This is an automated PR created with https://huggingface.co/spaces/safetensors/convert
This new file is equivalent to `pytorch_model.bin` but safe in the sense that
no arbitrary code can be put into it.
These files also happen to load much faster than their pytorch counterpart:
https://colab.research.google.com/github/huggingface/notebooks/blob/main/safetensors_doc/en/speed.ipynb
The widgets on your model page will run using this model even if this is not merged
making sure the file actually works.
If you find any issues: please report here: https://huggingface.co/spaces/safetensors/convert/discussions
Feel free to ignore this PR.
````
from llama_cpp import Llama
from paddleocr import PaddleOCR
text = ""
ocr = PaddleOCR(use_angle_cls=True, lang='en')
result = ocr.ocr("test_image.jpg", cls=True)
for idx in range(len(result)):
res = result[idx]
for line in res:
text = text + line[-1][0]+ "\n"
llm = Llama(model_path="inv.Q8_0.gguf",n_ctx=2048)
def extract_xml_answer(text: str) -> str:
answer = text.split("<answer>")[-1]
answer = answer.split("</answer>")[0]
return answer.strip()
messages = [
{"role": "system", "content": """Respond in the following format:
<reasoning>
...
</reasoning>
<answer>
```json
```
</answer>"""},
{"role": "user", "content": f"{text}+""""
Extract the data in JSON format using the schema:
{
"invoice_no":"string",
"issued_to": {
"name": "string",
"address": "string" // Address of the client
},
"pay_to": {
"bank_name": "string", // Name of the bank
"name": "string", // Name
"account_no": "number"
},
"items":[…
Took a small 1.5B model fine-tuned with RL (GRPO on Qwen2.5-Coder) and asked it to extract structured JSON from OCR text based on any user-defined schema. It needs more work but it still works! You can find the model and the gguf.
````
from llama_cpp import Llama
from paddleocr import PaddleOCR
text = ""
ocr = PaddleOCR(use_angle_cls=True, lang='en')
result = ocr.ocr("test_image.jpg", cls=True)
for idx in range(len(result)):
res = result[idx]
for line in res:
text = text + line[-1][0]+ "\n"
llm = Llama(model_path="inv.Q8_0.gguf",n_ctx=2048)
def extract_xml_answer(text: str) -> str:
answer = text.split("<answer>")[-1]
answer = answer.split("</answer>")[0]
return answer.strip()
messages = [
{"role": "system", "content": """Respond in the following format:
<reasoning>
...
</reasoning>
<answer>
```json
```
</answer>"""},
{"role": "user", "content": f"{text}+""""
Extract the data in JSON format using the schema:
{
"invoice_no":"string",…
