Spaces:
Paused
Paused
File size: 3,783 Bytes
7c99e50 f52050f 7c99e50 f52050f 7c99e50 f52050f 7c99e50 f52050f 7c99e50 4f59075 f52050f 7c99e50 f52050f 7c99e50 964f66f 7c99e50 4f59075 c659e77 7c99e50 f52050f 7c99e50 |
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 |
import re
import gradio as gr
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info
import torch
from ast import literal_eval
# default: Load the model on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained(
"Qwen/Qwen2-VL-7B-Instruct", torch_dtype="auto", device_map="auto"
)
# default processer
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
device = "cuda" if torch.cuda.is_available() else "cpu"
model.to(device)
other_benifits = '''Extract the following information in the given format:
{'other_benefits_and_information': {
'401k eru: {'This Period':'', 'Year-to-Date':''}},
'quota summary':
{
'sick:': '',
'vacation:': '',
}
'payment method': '',
'Amount': ''
}
'''
tax_deductions = '''Extract the following information in the given format:
{
'tax_deductions': {
'federal:': {
'withholding tax:': {'Amount':'', 'Year-To_Date':""},
'ee social security tax:': {'Amount':'', 'Year-To_Date':""},
'ee medicare tax:': {'Amount':'', 'Year-To_Date':""}},
'california:': {
'withholding tax:': {'Amount':'', 'Year-To_Date':""},
'ee disability tax:': {'Amount':'', 'Year-To_Date':""}}},
}
'''
def demo(image_name, prompt):
messages = [
{
"role": "user",
"content": [
{
"type": "image",
"image": image_name,
},
{"type": "text", "text": prompt},
],
}
]
# Preparation for inference
text = processor.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
text=[text],
images=image_inputs,
videos=video_inputs,
padding=True,
return_tensors="pt",
)
inputs = inputs.to("cuda")
# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=1500)
generated_ids_trimmed = [
out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
try:
# almost_json = output_text[0].replace('```\n', '').replace('\n```', '')
almost_json = output_text[0].split('```\n')[-1].split('\n```')[0]
json = literal_eval(almost_json)
except:
try:
# almost_json = output_text[0].replace('```json\n', '').replace('\n```', '')
almost_json = output_text[0].split('```json\n')[-1].split('\n```')[0]
json = literal_eval(almost_json)
except:
json = output_text[0]
return json
def process_document(image):
one = demo(image, other_benifits)
two = demo(image, tax_deductions)
return one, two
# article = "<p style='text-align: center'><a href='https://www.xelpmoc.in/' target='_blank'>Made by Xelpmoc</a></p>"
demo = gr.Interface(
fn=process_document,
inputs="image",
outputs="json",
title="Template-Free OCR model",
# article=article,
enable_queue=True,
examples=[["example.png"], ["example_2.png"]],
cache_examples=False)
demo.launch() |