Pre-release
This release is a pre-release and may not be stable for production use.
RFConv
Rectified Convolution
Pretrained Model
| model | baseline | rectified |
|---|---|---|
| ResNet-50 | 76.66 | 77.10 |
| ResNet-101 | 78.13 | 78.74 |
| ResNeXt-50_32x4d | 78.17 | 78.48 |
| ResNeSt-50_2s8x | 78.73 | 79.38 |
Verify Models:
Prepare ImageNet dataset:
cd scripts/dataset/
# assuming you have downloaded the dataset in the current folder
python prepare_imagenet.py --download-dir ./
Test Model
# use resnest50 as an example
cd scripts/
python verify.py --model resnet50 --crop-size 224
Release files for rfconv 0.0.2b20210330
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| rfconv-0.0.2b20210330.tar.gz | 8.2 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| rfconv-0.0.2b20210330-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 24.3 kB
Release files / rfconv-0.0.2b20210330.tar.gz
| Download URL | rfconv-0.0.2b20210330.tar.gz |
|---|---|
| Size | 8.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
60031f1aac10db1805ae7fff9159553cf8ef4dc11318221445c2e26a6c971f57
|
|
BLAKE2b-256 checksum How to use checksums |
7578fd224bc8080cd7683bcdb254b100b3486af803ce5648fd5718eda1f8e4f7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.4.1 importlib_metadata/3.10.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.59.0 CPython/3.7.10
|
Release files / rfconv-0.0.2b20210330-py3-none-any.whl
| Download URL | rfconv-0.0.2b20210330-py3-none-any.whl |
|---|---|
| Size | 16.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
420fd7fd3613b8f56f74b5bbd51e02214df5b76a3ecd227371cac1b4f9746f73
|
|
BLAKE2b-256 checksum How to use checksums |
e3d3d29206b58b27a744d222b5f3d818ef582b3c6e94194a80e2f000e766adf1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.4.1 importlib_metadata/3.10.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.59.0 CPython/3.7.10
|