Code submitted as mini project for partial fullfilment of M.Tech degree.
Ensure that you have python2/3.
This project is organized into 3 folders literature
, logs
, report_ppt
and 5 files gru.py
, main.py
, pos_tagger.py
, train.py
, TreeBankDataSet.py
PosTagging
├── gru.py
├── literature
│ └── Empirical Analysis of RNN.pdf
├── logs
│ └── logs_400_400_0.1_10_40_s_Adam_treebank_nll_0.8_file.log
├── main.py
├── pos_tagger.py
├── README.md
├── report_ppt
│ ├── POS Tagging using GRU.pdf
│ ├── ppt.pdf
│ └── report_sc18m002_pos_tagging.lyx~
├── train.py
└── TreeBankDataSet.py
literature
folder contains the base research paper from which the idea is inspired.
logs
folder contains generated logs for experiment.
report_ppt
folder contains final report and presentation.
This project requires an NVIDIA-GPU with CUDA 9.0 to run PyTorch with GPU computing capabilities.
We have used TreeBankDataSet
for the expermients.
main.py
is starting script for running the experimenttrain.py
is training scriptpos_tagger.py
architecture of generic pos_taggergru.py
, contains architecture of GRU cell
Start the training process by calling main.py