File size: 6,179 Bytes
d0c9c37
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ecddc77
d0c9c37
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ecddc77
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d0c9c37
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
import spaces
import gradio as gr
from marker.markdown_extractor import MarkdownExtractorConfig, MarkdownExtractor
from pdf-extractor.pdf_extractor import PDFExtractorConfig, PDFExtractor
from gemini.gemini_extractor import GeminiExtractorConfig, GeminiExtractor
from openai.oai_extractor import OAIExtractorConfig, OAIExtractor
from indexify_extractor_sdk import Content

markdown_extractor = MarkdownExtractor()
pdf_extractor = PDFExtractor()
gemini_extractor = GeminiExtractor()
oai_extractor = OAIExtractor()

@spaces.GPU
def use_marker(pdf_filepath):
	if pdf_filepath is None:
		raise gr.Error("Please provide some input PDF: upload an PDF file")
	
	with open(pdf_filepath, "rb") as f:
		pdf_data = f.read()
		
	content = Content(content_type="application/pdf", data=pdf_data)
	config = MarkdownExtractorConfig(batch_multiplier=2)
	
	result = markdown_extractor.extract(content, config)
	return result

@spaces.GPU
def use_pdf_extractor(pdf_filepath):
	if pdf_filepath is None:
		raise gr.Error("Please provide some input PDF: upload an PDF file")
	
	with open(pdf_filepath, "rb") as f:
		pdf_data = f.read()
		
	content = Content(content_type="application/pdf", data=pdf_data)
    config = PDFExtractorConfig(output_types=["text", "table"])
	
	result = pdf_extractor.extract(content, config)
	return result

@spaces.GPU
def use_gemini(pdf_filepath, key):
	if pdf_filepath is None:
		raise gr.Error("Please provide some input PDF: upload an PDF file")
	
	with open(pdf_filepath, "rb") as f:
		pdf_data = f.read()
		
	content = Content(content_type="application/pdf", data=pdf_data)
	config = GeminiExtractorConfig(prompt="Extract all text from the document.", model_name="gemini-1.5-flash", key=key)
	
	result = gemini_extractor.extract(content, config)
	return result

@spaces.GPU
def use_openai(pdf_filepath, key):
	if pdf_filepath is None:
		raise gr.Error("Please provide some input PDF: upload an PDF file")
	
	with open(pdf_filepath, "rb") as f:
		pdf_data = f.read()
		
	content = Content(content_type="application/pdf", data=pdf_data)
	config = OAIExtractorConfig(prompt="Extract all text from the document.", model_name="gpt-4o", key=key)
	
	result = oai_extractor.extract(content, config)
	return result

with gr.Blocks(title="PDF data extraction with Marker & Indexify") as marker_demo:
	gr.HTML("<h1 style='text-align: center'>PDF data extraction with Marker & <a href='https://getindexify.ai/'>Indexify</a></h1>")
	gr.HTML("<p style='text-align: center'>Indexify is a scalable realtime and continuous indexing and structured extraction engine for unstructured data to build generative AI applications</p>")
	gr.HTML("<h3 style='text-align: center'>If you like this demo, please ⭐ Star us on <a href='https://github.com/tensorlakeai/indexify' target='_blank'>GitHub</a>!</h3>")
	gr.HTML("<h4 style='text-align: center'>Here's an example notebook that demonstrates how to build a continous <a href='https://github.com/tensorlakeai/indexify/blob/main/docs/docs/examples/efficient_rag.ipynb' target='_blank'>extraction pipleine</a> with Indexify</h4>")

	with gr.Row():
		with gr.Column():
			gr.HTML(
				"<p><b>Step 1:</b> Upload a PDF file from local storage.</p>"

				"<p style='color: #A0A0A0;'>Use this demo for single PDF file only. "
				"You can extract from PDF files continuously and try various other extractors locally with "
				"<a href='https://getindexify.ai/'>Indexify</a>.</p>"
			)

			pdf_file = gr.File(type="filepath")

		with gr.Column():
			gr.HTML("<p><b>Step 2:</b> Run the extractor.</p>")

			go_button = gr.Button(
				value="Run extractor",
				variant="primary",
			)

			model_output_text_box = gr.Textbox(
				label="Extractor Output",
				elem_id="model_output_text_box",
			)

	with gr.Row():

		gr.HTML(
			"<p style='text-align: center'>"
				"Developed with 🫶 by <a href='https://getindexify.ai/' target='_blank'>Indexify</a> | "
				"a <a href='https://www.tensorlake.ai/' target='_blank'>Tensorlake</a> product"
			"</p>"
		)

	go_button.click(
		fn=use_marker, 
		inputs = [pdf_file],
		outputs = [model_output_text_box]
	)

with gr.Blocks(title="PDF data extraction with PDF Extractor & Indexify") as pdf_demo:
	gr.HTML("<h1 style='text-align: center'>PDF data extraction with PDF Extractor & <a href='https://getindexify.ai/'>Indexify</a></h1>")
	gr.HTML("<p style='text-align: center'>Indexify is a scalable realtime and continuous indexing and structured extraction engine for unstructured data to build generative AI applications</p>")
	gr.HTML("<h3 style='text-align: center'>If you like this demo, please ⭐ Star us on <a href='https://github.com/tensorlakeai/indexify' target='_blank'>GitHub</a>!</h3>")
	gr.HTML("<h4 style='text-align: center'>Here's an example notebook that demonstrates how to build a continous <a href='https://github.com/tensorlakeai/indexify/blob/main/docs/docs/examples/SEC_10_K_docs.ipynb' target='_blank'>extraction pipleine</a> with Indexify</h4>")

	with gr.Row():
		with gr.Column():
			gr.HTML(
				"<p><b>Step 1:</b> Upload a PDF file from local storage.</p>"

				"<p style='color: #A0A0A0;'>Use this demo for single PDF file only. "
				"You can extract from PDF files continuously and try various other extractors locally with "
				"<a href='https://getindexify.ai/'>Indexify</a>.</p>"
			)

			pdf_file = gr.File(type="filepath")

		with gr.Column():
			gr.HTML("<p><b>Step 2:</b> Run the extractor.</p>")

			go_button = gr.Button(
				value="Run extractor",
				variant="primary",
			)

			model_output_text_box = gr.Textbox(
				label="Extractor Output",
				elem_id="model_output_text_box",
			)

	with gr.Row():

		gr.HTML(
			"<p style='text-align: center'>"
				"Developed with 🫶 by <a href='https://getindexify.ai/' target='_blank'>Indexify</a> | "
				"a <a href='https://www.tensorlake.ai/' target='_blank'>Tensorlake</a> product"
			"</p>"
		)

	go_button.click(
		fn=use_pdf_extractor, 
		inputs = [pdf_file],
		outputs = [model_output_text_box]
	)

demo = gr.TabbedInterface([marker_demo, pdf_demo, gemini_demo, openai_demo], ["Marker Extractor", "PDF Extractor", "Gemini Extractor", "OpenAI Extractor"], theme=gr.themes.Soft())

demo.queue()
demo.launch()