Instructions to use hugginglearners/multi-object-classification with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- fastai
How to use hugginglearners/multi-object-classification with fastai:
from huggingface_hub import from_pretrained_fastai learn = from_pretrained_fastai("hugginglearners/multi-object-classification") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from hugginglearners/multi-object-classification: direct link, hf CLI and curl.
- Browser
- Download file 628 Bytes
-
https://huggingface.co/hugginglearners/multi-object-classification/resolve/main/README.md
- Command line
-
hf download hf://hugginglearners/multi-object-classification/README.md
-
curl -L -o README.md https://huggingface.co/hugginglearners/multi-object-classification/resolve/main/README.md
628 Bytes
metadata
tags:
- fastai
- image-classification
Model description
This repo contains the trained model for Multi-object classification
Full credits go to Nhu Hoang
Motivation: Classifying multiple objects is a challenging task without using an object detection algorithm. This model was trained on resnet34 backbone and achieved a good accuracy.
Training and evaluation data
Training hyperparameters
The following hyperparameters were used during training:
| Hyperparameters | Value |
|---|---|
| name | Adam |
| learning_rate | 3e-3 |
| training_precision | float16 |