Tae Hwan Jung

Added ppt Attachment

1

Added power point attachment to make modifications and deletions free :D As long as you leave the source, you are free to cite and edit it. #### License

Comparison with HDF5(pytables)

4

I compared numpy array of shapes (200000, 784) with the use of a dense-array in tileDB and the 'create_array' of pytable. However, the speed of the tileDB's I/O is significantly...

Hello Authors! Thanks for sharing nice work. Your code are awesome. I try your emoji_hunt demo with my trained model. I think this is important issue to implement Ferderated Learning....

loss don't go down when finetune and about requires_grad option in backward()

3

I tried finetuning with SS-Youtube dataset on `examples/finetune_youtube_last`. but There are some minor errors in there :(... First. The loss are not decrease well about each epochs when finetune using...

I will write test code related to this and conduct experiments. To implement it correctly, you should check https://github.com/graykode/matorage/blob/master/matorage/uploader.py#L41 to see if the upload is a success(200) code and put...

metadata file change from FileStorage to Database

5

Now metadata attribute is managed as a JSON file. However, as a long-term plan, we will modify it to work concurrency with the database. e.q : data ```json { "additional":...

support inference large models such as gpt-3 in storage calculation.

1

In deep learning, the popularity of large models (gpt-3, T5, megatron LM) is growing. However, due to this, the polarization of wealth in AI is intensifying. As an example that...

This is Worked ```shell pip install twkorean pip uninstall --yes twkorean git clone https://github.com/jaepil/twkorean cd twkorean python setup.py install cd .. ```

In ALBERT(Lan at el), There is not detail about 80% mask  But, from n-gram masking (Joshi et al., 2019), they said about 80/10/10 > As in BERT, we also...