Tejas Vaidhya

- I think the addition of a Contact Form is a great addition may be at the end of about_me section - maybe keeping the users on the same page...

Reason: Moving of codebase to `torchtext.legacy` in v0.9.0 source: [link](https://pypi.org/project/torchtext/) in section [BC Breaking]

Dependencies - Transformers.jl , WordTokenizer.jl [**Pretrained weights**](https://drive.google.com/drive/folders/1HHTlS_jBYRE4cG0elITEH7fAkiNmrEgz?usp=sharing) . - The pretrained weights are converted from the TensorFlow checkpoint released by google-research. - The code for conversion is given in `tfckpt2bsonforalbert.jl`...

During Training on CoNLL 2003 Dataset (with around 203k POS tagged word ) the program is getting terminated. Probably because of excessive memory allocation. Working on Better implementation Training code...

- This implementation is the same as `Transformers.Bert` with a tiny embeddings tweaks. - [RoBERTa](https://arxiv.org/abs/1907.11692) has the same architecture as BERT, but uses a byte-level BPE(implemented in `BPE.jl`) as a...

[WIP] Albert

10

Hi everyone I am adding **ALBERT [WIP]** Currently only raw code is given in PR. Dependencies - **Transformers.jl , WordTokenizer.jl** I am not exporting any function.I am still in middle...

Adding gmb dataset

3

Adding **GMB** Dataset. The dataset an extract from GMB corpus which is tagged, annotated and built specifically to train the classifier to predict named entities such as name, location, etc.