Metalearning

Stanford metalearning course Facebook package with metalearning utilities in pytorch Chelsea F...

This describes efforts to work on integration of DAML into the existing content scoring pipeline ...

import torch import random import numpy as np from config import global_config as cfg from re...

The main point of interest is train_maml in model.py. prev_min_loss gets set to a high value on ...

The DAML implementation uses this filter when constructing the optimizer Adam(lr = self.meta_lr, ...

Done Switch optimizer over to the ifltered optimizer. todo determine if self.model needs to be m...

2-21-2020: Finished a bunch of updates to the outer loop of maml to make it better reflect the d...

Allennlp has some code in tests that is gearing for multitask setups. Ideally, we'd like to be co...

https://github.com/allenai/allennlp/blob/master/allennlp/tests/training/multi_task_trainer_test.py

data constructors verify that the metatrainer constructor is being initialized correctly data s...

The BertRegressor model is now called "general_regressor" Doesn't seem that transition away from...

Current status: Multi-task training is working using interleaving dataset reader, off the shelf ...

Clear Search