|
a |
|
b/experiment/ocr_gemini.ipynb |
|
|
1 |
{ |
|
|
2 |
"cells": [ |
|
|
3 |
{ |
|
|
4 |
"cell_type": "code", |
|
|
5 |
"execution_count": 7, |
|
|
6 |
"metadata": {}, |
|
|
7 |
"outputs": [ |
|
|
8 |
{ |
|
|
9 |
"name": "stdout", |
|
|
10 |
"output_type": "stream", |
|
|
11 |
"text": [ |
|
|
12 |
"C:\\Users\\hoang\\OneDrive\\Desktop\\nghich_prj\\Prescription-Text-Extraction-Pharmacy-Information-Retrieval\n" |
|
|
13 |
] |
|
|
14 |
} |
|
|
15 |
], |
|
|
16 |
"source": [ |
|
|
17 |
"%cd C:\\Users\\hoang\\OneDrive\\Desktop\\nghich_prj\\Prescription-Text-Extraction-Pharmacy-Information-Retrieval" |
|
|
18 |
] |
|
|
19 |
}, |
|
|
20 |
{ |
|
|
21 |
"cell_type": "code", |
|
|
22 |
"execution_count": 11, |
|
|
23 |
"metadata": {}, |
|
|
24 |
"outputs": [], |
|
|
25 |
"source": [ |
|
|
26 |
"import google.generativeai as genai\n", |
|
|
27 |
"import os\n", |
|
|
28 |
"from dotenv import load_dotenv\n", |
|
|
29 |
"from PIL import Image\n", |
|
|
30 |
"\n", |
|
|
31 |
"\n", |
|
|
32 |
"load_dotenv()\n", |
|
|
33 |
"genai.configure(api_key=os.environ[\"GEMINI_API_KEY\"])" |
|
|
34 |
] |
|
|
35 |
}, |
|
|
36 |
{ |
|
|
37 |
"cell_type": "code", |
|
|
38 |
"execution_count": 9, |
|
|
39 |
"metadata": {}, |
|
|
40 |
"outputs": [], |
|
|
41 |
"source": [ |
|
|
42 |
"model = genai.GenerativeModel(\"gemini-1.5-flash\")" |
|
|
43 |
] |
|
|
44 |
}, |
|
|
45 |
{ |
|
|
46 |
"cell_type": "code", |
|
|
47 |
"execution_count": 10, |
|
|
48 |
"metadata": {}, |
|
|
49 |
"outputs": [], |
|
|
50 |
"source": [ |
|
|
51 |
"PROMPT= \"\"\"\n", |
|
|
52 |
"Extract information for all medications and instructions listed in the given prescription in the following format:\n", |
|
|
53 |
"[\n", |
|
|
54 |
" {\n", |
|
|
55 |
" name: \"medication name\",\n", |
|
|
56 |
" \"amount\": \"dosage\",\n", |
|
|
57 |
" \"guide\": \"usage instructions, e.g., 1 tablet in the morning\",\n", |
|
|
58 |
" \"note\": \"additional instructions\"\n", |
|
|
59 |
" }\n", |
|
|
60 |
"]\n", |
|
|
61 |
"\n", |
|
|
62 |
"Just return the desired format.\n", |
|
|
63 |
"\"\"\"" |
|
|
64 |
] |
|
|
65 |
}, |
|
|
66 |
{ |
|
|
67 |
"cell_type": "code", |
|
|
68 |
"execution_count": 14, |
|
|
69 |
"metadata": {}, |
|
|
70 |
"outputs": [], |
|
|
71 |
"source": [ |
|
|
72 |
"img = genai.upload_file(path=r\"C:\\Users\\hoang\\OneDrive\\Desktop\\nghich_prj\\Prescription-Text-Extraction-Pharmacy-Information-Retrieval\\assets\\ocr_sample\\bo-y-te-lui-lo-trinh-bat-buoc-benh-vien-hang-3-tro-len-phai-bo-ke-don-thuoc-viet-tay-202424227.jpg\", display_name=\"Prescription\")\n", |
|
|
73 |
"res = model.generate_content([img, PROMPT])" |
|
|
74 |
] |
|
|
75 |
}, |
|
|
76 |
{ |
|
|
77 |
"cell_type": "code", |
|
|
78 |
"execution_count": 15, |
|
|
79 |
"metadata": {}, |
|
|
80 |
"outputs": [ |
|
|
81 |
{ |
|
|
82 |
"name": "stdout", |
|
|
83 |
"output_type": "stream", |
|
|
84 |
"text": [ |
|
|
85 |
"[\n", |
|
|
86 |
" {\n", |
|
|
87 |
" \"name\": \"Bitex beules\",\n", |
|
|
88 |
" \"amount\": \"800mg\",\n", |
|
|
89 |
" \"guide\": \"4 viên/ ngày\",\n", |
|
|
90 |
" \"note\": \"sáng 2 viên, tối 2 viên\"\n", |
|
|
91 |
" },\n", |
|
|
92 |
" {\n", |
|
|
93 |
" \"name\": \"Keralian\",\n", |
|
|
94 |
" \"amount\": \"600\",\n", |
|
|
95 |
" \"guide\": \"4 viên/ ngày\",\n", |
|
|
96 |
" \"note\": \"sáng 2 viên, tối 2 viên\"\n", |
|
|
97 |
" },\n", |
|
|
98 |
" {\n", |
|
|
99 |
" \"name\": \"Debelecol\",\n", |
|
|
100 |
" \"amount\": \"600mg\",\n", |
|
|
101 |
" \"guide\": \"15 ml\",\n", |
|
|
102 |
" \"note\": \"trước bữa sáng\"\n", |
|
|
103 |
" }\n", |
|
|
104 |
"]\n" |
|
|
105 |
] |
|
|
106 |
} |
|
|
107 |
], |
|
|
108 |
"source": [ |
|
|
109 |
"print(res.text)" |
|
|
110 |
] |
|
|
111 |
} |
|
|
112 |
], |
|
|
113 |
"metadata": { |
|
|
114 |
"kernelspec": { |
|
|
115 |
"display_name": "base", |
|
|
116 |
"language": "python", |
|
|
117 |
"name": "python3" |
|
|
118 |
}, |
|
|
119 |
"language_info": { |
|
|
120 |
"codemirror_mode": { |
|
|
121 |
"name": "ipython", |
|
|
122 |
"version": 3 |
|
|
123 |
}, |
|
|
124 |
"file_extension": ".py", |
|
|
125 |
"mimetype": "text/x-python", |
|
|
126 |
"name": "python", |
|
|
127 |
"nbconvert_exporter": "python", |
|
|
128 |
"pygments_lexer": "ipython3", |
|
|
129 |
"version": "3.9.13" |
|
|
130 |
} |
|
|
131 |
}, |
|
|
132 |
"nbformat": 4, |
|
|
133 |
"nbformat_minor": 2 |
|
|
134 |
} |