moose
moose copied to clipboard
tf-encrypted
Secure distributed dataflow framework for encrypted machine learning and data processing
When we presented the whitepaper to the customer, there was a few pieces of feedback from the team who reviewed it. That feedback is as follows: - Add information about...
Currently we need to specify for all axes how we do slicing. In numpy, and other higher level languages one can simply do `x[:1]`. We'd like add the numpy slicing...
Currently we only support slicing for `HostRingTensor` or native `HostTensor`. This issue is relating to add support for slicing `HostBitTensor(BitArray)`. One can solve it by borrowing the slicing implementation from...
In https://github.com/tf-encrypted/moose/pull/1051, we have implemented an initial implementation of square root using `log2` and `pow2`. Instead we would you to implement the algorithm described in the whitepaper which is similar...
In the last few months, we have added lots of operation. We would like to assess the completeness for each operations in terms of types we support. [TODO] give more...
We see that ~7% for the generated LLVM code is related to `rmp_serde` ``` 357926 (6.1%) 258 (0.2%) ::deserialize_any 139718 (2.4%) 4356 (3.9%) core::ops::function::FnOnce::call_once 104948 (1.8%) 1484 (1.3%) alloc::alloc::box_free 85110...
this includes: - Adding `edsl.fill` and/or `edsl.zeros` - Supporting Fill, Zeros, & Ones on Replicated (and/or Mirrored) placements - Supporting these ops for the various missing types/dtypes
Metadata
Owner
Metadata
Secure distributed dataflow framework for encrypted machine learning and data processing