|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
- yo |
|
|
metrics: |
|
|
- accuracy |
|
|
pipeline_tag: image-classification |
|
|
tags: |
|
|
- Yoruba |
|
|
- tradition |
|
|
- cap |
|
|
- fila |
|
|
- gobi |
|
|
- Nigeria |
|
|
- Oodua |
|
|
- gods |
|
|
--- |
|
|
# Fila Yoruba Detector |
|
|
|
|
|
## Model Description |
|
|
This model is a binary classification model that detects whether an image contains a traditional Yoruba people's **fila** (hat) or not. It was trained on a small set of images to recognize two classes: **fila** and **not-fila**. |
|
|
|
|
|
## Intended Use |
|
|
This model is designed to classify images of traditional Yoruba attire, specifically detecting the presence of a **fila**. It can be used in applications where recognizing this cultural item is important, such as in cultural heritage analysis or fashion recognition. |
|
|
|
|
|
## Training Data |
|
|
The model was trained on a dataset of images containing two classes: |
|
|
- **fila**: Images that contain the traditional Yoruba hat. |
|
|
- **not-fila**: Images that do not contain the traditional Yoruba hat. |
|
|
|
|
|
The training data consisted of a limited number of images and may not generalize well to other image sets outside of the Yoruba context. |
|
|
|
|
|
## Evaluation Metrics |
|
|
The model's performance was evaluated using standard classification metrics: |
|
|
- **Accuracy**: 85% |
|
|
- **Precision**: 88% |
|
|
- **Recall**: 80% |
|
|
- **F1 Score**: 84% |
|
|
|
|
|
Note: These metrics may vary depending on the images it is tested on. |
|
|
|
|
|
## Limitations |
|
|
- The model was trained on a limited set of images, which might affect its generalization to other types of images. |
|
|
- The model may have difficulty recognizing **fila** in images with poor lighting or obscured views of the hat. |
|
|
- The performance may degrade if the model encounters images from other cultures or different types of headgear. |
|
|
|
|
|
## Ethical Considerations |
|
|
- Ensure that the model is used in contexts that respect cultural diversity and avoid reinforcing stereotypes. |
|
|
- The model might not perform equally well across all demographic groups, and care should be taken to avoid misuse in sensitive contexts. |
|
|
|
|
|
## How to Use |
|
|
To use this model, you can load it from Hugging Face using the following code: |
|
|
|
|
|
```python |
|
|
from huggingface_hub import hf_hub_download |
|
|
import tensorflow as tf |
|
|
|
|
|
# Download and load the model |
|
|
model_path = hf_hub_download(repo_id="dolaposalim/model-name", filename="filadentification.h5") |
|
|
model = tf.keras.models.load_model(model_path) |
|
|
|
|
|
# Predict function |
|
|
def predict(image): |
|
|
# Preprocess the image and make predictions |
|
|
image = image.resize((256, 256)) # Resize to model input size |
|
|
image = np.array(image) / 255.0 # Normalize |
|
|
image = np.expand_dims(image, axis=0) # Add batch dimension |
|
|
prediction = model.predict(image) |
|
|
return prediction``` |
