https://github.com/qubvel/resnet_152

Pretrained on ImageNet ResNet-152 model in Keras
https://github.com/qubvel/resnet_152

keras pre-trained pretrained resnet-152

Last synced: 10 days ago
JSON representation

Pretrained on ImageNet ResNet-152 model in Keras

• Host: GitHub
• URL: https://github.com/qubvel/resnet_152
• Owner: qubvel
• Created: 2018-05-11T14:41:16.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2018-05-12T17:31:31.000Z (over 6 years ago)
• Last Synced: 2024-10-06T12:24:19.319Z (about 1 month ago)
• Topics: keras, pre-trained, pretrained, resnet-152
• Language: Python
• Homepage:
• Size: 110 KB
• Stars: 5
• Watchers: 2
• Forks: 6
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# ResNet-152 model for Keras

Reference:

- [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385)

Extention of model taken [here](https://gist.github.com/flyyufelix/7e2eafb149f72f4d38dd661882c554a6).  

Model made in Keras style with pretrained weights from ImageNet provided in release (load automatically during model initialization).  

### Example  

##### Inference example for this picture:  



##### Code

```python

import numpy as np

from skimage.io import imread

from skimage.transform import resize

from keras.applications.imagenet_utils import decode_predictions

from keras.applications.imagenet_utils import preprocess_input

from model import ResNet152

# create model

model = ResNet152()

# define function for input preprocessing

def preprocess(x):

    x = resize(x, (224,224), mode='constant') * 255

    x = preprocess_input(x)

    if x.ndim == 3:

        x = np.expand_dims(x, 0)

    return x

# prepare image

img = imread('./imgs/cat.jpg')

x = preprocess(img)

# make prediction and decode it

y = model.predict(x)

pred_title = decode_predictions(y, top=1)[0][0][1]

# print result

print(pred_title)

### tiget_cat

```