stepfun-ai/GOT-OCR2_0 · Discussions

#49 opened 10 months ago by

Asdmkasomo

OCR Prompt Issue: Unexpected Underscores, Hyphens, or Missing Spaces in Output

#48 opened about 1 year ago by

Thaipitak

<unknown>:623: SyntaxWarning: invalid escape sequence '\l'

#47 opened about 1 year ago by

XiaHan19

Dataset sample

#46 opened about 1 year ago by

raminh921

Error: 'DynamicCache' object has no attribute 'get_max_length'

👍 1

6

#45 opened about 1 year ago by

tymasf

Poor performance with simple table extraction task

🔥 1

#43 opened over 1 year ago by

hanshupe

Update README.md

#42 opened over 1 year ago by

qubvel-hf

Fix code to comport with newer Transformers library

👍 3

#41 opened over 1 year ago by

ctranslate2-4you

Output in Markdown Format

#39 opened over 1 year ago by

sarathrajendran552

Batch inference

#38 opened over 1 year ago by

Stopwolf

input_ids = torch.as_tensor(inputs.input_ids).cuda() 是否有问题？

#37 opened over 1 year ago by

Moses25

Update modeling_GOT.py

#36 opened over 1 year ago by

lijicheng168

How can use GOT-OCR2_0 on vLLM or Ollama?

👀 1

#35 opened over 1 year ago by

MhmdSyd

Allow to use PIL Image in chat function

#34 opened over 1 year ago by

akshayballal

RuntimeError: Current CUDA Device does not support bfloat16. Please switch dtype to float16.

4

#32 opened over 1 year ago by

taoyang2

This is a license-related question.

#31 opened over 1 year ago by

mkh0813

Different Results from the demo

#30 opened over 1 year ago by

Ryanchynoweth

transformers.js implementation

#29 opened over 1 year ago by

subjugate

support device type `cpu` generate

#26 opened almost 2 years ago by

weege007

MODEL NOT WORKING!!!!!!!

#25 opened almost 2 years ago by

Pushpendra817

add stream params for chat and char_crop

#24 opened almost 2 years ago by

weege007

how to train it on custom dataset ?

➕ 4

#23 opened almost 2 years ago by

GardensOfBabylon29

support for onnx and openvino conversion

#22 opened almost 2 years ago by

Jession01

is it available to change output text format to Markdown?

10

#21 opened almost 2 years ago by

MhmdSyd

How to this model for another languages

#20 opened almost 2 years ago by

summon1d

Charing

#19 opened almost 2 years ago by

Hocine423

Upload modeling_got.py

#18 opened almost 2 years ago by

srimanth-d

Tesseract vs. GOT-OCR2_0: Which Performs Better for Text Extraction from Images?

#17 opened almost 2 years ago by

bubbleMilkTea

cpu

#16 opened almost 2 years ago by

goodasdgood

[output format] can the model's output be easily parsed? thx

#15 opened almost 2 years ago by

Eric108

Device Support

#14 opened almost 2 years ago by

YogeLiu

The results of the work are amazing，have some plans to support like llamacpp or other framework?

🚀🔥 3

#13 opened almost 2 years ago by

fsql

Alternative to filename string for image

#12 opened almost 2 years ago by

codysmith

Inquiry Regarding Model Performance and Attention Mask

#11 opened almost 2 years ago by

kartoshkafri

Bounding boxes in results

👍 4

#9 opened almost 2 years ago by

solankibhargav

Can it be used without GPU and CPU (intel or AMD) only?

4

#8 opened almost 2 years ago by

jawad1347

Table extraction

#7 opened almost 2 years ago by

Jkppp

megfile library error

#5 opened almost 2 years ago by

don412

Support for CPU and MAC version

#4 opened almost 2 years ago by

mallapraveen

ModuleNotFoundError: No module named 'fcntl'