Skorch Usage
Last updated
Last updated
We use skorch to implement multi-layer, fully-connected torch neural networks in Concrete-ML in a way that is compatible with the scikit-learn API.
This wrapper implements torch training boilerplate code, alleviating the work that needs to be done by the user. It is possible to add hooks during the training phase, for example once an epoch is finished.
Skorch allows the user to easily create a classifier or regressor around a neural network (NN), implemented in Torch as a nn.Module
. We provide a simple, fully-connected, multi-layer NN with a configurable number of layers and optional pruning (see pruning).
The SparseQuantNeuralNetImpl
class implements this neural network. Please see the documentation on this class in the API guide.
The constructor of this class takes some parameters that influence FHE compatibility:
n_w_bits
(default 3): number of bits for weights
n_a_bits
(default 3): number of bits for activations and inputs
n_accum_bits
(default 7): maximum accumulator bit width to impose through pruning
n_hidden_neurons_multiplier
(default 4): explained below
A linear or convolutional layer of an NN will compute a linear combination of weights and inputs (we also call this a 'multi-sum'). For example, a linear layer will compute:
where is the k-th neuron in the layer. In this case, the sum is taken on a single dimension. A convolutional layer will compute:
where is the k-th filter of the convolutional layer and , , are the number of input channels, the kernel height and the kernel width, respectively.
Following the formulas for the resulting bit width of quantized linear combinations described here, notably the maximum dimensionality of the input and weights that can make the result exceed 7 bits:
The pruning mechanism is already implemented in SparseQuantNeuralNetImpl
, and the user only needs to determine the parameters listed above. They can choose them in a way that is convenient, e.g. maximizing accuracy.
The skorch wrapper requires that all the parameters that will be passed to the wrapped nn.Module
be prefixed with module__
. For example, the code to create an FHE-compatible Concrete-ML fully-connected NN classifier for a dataset with 10 input dimensions and two classes, will thus be:
We could then increase n_hidden_neurons_multiplier
to improve performance, taking care to verify that the compiled NN does not exceed 7 bits of accumulator bit width.
A similar example is given in the classifier comparison notebook.
where is the maximum precision allowed.
For example, we set and with . The worst case is a scenario where all inputs and weights are equal to their maximal value . The formula above tells us that, in this case, we can afford at most elements in the multi-sums detailed above.
In a practical setting, the distribution of the weights of a neural network is Gaussian. Thus, there will be weights that are equal to 0 and many weights will have small values. In a typical scenario, we can exceed the worst-case number of active neurons. The parameter n_hidden_neurons_multiplier
is a factor that is multiplied with to determine the total number of non-zero weights that should be kept in a neuron.