Abuzaid01/asl-sign-language-classifier

ASL Sign Language Classification Model

This model is trained to recognize American Sign Language (ASL) alphabets using the ASL Alphabet Dataset.
It uses a ResNet50 backbone for image classification.

Model Details

• Base Architecture: ResNet50
• Number of Classes: 29
• Test Accuracy: 0.9999
• Dataset: ASL Alphabet (A–Z, space, delete, nothing)

Usage

from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import torch

# Load model and processor
model = AutoModelForImageClassification.from_pretrained("Abuzaid01/asl-sign-language-classifier")
processor = AutoImageProcessor.from_pretrained("Abuzaid01/asl-sign-language-classifier")

# Load an image
image = Image.open("asl_sample.jpg")

# Preprocess
inputs = processor(images=image, return_tensors="pt")

# Predict
with torch.no_grad():
    outputs = model(**inputs)
    logits = outputs.logits
    predicted_class = logits.argmax(-1).item()

print("Predicted class:", model.config.id2label[predicted_class])