tensorflow-image-models 0.0.11

Tensorflow image models

Install

pip install tensorflow_image_models==0.0.11

---

Usage

from tensorflow_image_models import list_models

list_models()

models list:
DPN92
DPN98
DPN131
DPN107
EfficientNetB0
EfficientNetB1
.....
.....
VGG16
VGG19
.....
.....
ViTGaint_14
ViTGaint_16

Create Model

from tensorflow_image_models import EfficientNet
model = EfficientNet.EfficientNetB0(classes=10)
model.summary()

from tensorflow_image_models import Inception
model = Inception.InceptionV4(classes=3, input_shape=(299,299,3)
model.summary()

---

Here is a list of input shape expected for each model

EfficientNetB0 = (224, 224, 3)
EfficientNetB1 = (240, 240, 3)
EfficientNetB2 = (260, 260, 3)
EfficientNetB3 = (300, 300, 3)
EfficientNetB4 = (380, 380, 3)
EfficientNetB5 = (456, 456, 3)
EfficientNetB6 = (528, 528, 3)
EfficientNetB7 = (600, 600, 3)
AlexNet = (224, 224, 3)
InceptionResNetV2 = (299, 299, 3)
InceptionV3 = (299, 299, 3)
LeNet = (32, 32, 1)
MobileNetV2 = (224, 224, 3)
ResNet18 = (224, 224, 3)
ResNet34 = (224, 224, 3)
VoVNet27 = (224, 224, 3)
VoVNet39 = (224, 224, 3)
VoVNet57 = (224, 224, 3)
ViTTiny = (224, 224, 3)
ViTSmall_8 = (224, 224, 3)
ViTSmall_16 = (224, 224, 3)
ViTSmall_32 = (224, 224, 3)
ViTBase_8 = (224, 224, 3)
ViTBase_16 = (224, 224, 3)
ViTBase_32 = (224, 224, 3)
ViTLarge_14 = (224, 224, 3)
ViTLarge_16 = (224, 224, 3)
ViTLarge_32 = (224, 224, 3)
ViTHuge_14 = (224, 224, 3)
ViTHuge_16 = (224, 224, 3)
ViTGaint_14 = (224, 224, 3)
ViTGaint_16 = (224, 224, 3)

The rest of the models receive any shape, and the default shape is (128,128,3)

---

License
This project is licensed under the MIT License

---

Author
DEEPOLOGY LAB