-
I was reading the maml example and saw this:
https://github.com/facebookresearch/higher/blob/941ae9f310994e728064f8edd7f17c57550e7c67/examples/maml-omniglot.py#L165
which seems right locally (si…
-
Can MAML interface work if I put pytorch LSTM module in the model?
-
Our documentation consists of "manual" written help and generated help.
From the manual written help we can add links to the topic pages in the generated documentation.
I would also like to add link…
-
Andrew R. raised this on discuss.ray.io:
```
I’ve noticed that several of the RLlib algorithms have a top-level config parameter named vf_share_layers.
However, vf_share_layers is also pres…
-
Hi, thank you for sharing your code.
In the conv_block() function in layers.py, tf.contrib.layers.batch_norm() is used without the is_training flag modification.
Since the flag controls to updat…
-
PlatyPS generates module-based XML help file names that don't work on systems < PS 4.0. It also doesn't add or mention the .ExternalHelp comment keyword that's required for functions in Get-Help 3.0.
…
juneb updated
3 years ago
-
It seems MAML is sort of unreliable/unstable. Is there a more stable implementation of it like MAML++?
-
All the cases you use in maml/datasets.py are downloaded datasets, and call torchmeta.datasets.Omniglot or MiniImagenet to define train/val/test data. Just wondering how can I define my own dataset to…
-
Thank you for the code for this article
I have a question. In the article, each episode only includes one task, but in the code, it seems to contain multiple tasks. Therefore, each epoch contains 100…
-