<center>
[[torchood](https://github.com/LSTM-Kirigaya/torchood)]: pytorch implementation for universal out of distribution methods
```bash
pip install torchood
```
</center>
## Support Methods
| Support Methods | Source |
|:---|:---|
| Robust Classification with Convolutional Prototype Learning (Prototype Networks) | [CVPR 2018: Rbust classification with convolutional prototype learning](https://arxiv.org/pdf/1805.03438) |
| Predictive Uncertainty Estimation via Prior Networks(Prior Networks) | [NeurIPS 2018: Predictive Uncertainty Estimation via Prior Networks]() |
| Evidential Deep Learning to Quantify Classification Uncertainty (EDL) | [NeurIPS 2018: Evidential Deep Learning to Quantify Classification Uncertainty](https://papers.nips.cc/paper/2018/hash/a981f2b708044d6fb4a71a1463242520-Abstract.html) |
| Posterior Network (PostNet) | [NeurIPS 2020: Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts]() |
| Evidential Neural Network (ENN) | [NeurIPS 2018: Evidential Deep Learning to Quantify Classification Uncertainty](https://arxiv.org/pdf/1806.01768) |
| Evidence Reconciled Neural Network(ERNN) | [MICCAI 2023: Evidence Reconciled Neural Network for Out-of-Distribution Detection in Medical Images](https://conferences.miccai.org/2023/papers/249-Paper2401.html) |
| Redundancy Removing Evidential Neural Network(R2ENN) | - |
---
## Usage
```bash
pip install torchood
```
```python
import torchood
class UserDefineModel(nn.Module):
...
classifier = UserDefineModel(...)
classifier = torchood.EvidenceNeuralNetwork(classifier)
```
### Train
```python
for data, label in train_loader:
# transform data & label to correct device
# ...
_, evidence, _ = classifier(data)
loss = classifier.criterion(evidence, label)
optimizer.zero_grad()
loss.backward()
optimizer.step()
```
### Infer
```python
prob, uncertainty = classifier.predict(inputs)
```
Raw data
{
"_id": null,
"home_page": null,
"name": "torchood",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": null,
"keywords": "Pytorch, ood, out of distribution, out-of-distribution, classification",
"author": null,
"author_email": "lstm-kirigaya <1193466151@qq.com>",
"download_url": "https://files.pythonhosted.org/packages/06/7a/dc31ca4fb7344d0882e5e9b14f516122fbc8f50de6ed2f789e36619384f1/torchood-0.0.1.tar.gz",
"platform": null,
"description": "\n<center>\n\n\n\n\n[[torchood](https://github.com/LSTM-Kirigaya/torchood)]: pytorch implementation for universal out of distribution methods\n\n```bash\npip install torchood\n```\n\n</center>\n\n## Support Methods\n\n| Support Methods | Source |\n|:---|:---|\n| Robust Classification with Convolutional Prototype Learning (Prototype Networks) | [CVPR 2018: Rbust classification with convolutional prototype learning](https://arxiv.org/pdf/1805.03438) |\n| Predictive Uncertainty Estimation via Prior Networks(Prior Networks) | [NeurIPS 2018: Predictive Uncertainty Estimation via Prior Networks]() |\n| Evidential Deep Learning to Quantify Classification Uncertainty (EDL) | [NeurIPS 2018: Evidential Deep Learning to Quantify Classification Uncertainty](https://papers.nips.cc/paper/2018/hash/a981f2b708044d6fb4a71a1463242520-Abstract.html) |\n| Posterior Network (PostNet) | [NeurIPS 2020: Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts]() |\n| Evidential Neural Network (ENN) | [NeurIPS 2018: Evidential Deep Learning to Quantify Classification Uncertainty](https://arxiv.org/pdf/1806.01768) |\n| Evidence Reconciled Neural Network(ERNN) | [MICCAI 2023: Evidence Reconciled Neural Network for Out-of-Distribution Detection in Medical Images](https://conferences.miccai.org/2023/papers/249-Paper2401.html) |\n| Redundancy Removing Evidential Neural Network(R2ENN) | - |\n\n---\n\n## Usage\n\n```bash\npip install torchood\n```\n\n```python\nimport torchood\n\nclass UserDefineModel(nn.Module):\n ...\n\nclassifier = UserDefineModel(...)\nclassifier = torchood.EvidenceNeuralNetwork(classifier)\n```\n\n### Train\n\n```python\nfor data, label in train_loader:\n # transform data & label to correct device\n # ...\n\n _, evidence, _ = classifier(data)\n loss = classifier.criterion(evidence, label)\n\n optimizer.zero_grad()\n loss.backward()\n optimizer.step()\n```\n\n### Infer\n\n```python\nprob, uncertainty = classifier.predict(inputs)\n```\n\n\n",
"bugtrack_url": null,
"license": "Apache 2.0",
"summary": "implement of local laplace filter algorithm",
"version": "0.0.1",
"project_urls": {
"text": "https://github.com/LSTM-Kirigaya"
},
"split_keywords": [
"pytorch",
" ood",
" out of distribution",
" out-of-distribution",
" classification"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "cf3574d0c976886aa16f011313e43c3cad6cb06254eadfb69a10b7e98cdf32fb",
"md5": "872a3028586954b8c0a796bcf83355f9",
"sha256": "6e6ef8068eb38f9887aa60599c1472fce0533f5cd6db5c9694329ccdf2f8f32e"
},
"downloads": -1,
"filename": "torchood-0.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "872a3028586954b8c0a796bcf83355f9",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10",
"size": 8188,
"upload_time": "2024-12-16T11:01:33",
"upload_time_iso_8601": "2024-12-16T11:01:33.498675Z",
"url": "https://files.pythonhosted.org/packages/cf/35/74d0c976886aa16f011313e43c3cad6cb06254eadfb69a10b7e98cdf32fb/torchood-0.0.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "067adc31ca4fb7344d0882e5e9b14f516122fbc8f50de6ed2f789e36619384f1",
"md5": "0813a1b5082a54cc84aaec03565a157f",
"sha256": "2b254a7cb001bb822eff11d5e2fc0512a5e665b17db2a283493cc188b02be1ff"
},
"downloads": -1,
"filename": "torchood-0.0.1.tar.gz",
"has_sig": false,
"md5_digest": "0813a1b5082a54cc84aaec03565a157f",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 7632,
"upload_time": "2024-12-16T11:01:36",
"upload_time_iso_8601": "2024-12-16T11:01:36.407823Z",
"url": "https://files.pythonhosted.org/packages/06/7a/dc31ca4fb7344d0882e5e9b14f516122fbc8f50de6ed2f789e36619384f1/torchood-0.0.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-12-16 11:01:36",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "torchood"
}
JSON