NLP - POS tagger with Perceptron algorithm

Code for assignments from prof. Michael Collin's NLP course, COMSW4705 Spring 2015 at Columbia U.

Use the shell script hw4.sh to run the python scripts for questions 1-4.

Runtime ~3 minutes

It also runs eval_tagger.py for each of the output files generated:

For q4: q4_output

For q5: q5_output

For q6: q6_output_combo1 q6_output_combo2 q6_output_combo3 q6_output_combo4

Note that many temporary files are created then deleted

See report.pdf for description and instructions.

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commits
README.md		README.md
eval_tagger.py		eval_tagger.py
hw4.sh		hw4.sh
question4.py		question4.py
question5.py		question5.py
question6.py		question6.py
report.pdf		report.pdf
suffix_tagger.model		suffix_tagger.model
tag.model		tag.model
tag_dev.dat		tag_dev.dat
tag_dev.key		tag_dev.key
tag_train.dat		tag_train.dat
tagger_config.py		tagger_config.py
tagger_config.pyc		tagger_config.pyc
tagger_decoder.py		tagger_decoder.py
tagger_history_generator.py		tagger_history_generator.py
utils.py		utils.py
utils.pyc		utils.pyc

jlouback/nlp-perceptron