Implementation quick question

[macrocredit]

Hi - Great work!

I have one question:

loss, jvp = fc.jvp(f, (tuple(params),), (v_params,))

Do you know why jvp is a scalar? I would have thought that this is a matrix. Also, is there a reason why we are calling tuple(params) instead of params?

Thank you.

[FITZET]

i also found that the jvp is a scalar, and i'm not sure how the jvp is calculated. i want a formula to show the computing flow layer by layer, do you know how to get the formula?

[inikishev]

Hi - Great work!

I have one question:

loss, jvp = fc.jvp(f, (tuple(params),), (v_params,))

Do you know why jvp is a scalar? I would have thought that this is a matrix. Also, is there a reason why we are calling tuple(params) instead of params?

Thank you.

its dot product of jacobian with a random vector, which is the same as directional derivative in the direction of that vector. Params are a tuple because I believe jvp in pytorch requires a tuple