entity_group": "ilce",
+    "score": 0.854024350643158,
+    "word": "Antakya",
+    "start": 72,
+    "end": 79
+  }
+]"""
+
+
+
+
+
+
+# this is not in use due to content filter
 def openai_response(ocr_input):
     prompt = f"""Tabular Data Extraction You are a highly intelligent and accurate tabular data extractor from 
             plain text input and especially from emergency text that carries address information, your inputs can be text 
@@ -144,6 +173,7 @@ def openai_response(ocr_input):
     return resp
 
 
+
 with gr.Blocks() as demo:
     gr.Markdown(
     """
@@ -173,7 +203,7 @@ with gr.Blocks() as demo:
 
     submit_button.click(get_parsed_address, inputs = img_area, outputs = open_api_text, api_name="upload_image")
 
-    ocr_result.change(openai_response, ocr_result, open_api_text, api_name="upload-text")
+    ocr_result.change(query, ocr_result, open_api_text, api_name="upload-text")
 
     open_api_text.change(text_dict, open_api_text, [city, distinct, neighbourhood, street, address, tel, name_surname, no])