-
Hi Ian,
Can I modify the RNN to take in 2D series instead of 1D, is it feasible?
Or is there some other solution which think is possible, like concatenating the values somehow?
My current solution …
-
# Outline
## Transfer learning for echelle spectroscopy
- **Introduction**
- [ ] Spectral fitting past and present
- [ ] Ideal goal is both physical self-consistency and high predictive accura…
gully updated
2 years ago
-
Hi all,
I am currently reading your code. I am new to PyTorch and Computer. I am really like your work.
When I am reading the PyTorch online, I find that " torch.nn only supports mini-batches. T…
-
As discussed in #62 with @lacava (and discussed a bit in #24 by others last year), I think a wall clock time benchmark would be a really nice complement to comparing over a fixed number of evaluations…
-
**Describe the bug**
while running hivemind albert experiment, we have one monitor peer and two worker peers. One of the nodes is working fine
But the other peer is stack at downloading parameters…
-
To recap for other readers: it is currently possible to parallelize the optimization step by setting the (non-default) `n_sgd_threads > 1`. However, this is not ideal because the code is subject to ra…
-
At first add more tests!
Try to track down the bug.
-
In `foolbox/foolbox/attacks/gradient_descent_base.py `, `class BaseGradientDescent(FixedEpsilonAttack, ABC):`, there is `get_loss_fn` function.
def get_loss_fn(
self, model: Model, la…
-
I have a question regarding `SGD Optimizer`. Same issue has been raised in [Stack Overflow](https://stackoverflow.com/questions/67636925/need-clarification-regarding-sgd-optimizer) as well.
There a…
-
I can't remember where, but there was a recent PR offering cosmetic (e.g. PEP8) fixes, and we usually reject these PRs as causing merge conflicts with existing open PRs. @amueller wanted to know exact…