hidden-markov
The model parameters as trained by the program will be stored in model_params.csv, in the following format:
| type | first_term | second_term | value |
|---|---|---|---|
| e | x | y | e( x given y ) |
| q | y(i) | y(i-1) | q( y(i) given y(i-1) ) |
The training data will also be evaluated, and stored in eval_trgdata.csv in the following format:
| type | first_term | second_term | value |
|---|---|---|---|
| e | x | y | count( y -> x ) |
| q | y(i) | y(i-1) | count( y(i-1) -> y(i) ) |
| (none) | y | (none) | count( y ) |
To run challenge: python run_viterbi_optimizer.py -i data/combined_train -ti data/test.in -to test.p3.out