Spaces:
Build error
Build error
File size: 3,803 Bytes
0d99179 |
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 |
from .model import InformationExtractedFromABillReceipt as PydanticModel
from langchain.chains import LLMChain
from langchain.chat_models import ChatOpenAI
from langchain.output_parsers import PydanticOutputParser, OutputFixingParser
from langchain.prompts import (
ChatPromptTemplate,
HumanMessagePromptTemplate,
SystemMessagePromptTemplate,
)
model = ChatOpenAI(
temperature=0,
n=1,
model_kwargs={
'stop': None,
'top_p': 1,
'frequency_penalty': 0,
'presence_penalty': 0,
}
)
# Build category chain
system_message_prompt = SystemMessagePromptTemplate.from_template(
"You are an information extraction engine that outputs details from OCR processed "
"documents like uids, total, tax, name, currency, date, seller details, summary. You "
"may use context to make an educated guess about the currency. Use null if you are "
"unable to find certain details\n"
"{format_instructions}"
)
human_message_prompt = HumanMessagePromptTemplate.from_template("{text}")
chat_prompt = ChatPromptTemplate.from_messages(
[system_message_prompt, human_message_prompt]
)
output_parser = PydanticOutputParser(pydantic_object=PydanticModel)
fixing_parser = OutputFixingParser.from_llm(llm=model, parser=output_parser)
chain = LLMChain(llm=model, prompt=chat_prompt, output_parser=fixing_parser)
if __name__ == "__main__":
text = """amazonin
we)
Sold By :
Spigen India Pvt. Ltd.
* Rect/Killa Nos. 38//8/2 min, 192//22/1,196//2/1/1,
37//15/1, 15/2,, Adjacent to Starex School, Village
- Binola, National Highway -8, Tehsil - Manesar
Gurgaon, Haryana, 122413
IN
PAN No: ABACS5056L
GST Registration No: O6ABACS5056L12Z5
Order Number: 407-5335982-7837125
Order Date: 30.05.2023
Tax Invoice/Bill of Supply/Cash Memo
(Original for Recipient)
Billing Address :
Praveen Bohra
E-303, ParkView City 2, Sector 49, Sohna Road
GURGAON, HARYANA, 122018
IN
State/UT Code: 06
Shipping Address :
Praveen Bohra
Praveen Bohra
E-303, ParkView City 2, Sector 49, Sohna Road
GURGAON, HARYANA, 122018
IN
State/UT Code: 06
Place of supply: HARYANA
Place of delivery: HARYANA
Invoice Number : DEL5-21033
Invoice Details : HR-DEL5-918080915-2324
Invoice Date : 30.05.2023
Description at Tax |Tax /|Tax Total
p y Rate |Type |Amount|Amount
Black) | BO8BHLZHBH ( ACS01744INP )
HSN:39269099
1 |Spigen Liquid Air Back Cover Case for iPhone 12 Mini (TPU | Matte
1846.62] 1 |%846.62| 9% |CGST! %76.19 |%999.00
9% |SGST| %76.19
TOTAL:
Amount in Words:
Nine Hundred Ninety-nine only
Whether tax is payable under reverse charge - No
For Spigen India Pvt. Ltd.:
sSoigenrn
Authorized Signatory
Payment Transaction ID: Date & Time: 30/05/2023, 10:48:43 Invoice Value: Mode of Payment: Credit
2rs9ZEF8BwU9VmWiCc2Us hrs 999.00 Card
*ASSPL-Amazon Seller Services Pvt. Ltd., ARIPL-Amazon Retail India Pvt. Ltd. (only where Amazon Retail India Pvt. Ltd. fulfillment center is co-located)
Customers desirous of availing input GST credit are requested to create a Business account and purchase on Amazon.in/business from Business eligible offers
Please note that this invoice is not a demand for payment
Page 1 of 1"""
# result = chain.prompt.format_prompt(text=text, format_instructions=fixing_parser.get_format_instructions())
# print(result.json(indent=4))
result = chain.generate(input_list=[{"text": text, "format_instructions": fixing_parser.get_format_instructions()}])
print(result)
result = fixing_parser.parse_with_prompt(result.generations[0][0].text, chain.prompt.format_prompt(text=text, format_instructions=fixing_parser.get_format_instructions()))
print(result)
# result = chain.run(text=text, format_instructions=output_parser.get_format_instructions(), verbose=True)
# print(result)
|