AshProg/bird-classifier-convnext

Bird Species Classification - ConvNeXt-Base

Model Description

This model classifies 200 bird species using ConvNeXt-Base architecture with transfer learning.

Performance

• Test Accuracy: 83.64%
• Average Per-Class Accuracy: 83.29%
• Architecture: ConvNeXt-Base (87M parameters)
• Dataset: CUB-200-2011 (200 bird species)

Training Details

Model Architecture

• Base Model: ConvNeXt-Base pretrained on ImageNet-1K
• Classifier: Custom 2-layer classifier with dropout
• Input Size: 224x224 RGB images

Training Strategy

• Phase 1 (40 epochs): Frozen backbone, train classifier only

  • Learning Rate: 0.001
  • Batch Size: 32

• Phase 2 (20 epochs): Full fine-tuning

  • Learning Rate: 0.0001
  • Batch Size: 32

Regularization

• Dropout: 0.6, 0.5
• Label Smoothing: 0.2
• Weight Decay: 0.005
• Data Augmentation: rotation, flip, color jitter, random erasing

Usage

import torch
import torch.nn as nn
from torchvision import models, transforms
from PIL import Image

# Load model
model = models.convnext_base(weights=None)
num_features = model.classifier[2].in_features
model.classifier[2] = nn.Sequential(
    nn.Dropout(0.6),
    nn.Linear(num_features, 1024),
    nn.ReLU(),
    nn.Dropout(0.5),
    nn.Linear(1024, 200)
)

# Load weights
model.load_state_dict(torch.load('final_model.pth', map_location='cpu'))
model.eval()

# Preprocessing
transform = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])

# Predict
image = Image.open('bird.jpg').convert('RGB')
image_tensor = transform(image).unsqueeze(0)

with torch.no_grad():
    outputs = model(image_tensor)
    probabilities = torch.nn.functional.softmax(outputs[0], dim=0)
    top5_prob, top5_indices = torch.topk(probabilities, 5)
    
print("Top 5 Predictions:")
for prob, idx in zip(top5_prob, top5_indices):
    print(f"Class {idx}: {prob.item()*100:.2f}%")

Try it out!

Try the live demo: Bird Species Classifier

Model Files

• final_model.pth (1.06 GB): Full model weights

Citation

Dataset: CUB-200-2011

@techreport{WahCUB_200_2011,
    Title = {{The Caltech-UCSD Birds-200-2011 Dataset}},
    Author = {Wah, C. and Branson, S. and Welinder, P. and Perona, P. and Belongie, S.},
    Year = {2011},
    Institution = {California Institute of Technology},
    Number = {CNS-TR-2011-001}
}

Contact

For questions or issues, please open an issue on the Space repository.