Skip to content

Latest commit

 

History

History

resnext50_32x4d

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
models
models
 
 
utils
utils
 
 
README.md
README.md
 
 
infer.py
infer.py
 
 
infer.sh
infer.sh
 
 
requirements.txt
requirements.txt
 
 
train.py
train.py
 
 
train.sh
train.sh
 
 

README.md

ResNext50_32x4d

Train on imagenette Dataset

Prepare Traning Data And Pretrain Models

Download Ofrecord

wget https://oneflow-public.oss-cn-beijing.aliyuncs.com/datasets/imagenette_ofrecord.tar.gz
tar zxf imagenette_ofrecord.tar.gz

Download Pretrain Models

wget https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/cv/classification/resnext50_32x4d/resnext50_32x4d_oneflow_model.tar.gz

Run Oneflow Training script

bash train.sh

Inference on Single Image

bash infer.sh

Util

convert pytorch pretrained model to oneflow pretrained model

wget https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth
import torch
import oneflow as flow 
from models.resnext50_32x4d import resnext50_32x4d

parameters = torch.load("resnext50_32x4d-7cdf4587.pth")
new_parameters = dict()
for key,value in parameters.items():
     if "num_batches_tracked" not in key:
          val = value.detach().cpu().numpy()
          new_parameters[key] = val




resnext50_32x4d_module = resnext50_32x4d()
resnext50_32x4d_module.load_state_dict(new_parameters)
flow.save(resnext50_32x4d_module.state_dict(), "resnext50_32x4d_oneflow_model")