segmentation_models
segmentation_models copied to clipboard
qubvel
Segmentation models with pretrained backbones. Keras and TensorFlow Keras.
 import efficientnet.keras as efn import keras.backend as backend with strategy.scope(): def build_model(): **eff0 = efn.EfficientNetB0(weights='imagenet',include_top=False,input_shape=(224,224,3))** X = GlobalAveragePooling2D()(eff0.output) X = Dropout(0.5)(X) X = Dense(64)(X) X = Activation('relu')(X) X =...
@qubvel I've ported your DeeplabV3+ implementation from `segmentation_models_pytorch`. I would appreciate your feedback. :)
Hi, I'm working on a two class problem (background & region of interest). I need to set the loss function's weights to [0, 1] to ensure the model is learning...
Hi, I've just acquired a new machine with the lastest GPU RTX3090. This GPU is only supported for CUDA 11 meaning that I can only use the latest version of...
After training a UNet on 224x224 images I store it as a SavedModel binary using TF 2.7. Later, I load this model in the same environment in order to do...
Hello! I am trying to run this command `model = sm.Unet("seresnet34")` after install the pip package and all the dependencies, and I get the `AttributeError: 'str' object has no attribute...
@qubvel Thanks for this great repo. Here is some segmentation model architecture that is worth adding. Please consider adding them to this great repo. - `Unet++` : [Code ](https://github.com/MrGiovanni/UNetPlusPlus/tree/master/keras) -...
Code base from https://github.com/keras-team/keras-applications/blob/master/keras_applications/xception.py
When I am trying to import segmentation models in google colab after completing the installation of the segmentation model in the colab notebook. It throws the following error : "ImportError:...
