-
For the [MAML reproducing](https://github.com/learnables/learn2learn/tree/master/examples/vision), the document says that **Note that the original MAML paper trains with 5 fast adaptation step, but te…
-
Very insightful paper! Still I have a question about the ant-direction task. In the pearl code, the survive_reward is set as 1.0 in ant_dir.py. But it seems that in the original MAML implementation, t…
ylfzr updated
3 years ago
-
In MAML we uses "Higher" module which is a library providing support for higher-order optimization developed by Facebook.
https://github.com/facebookresearch/higher
It turns existing `torch.nn.Modul…
-
- [x] Introduction
- [x] MAML
- [x] First-Order
- [x] iMAML
- [x] Comparison
-
Hi, I was trying to run the example of maml machine learning package notebooks/pes/gap example (https://github.com/materialsvirtuallab/maml/tree/master/notebooks/pes/gap) which calls gap_fit program, …
-
### Describe problem
The `maml` documentation must be added to the project of the `Sandcastle Help File Builder`.
The word documentation keep as is.
### Describe the solution you'd like
Ther…
-
I'm trying out the sample shown [here](https://docs.microsoft.com/en-gb/dotnet/machine-learning/tutorials/sentiment-analysis). However, whenever I try to train the model I get an error: "The size of i…
-
Hi,
I run the base model by the following command:
> python experiments/spider_dg/run.py train configs/spider/run_config/run_en_spider_bert_baseline.jsonnet
After 20K training steps, I got t…
-
- [x] task loss spaces on top
- [x] meta loss space for 0 steps (aka sum)
- [x] meta loss space for n > 0 steps (aka maml)
Optional
- [x] possible to increase steps dynamically (svgs/contours)
-
- [x] Futher readings
- [x] Refer to code again
- [x] short conclusion
- [x] ~Make maml-tf2 repository public~