How to use one hot encoded data in Keras input
Issue I’m trying to build a model that is using a one-hot encoded input next to a numeric feature input….
Category: Tensorflow
Training fails if model is saved beforehand
Issue I noticed that saving a Tensorflow model prior to training causes the training to perform poorly. Obviously the solution…
How to save tensorflow recommenders framework model
Issue good tuto, thank you to that, but I have a problem to save model to reuse it. I have…
Model was constructed with shape (None, 65536) but it was called on an input with incompatible shape (None, 65536, None)
Issue For reference the full error is here: WARNING:tensorflow:Model was constructed with shape (None, 65536) for input KerasTensor(type_spec=TensorSpec(shape=(None, 65536), dtype=tf.float32,…
Tensorflow: How to zip together multiple concatenated datasets?
Issue I am attempting to zip together multiple tf.data.Dataset objects for a multi-input keras model. Each tf.data.Dataset object is a…
Module: tf.keras.datasets. Set where (directory) to download the dataset
Issue I’m using tf.keras.datasets to download CIFAR 10 dataset and I wondering where the images are downloaded. I’ve been searching…
Custom activation function in Tensorflow with trainable params
Issue I am trying to implement a custom version of the PElu activation function in tensorflow. The custom thing about…
Colab+TPU not supporting TF 2.3.0 tf.keras.layers.experimental.preprocessing
Issue I was updating my model using TF 2.3.0 on Colab+TPU based on https://keras.io/examples/vision/image_classification_efficientnet_fine_tuning/, specifically following the Data augmentation and…
Error with custom layer when using tf.function
Issue I created a classification model using LSTM and Attention (custom layer). The model trains fine with model.fit or without…
CNN model with high accuracy and right evaluations but wrong predictions
Issue I am working on a tumor classification problem where I have 4 classes [glioma(1624 sample), meningioma(1645 sample), notumor(2000 sample)…