dropout | efficient dropout for deep convolutional neural networks | Machine Learning library

I tried with scikeras but I got errors because it doesn't accept not-numerical inputs (in our case the input is in str format). So I came back to the standard keras wrapper.

The focal point here is that the model is not built correctly. The TextVectorization must be put inside the Sequential model like shown in the official documentation.

So the build_model function becomes:

This seems to be a bug in Tensorflow 2.7 when using model.save combined with the parameter save_format="tf", which is set by default. The layers RandomFlip, RandomRotation, RandomZoom, and RandomContrast are causing the problems, since they are not serializable. Interestingly, the Rescaling layer can be saved without any problems. A workaround would be to simply save your model with the older Keras H5 format model.save("test", save_format='h5'):

If you plan to use the SpatialDropout1D layer, it has to receive a 3D tensor (batch_size, time_steps, features), so adding an additional dimension to your tensor before feeding it to the dropout layer is one option that is perfectly legitimate. Note, though, that in your case you could use both SpatialDropout1D or Dropout:

The number of parameters is at most and indication how fast a model trains or runs inference. It might depend on many other factors.

Here some examples, which might influence the throughput of your model:

1. The activation function: ReLu activations are faster then e.g. ELU or GELU which have exponetial terms. Not only is computing an exponention number slower than a linear number, but also the gradient is much more complex to compute since in Case of Relu is constant number, the slope of the activation (e.g.1).
2. the bit precission used for your data. Some HW accelerators can make faster computations in float16 than in float32 and also reading less bits decreses latency.
3. Some layers might not have parameters but perform fixed calculations. Eventhough no parameter is added to the network's weight, a computation still is performed.
4. The archetecture of your training HW. Certain filter sizes and batch sizes can be computed more efficiently than others.
5. sometimes the speed of the computing HW is not the bottleneck, the input pipeline for loading and preprocessing your data

It's hard to tell without testing but in your particular example I would guess, that the following might slow down your inference:

1. large perceptive field with a 7x7 conv
2. leaky_relu is slightly slower than relu
3. Probably your data input pipeline is the bottleneck, not the inference speed. If the inference speed is much faster than the data preparation, it might appear that both models have the same speed. But in reality the HW is idle and waits for data.

To understand whats going on, you could either change some parameters and evaluate the speed, or you could analyze your input pipeline by tracing your hardware using tensorboard. Here is a smal guide: https://www.tensorflow.org/tensorboard/tensorboard_profiling_keras

Best, Sascha

First of all, metrics such as Precision and Recall are focused on the positive class only, avoiding the problems encountered by multi-class focus metrics in the case of the class imbalance. Thus, we may not obtain enough information about the performance of the negative class if we keep considering all indicators. Haibo He et al suggest the metrics below to rate both items:

1. Geometric Mean.
2. F-Measure.
3. Macro-Averaged Accuracy.
4. Newer Combinations of Threshold Metrics: Mean-Class-Weighted Accuracy, Optimized Precision, Adjusted Geometric Mean, Index of Balanced Accuracy.

My suggestions:

1. Use the PR-curve and the F1-score.
2. Try with geometric transformations, photometric transformations, random occlusions (to avoid overfitting), SMOTE, Tomek links (for undersampling majorities), etc.
3. Random undersampling may delete relevant features of your dataset. Again, analyze your dataset using KNN and other similar techniques.
4. Check this book: H. He and Y. Ma, Imbalanced Learning: Foundations, Algorithms, and Applications, Hoboken, New Jersey: Wiley-IEEE Press, 2013.

As I don't exactly know what changes you have made in the code... I don't have idea about your dataset. But I can see that you are trying to train the whole datset with one epoch and passing the steps per epoch directly. I would recommend to write it like this

set some batch_size 2^n power (for example 16 or 32 or etc) if you don't want to batch the dataset just set batch_size to 1

If the hyperspectral dataset is given to you as a large image with many channels, I suppose that the classification of each pixel should depend on the pixels around it (otherwise I would not format the data as an image, i.e. without grid structure). Given this assumption, breaking up the input picture into 1x1 parts is not a good idea as you are loosing the grid structure.

I further suppose that the order of the channels is arbitrary, which implies that convolution over the channels is probably not meaningful (which you however did not plan to do anyways).

Instead of reformatting the data the way you did, you may want to create a model that takes an image as input and also outputs an "image" containing the classifications for each pixel. I.e. if you have 10 classes and take a (145, 145, 200) image as input, your model would output a (145, 145, 10) image. In that architecture you would not have any fully-connected layers. Your output layer would also be a convolutional layer.

That however means that you will not be able to keep your current architecture. That is because the tasks for MNIST/CIFAR10 and your hyperspectral dataset are not the same. For MNIST/CIFAR10 you want to classify an image in it's entirety, while for the other dataset you want to assign a class to each pixel (while most likely also using the pixels around each pixel).

Some further ideas:

• If you want to turn the pixel classification task on the hyperspectral dataset into a classification task for an entire image, maybe you can reformulate that task as "classifying a hyperspectral image as the class of it's center (or top-left, or bottom-right, or (21th, 104th), or whatever) pixel". To obtain the data from your single hyperspectral image, for each pixel, I would shift the image such that the target pixel is at the desired location (e.g. the center). All pixels that "fall off" the border could be inserted at the other side of the image.
• If you want to stick with a pixel classification task but need more data, maybe split up the single hyperspectral image you have into many smaller images (e.g. 10x10x200). You may even want to use images of many different sizes. If you model only has convolution and pooling layers and you make sure to maintain the sizes of the image, that should work out.

First, call model.compile() if you really miss it.

Second, check x.shape. I made mock data generator, and it works fine.

text input must of type str (single example), List[str] (batch or single pretokenized example) or List[List[str]] (batch of pretokenized examples).

Solution to the above error:

Just use text_input = 'text'

instead of

text_input = tf.keras.layers.Input(shape=(), dtype=tf.string, name='text')