Instructions to use Riksarkivet/rtmdet_regions with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- HTRflow
How to use Riksarkivet/rtmdet_regions with HTRflow:
# CLI usage # see docs: https://ai-riksarkivet.github.io/htrflow/latest/getting_started/quick_start.html htrflow pipeline <path/to/pipeline.yaml> <path/to/image>
# Python usage from htrflow.pipeline.pipeline import Pipeline from htrflow.pipeline.steps import Task from htrflow.models.framework.model import ModelClass pipeline = Pipeline( [ Task( ModelClass, {"model": "Riksarkivet/rtmdet_regions"}, {} ), ]) - Notebooks
- Google Colab
- Kaggle
How to setup in Colab
Sorry, I am new to huggingface and machine learning.
I want to run only the region detection model on a handwritten text image. How to setup that?
Is there any documentation? Thanks!
Hi π
There is currently no integrations to use openmmlabs model such as the rtmdet model unfortunately.. However, we are currently working on a open soure package to cover this for this kinda model and others:
see here π htrflow_core
Their still no offical stable release for the package, but see here for an "example" on how to use the RTMdet model.
Also, you could atucally just use it directly by using the mmdet package provided by openmmlab
DetInferencer(model=model_config, weights=model_weights)
code block:
from mmdet.apis import DetInferencer
from mmdet.structures import DetDataSample
model = DetInferencer( model=model_config, weights=model_weights, model_args*)
results : DetDataSample = model(images, batch_size=batch_size, draw_pred=False, return_datasample=True, **kwargs)
Anyway, I hope this helped! π€
Stay tuned for htrflow_core release π