Pysyft: Add Functionality to FixedPrecisionTensor

[iamtrask]

In https://github.com/OpenMined/PySyft/pull/1530 and https://github.com/OpenMined/PySyft/pull/1531, we created a new class called FixedPrecisionTensor which is able to take any pytorch tensor and compute over it using a limited range of precision. For example (https://github.com/OpenMined/PySyft/blob/master/examples/torch/Fixed%20Precision%20Tensor%20Testing.ipynb)

However, only addition has been implemented. In this work, we'd like to add the following functions.

• __mult__ - multiplication by another fixed precision tensor
• __mult__ - multiplication by another non-fixed precision tensor
• prod() - multiplication across a dimension
• cumprod() - cumulative multiplication across a dimension
• __add__ - addition by a non-fixed precision tensor
• sum() - addition across a dimension
• cumsum() - cumulative addition across a dimension
• inheritance - all undefined methods should inherit from the tensor underlying FixedPrecisionTensor (self.child). Test that this works using __getitem__

[jjmachan]

I would like to work on this. I'm just getting started with this project though.

[iamtrask]

Please reach out if you get stuck on anything - I'm here to help :)

[alhparsa]

I got this 😁