Transfer Learning library for Deep Neural Networks.

Dear authors,

According to your ICLR paper Empirical Bayes Transductive Meta-Learning with Synthetic Gradients, it appears p(w) is adjustable and is trained to achieve empirical bayes. However, in your code, sib.py line 29, I belive the p(w) is fixed as a zero-mean gaussian. Please correct me if I were wrong, but how is this implementation achieving empirical bayes?

Thank you in advance!

Issue #, if available:

Description of changes: Replaced Layer Factory with use of mxnet.symbol.load_json()

• Removed layer factory files and associated test file.
• Updated notebooks to fit new API

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Is your feature request related to a problem? Please describe. I would like to compare two simple vanilla CNN models trained on different datasets. How much changes do I need to do in the current code.

Describe the solution you'd like Comparing two simple vanilla CNN models trained on different datasets (e.g., MNIST and SVHN)

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context Add any other context or screenshots about the feature request here.

Issue #, if available:

Description of changes: We are phasing out gpyopt in favor of emukit. This PR replaces gpyopt with emukit in xfer-hpo notebook.

This PR replaces https://github.com/amzn/xfer/pull/59 which was done against a wrong branch

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Issue #, if available: n/a

Description of changes: We are phasing out gpyopt in favor of emukit. This PR replaces gpyopt with emukit in xfer-hpo notebook.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Github apparently has a bug in its markdown rendering when it won't properly show links that are not http or https. Example of where this is discussed: https://github.com/gollum/gollum/issues/759

The first link in the README.md is ftp so is affected. As a solution, the link could be to the web page that references the actual paper: http://pages.cs.wisc.edu/~shavlik/abstracts/torrey.handbook09.abstract.html

Bumps nbconvert from 5.3.1 to 6.3.0.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Issue #, if available: #71

Description of changes: Move travis config file and update to work with xfer-ml in subdirectory.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Hi,

MXNet nightly builds have been moved from PyPI to S3 (https://github.com/pypa/pypi-support/issues/243). If this project requires the nightly builds for CI validation, please make sure that your CI pipeline now takes the nightly builds from https://dist.mxnet.io/python. Furthermore, as MXNet 1.7 and 2.0 are currently in progress simultaneously, we are releasing nightly builds for both 1.x and 2.x. Take mxnet-cu102 for example:

If you need the nightly builds for 1.x:

pip install --pre "mxnet-cu102<2" -f https://dist.mxnet.io/python

If you need the nightly builds for 2.x:

pip install --pre "mxnet-cu102>=2" -f https://dist.mxnet.io/python

Feel free to close this issue if the necessary changes have been made, or If the project doesn't rely on MXNet nightly builds. Thanks.

Issue #, if available:

Description of changes: Moved xfer/xfer-ml/contrib/leap into xfer/leap. The reason for that is that the restructure has all files in flat structure under xfer, instead of inside xfer-ml (which makes sense since e.g. leap does not have dependencies in xfer-ml).

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Issue #, if available:

Description of changes: Summary: Restructuring xfer such that it becomes a more general Python repository for transfer and meta-learning. The previous state of xfer now becomes a sub-folder named xfer-ml which is only one of the tools contained in the now more general project xfer. Currently xfer-ml is the only library, but soon more will be added (hence the need for restructuring).

Details:

• Created xfer-ml subfolder
• Moved all files and folders (apart from NOTICE and CODE_OF_CONDUCT) inside xfer-ml with no further changes
• Re-wrote top-level README.md and xfer-ml/README.md to reflect the new nature of the structure.

In essence, only the new README's have content changes, the rest of the changes is just files moved to the new xfer-ml folder.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Bug

I think test_year is supposed to be train_year in line 55 right?

https://github.com/amzn/xfer/blob/dd4a6a27ca00406df83eec5916d3a76a1a798248/finite_ntk/data.py#L40-L55

Questions

I see the variables inside, extent and grid_x being declared but not used in the malaria experiments. I'm looking to replicate the experiments in JAX so I was wondering what the original purpose of these variables were. In particular, what is the sparse tensor for marking Nigeria and inside supposed to be doing?

https://github.com/amzn/xfer/blob/dd4a6a27ca00406df83eec5916d3a76a1a798248/finite_ntk/data.py#L80-L100

P.S. thank you for the work and for releasing the code!

Adds the code to reproduce Figure 7b (olivetti dataset) and Table 1 (unsupervised to supervised experiments) of https://arxiv.org/pdf/2103.01439.pdf that had not been included in the public version previously.

Let me know if I still need to modify licenses / attribution here.

@shuaitang @pgmoren

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

Maybe I am missing something. I don't see where current_pars argument is used inside the losses. Is it used, if not I guess why it is passed?

https://github.com/amzn/xfer/blob/dd4a6a27ca00406df83eec5916d3a76a1a798248/finite_ntk/experiments/cifar/losses.py#L22

Hello, a great work! Thank you for sharing the codes. I to begin the Step 2 in README : python main.py --config config/miniImageNet_1shot.yaml --seed 100 --gpu 0

Firstly, I congfig the miniImageNet_1shot.json file path in config.py like this:

def get_args(): """ Create argparser for frequent configurations.

:return: argparser object
"""
argparser = argparse.ArgumentParser(description=__doc__)
argparser.add_argument(
    '-c', '--config',
    metavar='C',
    default="/home/dy/PP/FSL/sib_meta_learn/data/Mini-ImageNet/val1000Episode_5_way_5_shot.json",
    help='The Configuration file')

Then, I run main,py, there's an error like this:

Would you please help me solve the problem? Appreciate!

Issue #, if available:

Description of changes:

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.