license: mit
datasets:
- chriamue/bird-species-dataset
language:
- en
metrics:
- accuracy
library_name: transformers
pipeline_tag: image-classification
tags:
- biology
- image-classification
- vision
model-index:
- name: bird-species-classifier
results:
- task:
type: ImageClassification
dataset:
type: chriamue/bird-species-dataset
name: Bird Species
config: default
split: validation
metrics:
- type: accuracy
value: 96.8
- type: loss
value: 0.1379
Model Card for "Bird Species Classifier"
Model Description
The "Bird Species Classifier" is a state-of-the-art image classification model designed to identify various bird species from images. It uses the EfficientNet architecture and has been fine-tuned to achieve high accuracy in recognizing a wide range of bird species.
How to Use
You can easily use the model in your Python environment with the following code:
from transformers import AutoFeatureExtractor, AutoModelForImageClassification
extractor = AutoFeatureExtractor.from_pretrained("chriamue/bird-species-classifier")
model = AutoModelForImageClassification.from_pretrained("chriamue/bird-species-classifier")
Applications
- Bird species identification for educational or ecological research.
- Assistance in biodiversity monitoring and conservation efforts.
- Enhancing user experience in nature apps and platforms.
Training Data
The model was trained on the "Bird Species" dataset, which is a comprehensive collection of bird images. Key features of this dataset include:
- Total Species: 525 bird species.
- Training Images: 84,635 images.
- Validation Images: 2,625 images.
- Test Images: 2,625 images.
- Image Format: Color images (224x224x3) in JPG format.
- Source: Sourced from Kaggle.
Training Results
The model achieved impressive results after 6 epochs of training:
- Accuracy: 96.8%
- Loss: 0.1379
- Runtime: 136.81 seconds
- Samples per Second: 19.188
- Steps per Second: 1.206
- Total Training Steps: 31,740
These metrics indicate a high level of performance, making the model reliable for practical applications.
Limitations and Bias
- The performance of the model might vary under different lighting conditions or image qualities.
- The model's accuracy is dependent on the diversity and representation in the training dataset. It may perform less effectively on bird species not well represented in the dataset.
Ethical Considerations
This model should be used responsibly, considering privacy and environmental impacts. It should not be used for harmful purposes such as targeting endangered species or violating wildlife protection laws.
Acknowledgements
We would like to acknowledge the creators of the dataset on Kaggle for providing a rich source of data that made this model possible.