Inputset for table lookups about concrete-numpy HOT 4 CLOSED

Alright, thank you very much for your help!

from concrete-numpy.

Hello,

I tried to recreate the error you get but I couldn't. Here is what I did:

Created a new virtual env using:

$ python3.8 -m venv .venv

Activated it:

$ source .venv/bin/activate

Installed concrete-numpy:

$ pip install concrete-numpy

Here is the state of the environment:

$ pip freeze
concrete-compiler==0.6.0
concrete-numpy==0.5.0
cycler==0.11.0
fonttools==4.33.3
kiwisolver==1.4.3
loguru==0.5.3
matplotlib==3.5.2
networkx==2.8.4
numpy==1.23.0
packaging==21.3
Pillow==9.1.1
pyparsing==3.0.9
python-dateutil==2.8.2
PyYAML==6.0
six==1.16.0
torch==1.11.0
typing_extensions==4.2.0

Finally ran this code:

import concrete.numpy as hnp
import numpy as np

squared = hnp.LookupTable([i ** 2 for i in range(4)])
cubed = hnp.LookupTable([i ** 3 for i in range(4)])

table = hnp.MultiLookupTable([
    [squared, cubed],
    [squared, cubed],
    [squared, cubed],
])

def f(x):
    return table[x]

t_lookup_inputset = [
    np.array([[0, 0], [0, 1], [0, 2]], dtype=np.uint8),
    np.array([[1, 1], [1, 2], [1, 3]], dtype=np.uint8),
    np.array([[2, 1], [2, 2], [2, 3]], dtype=np.uint8),
]

compiler = hnp.NPFHECompiler(
    f, {"x": "encrypted"}
)
print('compiling the circuit')
table_circuit = compiler.compile_on_inputset(t_lookup_inputset)
print('compile done')

inp = np.array([[0, 1], [1, 1], [1, 2]], dtype=np.uint8)
print(table_circuit.encrypt_run_decrypt(inp))

This is the output I get:

compiling the circuit
compile done
[[0 1]
 [1 1]
 [1 8]]

Could you verify that it's working? And if it's not working could you share reproduction steps.

Thanks.

from concrete-numpy.

Hi Metu,

Thank you for your swift reply! I just checked again, and it seems to be working without issues. I am using jupyter notebooks, so maybe I had some artifacts left from the other things I have tried. A small additional question, how do I fix the

WARNING: You are currently using the software variant of concrete-csprng which does not have access to a hardware source of randomness. To ensure the security of your application, please arrange to provide a secret by using the `concrete_csprng::set_soft_rdseed_secret` function.

Since I am working in Python and this seems to be a Rust function?

Thank you!

Best,
Mirza

from concrete-numpy.

This warning could happen if you are a non-x86 architecture, say on Apple Silicon for example (that is my case). And in that case you cant use hardware acceleration and have slower performances. It is indeed in one of the Rust dependency of concrete-numpy.

More info on the issue here: https://community.zama.ai/t/support-for-hardware-source-of-randomness-in-docker-vm/44

from concrete-numpy.