Spaces:
Runtime error
Runtime error
mertcobanov
commited on
Commit
•
46b1434
0
Parent(s):
first
Browse files- README.md +12 -0
- app.py +182 -0
- requirements.txt +5 -0
README.md
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: Deprem Ocr 2
|
3 |
+
emoji: 👀
|
4 |
+
colorFrom: green
|
5 |
+
colorTo: blue
|
6 |
+
sdk: gradio
|
7 |
+
sdk_version: 3.17.0
|
8 |
+
app_file: app.py
|
9 |
+
pinned: false
|
10 |
+
---
|
11 |
+
|
12 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
app.py
ADDED
@@ -0,0 +1,182 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from easyocr import Reader
|
3 |
+
from PIL import Image
|
4 |
+
import io
|
5 |
+
import json
|
6 |
+
import csv
|
7 |
+
import openai
|
8 |
+
import ast
|
9 |
+
import os
|
10 |
+
from deta import Deta
|
11 |
+
|
12 |
+
|
13 |
+
######################
|
14 |
+
import requests
|
15 |
+
import json
|
16 |
+
|
17 |
+
import os
|
18 |
+
import openai
|
19 |
+
|
20 |
+
|
21 |
+
|
22 |
+
class OpenAI_API:
|
23 |
+
def __init__(self):
|
24 |
+
self.openai_api_key = ''
|
25 |
+
|
26 |
+
def single_request(self, address_text):
|
27 |
+
|
28 |
+
openai.api_type = "azure"
|
29 |
+
openai.api_base = "https://damlaopenai.openai.azure.com/"
|
30 |
+
openai.api_version = "2022-12-01"
|
31 |
+
openai.api_key = os.getenv("API_KEY")
|
32 |
+
|
33 |
+
response = openai.Completion.create(
|
34 |
+
engine="Davinci-003",
|
35 |
+
prompt=address_text,
|
36 |
+
temperature=0.9,
|
37 |
+
max_tokens=256,
|
38 |
+
top_p=1.0,
|
39 |
+
n=1,
|
40 |
+
logprobs=0,
|
41 |
+
echo=False,
|
42 |
+
stop=None,
|
43 |
+
frequency_penalty=0,
|
44 |
+
presence_penalty=0,
|
45 |
+
best_of=1)
|
46 |
+
|
47 |
+
return response
|
48 |
+
|
49 |
+
########################
|
50 |
+
|
51 |
+
openai.api_key = os.getenv('API_KEY')
|
52 |
+
reader = Reader(["tr"])
|
53 |
+
|
54 |
+
|
55 |
+
def get_parsed_address(input_img):
|
56 |
+
|
57 |
+
address_full_text = get_text(input_img)
|
58 |
+
return openai_response(address_full_text)
|
59 |
+
|
60 |
+
|
61 |
+
def preprocess_img(inp_image):
|
62 |
+
gray = cv2.cvtColor(inp_image, cv2.COLOR_BGR2GRAY)
|
63 |
+
gray_img = cv2.bitwise_not(gray)
|
64 |
+
return gray_img
|
65 |
+
|
66 |
+
|
67 |
+
def get_text(input_img):
|
68 |
+
result = reader.readtext(input_img, detail=0)
|
69 |
+
return " ".join(result)
|
70 |
+
|
71 |
+
|
72 |
+
def save_csv(mahalle, il, sokak, apartman):
|
73 |
+
adres_full = [mahalle, il, sokak, apartman]
|
74 |
+
|
75 |
+
with open("adress_book.csv", "a", encoding="utf-8") as f:
|
76 |
+
write = csv.writer(f)
|
77 |
+
write.writerow(adres_full)
|
78 |
+
return adres_full
|
79 |
+
|
80 |
+
|
81 |
+
def get_json(mahalle, il, sokak, apartman):
|
82 |
+
adres = {"mahalle": mahalle, "il": il, "sokak": sokak, "apartman": apartman}
|
83 |
+
dump = json.dumps(adres, indent=4, ensure_ascii=False)
|
84 |
+
return dump
|
85 |
+
|
86 |
+
def write_db(data_dict):
|
87 |
+
# 2) initialize with a project key
|
88 |
+
deta_key = os.getenv('DETA_KEY')
|
89 |
+
deta = Deta(deta_key)
|
90 |
+
|
91 |
+
# 3) create and use as many DBs as you want!
|
92 |
+
users = deta.Base("deprem-ocr")
|
93 |
+
users.insert(data_dict)
|
94 |
+
|
95 |
+
|
96 |
+
def text_dict(input):
|
97 |
+
eval_result = ast.literal_eval(input)
|
98 |
+
write_db(eval_result)
|
99 |
+
|
100 |
+
return (
|
101 |
+
str(eval_result['city']),
|
102 |
+
str(eval_result['distinct']),
|
103 |
+
str(eval_result['neighbourhood']),
|
104 |
+
str(eval_result['street']),
|
105 |
+
str(eval_result['address']),
|
106 |
+
str(eval_result['tel']),
|
107 |
+
str(eval_result['name_surname']),
|
108 |
+
str(eval_result['no']),
|
109 |
+
)
|
110 |
+
|
111 |
+
def openai_response(ocr_input):
|
112 |
+
prompt = f"""Tabular Data Extraction You are a highly intelligent and accurate tabular data extractor from
|
113 |
+
plain text input and especially from emergency text that carries address information, your inputs can be text
|
114 |
+
of arbitrary size, but the output should be in [{{'tabular': {{'entity_type': 'entity'}} }}] JSON format Force it
|
115 |
+
to only extract keys that are shared as an example in the examples section, if a key value is not found in the
|
116 |
+
text input, then it should be ignored. Have only city, distinct, neighbourhood,
|
117 |
+
street, no, tel, name_surname, address Examples: Input: Deprem sırasında evimizde yer alan adresimiz: İstanbul,
|
118 |
+
Beşiktaş, Yıldız Mahallesi, Cumhuriyet Caddesi No: 35, cep telefonu numaram 5551231256, adim Ahmet Yilmaz
|
119 |
+
Output: {{'city': 'İstanbul', 'distinct': 'Beşiktaş', 'neighbourhood': 'Yıldız Mahallesi', 'street': 'Cumhuriyet Caddesi', 'no': '35', 'tel': '5551231256', 'name_surname': 'Ahmet Yılmaz', 'address': 'İstanbul, Beşiktaş, Yıldız Mahallesi, Cumhuriyet Caddesi No: 35'}}
|
120 |
+
Input: {ocr_input}
|
121 |
+
Output:
|
122 |
+
"""
|
123 |
+
|
124 |
+
openai_client = OpenAI_API()
|
125 |
+
response = openai_client.single_request(ocr_input)
|
126 |
+
resp = response["choices"][0]["text"]
|
127 |
+
print(resp)
|
128 |
+
resp = eval(resp.replace("'{", "{").replace("}'", "}"))
|
129 |
+
resp["input"] = ocr_input
|
130 |
+
dict_keys = [
|
131 |
+
'city',
|
132 |
+
'distinct',
|
133 |
+
'neighbourhood',
|
134 |
+
'street',
|
135 |
+
'no',
|
136 |
+
'tel',
|
137 |
+
'name_surname',
|
138 |
+
'address',
|
139 |
+
'input',
|
140 |
+
]
|
141 |
+
for key in dict_keys:
|
142 |
+
if key not in resp.keys():
|
143 |
+
resp[key] = ''
|
144 |
+
return resp
|
145 |
+
|
146 |
+
|
147 |
+
with gr.Blocks() as demo:
|
148 |
+
gr.Markdown(
|
149 |
+
"""
|
150 |
+
# Enkaz Bildirme Uygulaması
|
151 |
+
""")
|
152 |
+
gr.Markdown("Bu uygulamada ekran görüntüsü sürükleyip bırakarak AFAD'a enkaz bildirimi yapabilirsiniz. Mesajı metin olarak da girebilirsiniz, tam adresi ayrıştırıp döndürür. API olarak kullanmak isterseniz sayfanın en altında use via api'ya tıklayın.")
|
153 |
+
with gr.Row():
|
154 |
+
img_area = gr.Image(label="Ekran Görüntüsü yükleyin 👇")
|
155 |
+
ocr_result = gr.Textbox(label="Metin yükleyin 👇 ")
|
156 |
+
open_api_text = gr.Textbox(label="Tam Adres")
|
157 |
+
submit_button = gr.Button(label="Yükle")
|
158 |
+
with gr.Column():
|
159 |
+
with gr.Row():
|
160 |
+
city = gr.Textbox(label="İl")
|
161 |
+
distinct = gr.Textbox(label="İlçe")
|
162 |
+
with gr.Row():
|
163 |
+
neighbourhood = gr.Textbox(label="Mahalle")
|
164 |
+
street = gr.Textbox(label="Sokak/Cadde/Bulvar")
|
165 |
+
with gr.Row():
|
166 |
+
tel = gr.Textbox(label="Telefon")
|
167 |
+
with gr.Row():
|
168 |
+
name_surname = gr.Textbox(label="İsim Soyisim")
|
169 |
+
address = gr.Textbox(label="Adres")
|
170 |
+
with gr.Row():
|
171 |
+
no = gr.Textbox(label="Kapı No")
|
172 |
+
|
173 |
+
|
174 |
+
submit_button.click(get_parsed_address, inputs = img_area, outputs = open_api_text, api_name="upload_image")
|
175 |
+
|
176 |
+
ocr_result.change(openai_response, ocr_result, open_api_text, api_name="upload-text")
|
177 |
+
|
178 |
+
open_api_text.change(text_dict, open_api_text, [city, distinct, neighbourhood, street, address, tel, name_surname, no])
|
179 |
+
|
180 |
+
|
181 |
+
if __name__ == "__main__":
|
182 |
+
demo.launch()
|
requirements.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
openai
|
2 |
+
Pillow
|
3 |
+
easyocr
|
4 |
+
gradio
|
5 |
+
deta
|