Loading
Vous avez un workflow ComfyUI ? Obtenez une documentation lisible directement dans votre navigateur.Rendre un workflow lisible.Essayer maintenant ↗
A lightweight custom CNN for binary classification of chest X-rays (Normal vs Pneumonia) with Grad-CAM explainability.
Source du modèle
Description de la source
A lightweight custom CNN for binary classification of chest X-rays (Normal vs Pneumonia) with Grad-CAM explainability.
Sources
1 sourceVérifié 17 sept.
Artefacts du modèle
5 artefactspneumonia_classifier_aug_augmented_heavy_wypxe3qwqsx75ahb.pt
pt · 74,7 KB · SHA-256 1664549ec1d0…b6d3 · Hugging Face
Téléchargerpneumonia_classifier_aug_baseline_no_augmentation_zvyot3qwqgnsfahb.pt
pt · 74,7 KB · SHA-256 7b7e6c222351…9906 · Hugging Face
TéléchargerExtraits de sources
2 extraitsTry the live demo: Pneumonia AI Diagnostic Suite
| Metric | Score |
|---|---|
| Accuracy | 98% |
| Pneumonia Precision | 97% |
| Pneumonia Recall | 100% |
| Normal Precision | 100% |
| Normal Recall | 96% |
| Macro F1-Score | 0.98 |
| ROC AUC | 1.00 |
Evaluated on 57 test samples with optimized threshold (0.1) for high sensitivity.
| Fold | Accuracy | F1-Score |
|---|---|---|
| 1 | 92.86% | 0.927 |
| 2 | 97.62% | 0.976 |
| 3 | 92.86% | 0.933 |
| 4 | 95.24% | 0.950 |
| 5 | 97.62% | 0.977 |
| Mean | 95.24% | 0.953 |
| FP32 | INT8 | Improvement | |
|---|---|---|---|
| Size | 243 KB | 52 KB | 4.64x smaller |
| Latency | 108 ms | 27 ms | 3.99x faster |
import torch
from huggingface_hub import hf_hub_download
from pneumonia_classifier.ml.model.arch import Net
# Download model
model_path = hf_hub_download(
repo_id="24f2004275/pneumonia_classifier",
filename="pneumonia_classifier_cnn_uza7heywpgthvahb.pt"
)
# Load model
model = Net()
model.load_state_dict(torch.load(model_path, map_location="cpu", weights_only=False))
model.eval()
# Inference
from torchvision import transforms
transform = transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])
# image = Image.open("chest_xray.jpg").convert("RGB")
# tensor = transform(image).unsqueeze(0)
# with torch.no_grad():
# output = model(tensor)
# probs = torch.exp(output)
# prediction = "Pneumonia" if probs.argmax() == 1 else "Normal"
| File | Description |
|---|---|
pneumonia_classifier_cnn_uza7heywpgthvahb.pt | Primary model (FP32, best accuracy) |
pneumonia_classifier_cnn_int8_tgzwzsqwqw54dahb.pt | INT8 quantized variant |
pneumonia_classifier_cnn_int8_ra32tviwyo4u3ahb.pt | INT8 quantized variant (alt) |
pneumonia_classifier_aug_baseline_no_augmentation_zvyot3qwqgnsfahb.pt | Baseline (no augmentation) |
pneumonia_classifier_aug_augmented_heavy_wypxe3qwqsx75ahb.pt | Heavy augmentation variant |
This model integrates Grad-CAM (Gradient-weighted Class Activation Mapping) to visualize which regions of the chest X-ray the model focuses on for its predictions. The heatmap highlights areas of radiographical density associated with pneumonia.
MIT License
@misc{pneumonia_classifier,
title={Pneumonia Detection from Chest X-Rays using Custom CNN with Grad-CAM},
author={Ayush Dubey},
year={2024},
url={https://huggingface.co/24f2004275/pneumonia_classifier}
}
pneumonia_classifier_cnn_int8_ra32tviwyo4u3ahb.pt
pt · 58,4 KB · SHA-256 7baf81828226…6fe7 · Hugging Face
Téléchargerpneumonia_classifier_cnn_int8_tgzwzsqwqw54dahb.pt
pt · 74,4 KB · SHA-256 a5df6f830da4…040e · Hugging Face
Téléchargerpneumonia_classifier_cnn_uza7heywpgthvahb.pt
pt · 74,7 KB · SHA-256 d72a3192cbac…d1d2 · Hugging Face
Télécharger--- language: - en library_name: pytorch license: mit tags: - pneumonia - chest-xray - medical-imaging - cnn - image-classification - grad-cam - explainability - pytorch - custom-cnn pipeline_tag: image-classification metrics: - accuracy - f1 - precision - recall - roc_auc datasets: - chest-xray-pneumonia framework: - PyTorch - TorchVision --- # Pneumonia Classifier (Custom CNN) A lightweight custom CNN for binary classification of chest X-rays (Normal vs Pneumonia) with Grad-CAM explainability. ## Model Details - **Architecture**: Custom CNN (Net) — 9 Conv2d layers + Global Average Pooling - **Input**: RGB chest X-ray images, resized to 224×224 - **Output**: Binary classification (Normal / Pneumonia) with softmax probabilities - **Parameters**: ~35K (lightweight, edge-friendly) - **Framework**: PyTorch ## Demo Try the live demo: [Pneumonia AI Diagnostic Suite](https://pneumonia-classifie.streamlit.app/) ## Performance | Metric | Score | |--------|-------| | Accuracy | 98% | | Pneumonia Precision | 97% | | Pneumonia Recall | 100% | | Normal Precision | 100% | | Normal Recall | 96% | | Macro F1-Score | 0.98 | | ROC AUC | 1.00 | > Evaluated on 57 test samples with optimized threshold (0.1) for high sensitivity. ### Cross-Validation (5-Fold Stratified) | Fold | Accuracy | F1-Score | |------|----------|----------| | 1 | 92.86% | 0.927 | | 2 | 97.62% | 0.976 | | 3 | 92.86% | 0.933 | | 4 | 95.24% | 0.950 | | 5 | 97.62% | 0.977 | | **Mean** | **95.24%** | **0.953** | ## Training - **Optimizer**: SGD with Momentum (0.8) - **Loss**: Negative Log Likelihood (NLL) on log_softmax output - **Epochs**: 18 (early stopping, patience=5) - **Best Training Accuracy**: 98.25% - **Data Augmentation**: Baseline (no augmentation) — the unaugmented model performed better (95.24% CV vs 92.38% with heavy augmentation) ## Quantization (INT8) | | FP32 | INT8 | Improvement | |--|------|------|-------------| | Size | 243 KB | 52 KB | **4.64x** smaller | | Latency | 108 ms | 27 ms | **3.99x** faster | ## Usage ```python import torch from huggingface_hub import hf_hub_download from pneumonia_classifier.ml.model.arch import Net # Download model model_path = hf_hub_download( repo_id="24f2004275/pneumonia_classifier", filename="pneumonia_classifier_cnn_uza7heywpgthvahb.pt" ) # Load model model = Net() model.load_state_dict(torch.load(model_path, map_l...
Source context: 0 downloads · 0 likes · Pipeline image-classification · Library pytorch · Repo 24f2004275/pneumonia_classifier