Issue/3 - #8
Open
rodrigo-castellon wants to merge 38 commits into
Open
Conversation
…mmit last time, did not track that change)
…ive method previously considered
Owner
|
Seems like you went well beyond data loading! The code looks clear and very well documented. I had some minor comments. I haven't run the training myself yet, but it'd be very interesting to see how the multiple objects (right now, simply concatenated) are processed. Let's talk soon. Great work! |
jiwoncpark
requested changes
Apr 9, 2020
| # Instantiate posterior (for logging) | ||
| #post = getattr(inference.posterior, loss_fn.posterior_name)(val_data.Y_dim, device, valset.train_Y_mean, valset.train_Y_std) | ||
| # Instantiate model | ||
| net = getattr(models, cfg.model.architecture)(input_size=59) |
Owner
There was a problem hiding this comment.
Let's get the input size from the data
|
|
||
| __all__ = ['sigmoid'] | ||
|
|
||
| def sigmoid(x): |
Owner
There was a problem hiding this comment.
def sigmoid(x):
return np.where(x >= 0,
1 / (1 + np.exp(-x)),
np.exp(x) / (1 + np.exp(x)))```
| # Instantiate loss function | ||
| loss_fn = getattr(losses, cfg.model.likelihood_class)() | ||
| # Instantiate posterior (for logging) | ||
| #post = getattr(inference.posterior, loss_fn.posterior_name)(val_data.Y_dim, device, valset.train_Y_mean, valset.train_Y_std) |
Owner
There was a problem hiding this comment.
You can clean up all the h0rton code
| # Declare all collate functions below # | ||
| ####################################### | ||
|
|
||
| def collate_RNN(batch): |
Owner
There was a problem hiding this comment.
Maybe come up with a basic unit test?
| def __len__(self): | ||
| return self.n_data | ||
| def __getitem__(self, idx): | ||
| return torch.FloatTensor(self.X[idx]), torch.FloatTensor([self.Y[idx]]) |
Owner
There was a problem hiding this comment.
Again, a basic unit test here would be helpful (just to check the dimensions of the data and each batch, etc.)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Set up the data pipeline for an RNN (with some extensibility to other models); RNN can now train with no problem. Still, however, may need some future work done in the realm of manipulating the (ra, dec) point coordinates into better representations of position (PointNet?) or processing them for specific network architectures (graph net, for example).