MANN for M1 Mac Users

MANN, which stands for Multitask Artificial Neural Networks, is a Python package which enables creating sparse multitask models compatible with TensorFlow. This package contains custom layers and utilities to facilitate the training and optimization of models using the Reduction of Sub-Network Neuroplasticity (RSN2) training procedure developed by AI Squared, Inc.

Installation

This package is available through PyPi, but because of the current experimental nature of TenroFlow on the M1 platform, there are additional steps required for installation. Firstly, the steps provided at this link must be followed to install the TensorFlow metal plugin and version of TensorFlow supported on the M1 SOC. Once those steps are completed, the following command can be run to install this package:

pip install mann-m1mac

Please note that MacOS 12.0.1 or later is required.

Capabilities

We strive to maintain complete feature parity between this version of the MANN package and the version available for all other platforms.

The MANN package includes two subpackages, the mann.utils package and the mann.layers package. As the name implies, the mann.utils package includes utilities which assist in model training. The mann.layers package includes custom Keras-compatible layers which can be used to train sparse multitask models.

Utils

The mann.utils subpackage has three main functions: the mask_model function, the get_custom_objects function, and the convert_model function.

1. mask_model
   • The mask_model function is central to the RSN2 training procedure and enables masking/pruning a model so a large percentage of the weights are inactive.
   • Inputs to the mask_model function are a TensorFlow model, a percentile in integer form, a method - either one of 'gradients' or 'magnitude', input data, and target data.
2. get_custom_objects
   • The get_custom_objects function takes no parameters and returns a dictionary of all custom objects required to load a model trained using this package.
3. remove_layer_masks
   • The remove_layer_masks function takes a trained model with masked layers and converts it to a model without masking layers.

Layers

The mann.layers subpackage contains custom Keras-compatible layers which can be used to train sparse multitask models. The layers contained in this package are as follows:

1. MaskedDense
   • This layer is nearly identical to the Keras Dense layer, but it supports masking and pruning to reduce the number of active weights.
2. MaskedConv2D
   • This layer is nearly identical to the Keras Conv2D layer, but it supports masking and pruning to reduce the number of active weights.
3. MultiMaskedDense
   • This layer supports isolating pathways within the network and dedicating them for individual tasks and performing fully-connected operations on the input data.
4. MultiMaskedConv2D
   • This layer supports isolating pathways within the network and dedicating them for individual tasks and performing convolutional operations on the input data.
5. MultiDense
   • This layer supports multitask inference using a fully-connected architecture and is not designed for training. Once a model is trained with the MultiMaskedDense layer, that layer can be converted into this layer for inference by using the mann.utils.remove_layer_masks function.
6. MultiConv2D
   • This layer supports multitask inference using a convolutional architecture and is not designed for training. Once a model is trained with the MultiMaskedConv2D layer, that layer can be converted to this layer for inference by using the mann.utils.remove_layer_masks function.
7. SelectorLayer
   • This layer selects which of the multiple inputs fed into it is returned as a result. This layer is designed to be used specifically with multitask layers.
8. SumLayer
   • This layer returns the element-wise sum of all of the inputs.
9. FilterLayer
   • This layer can be turned on or off, and indicates whether the single input passed to it should be output or if all zeros should be returned.
10. MultiMaxPool2D
    • This layer implements Max Pool operations on multitask inputs.

Additional Documentation and Training Materials

Additional documentation and training materials will be added to the AI Squared Website as we continue to develop this project and its capabilities.

Feature Roadmap

• Transformers
  • We are in the process of adding the Transformer Layer into this package. Creating these layers will enable the training of multitask compressed models specifically for Natural Language Processing (NLP). Stay tuned!