-
Notifications
You must be signed in to change notification settings - Fork 2
Running BIUTEE
BIUTEE runs in three main phases:
- Preprocessing (LAP)
- Training
- Testing
BIUTEE can then train a model from the preprocessed training data. This model is then used by BIUTEE during inference (the testing phase). It is important to note that BIUTEE can be tested in a certain configuration only when using a model that was trained under that exact configuration (same knowledge resources, optimization, etc).
BIUTEE also has an additional running mode - the visual tracing tool.
In Linux/Unix, BIUTEE can be run from the command line using run_biutee.sh from $BIUTEE/workdir. Running with -h reveals the modes and options:
run_biutee.sh [OPTIONS] MODE Available modes: lap_train train lap_train,train lap_test test lap_test,test visual -i <filename> raw input file (rte pairs format) -s <filename> serialized input file (saves/loads the preprocessed pairs after LAP) -m <dir> models directory -o <filename> results file (only in test modes) -r <list> list of resources, separated by commas (no spaces) -t <int> number of threads -e <port> easyfirst parser's port -f optimize F1 (instead of accuracy) -g gap mode
For first-time use, we recommend running lap_train,train followed by lap_test,test. See how to upgrade BIUTEE and BIUTEE's knowledge resources for more information on -r.
BIUTEE comes along with a few pre-trained models. They are all located under $BIUTEE/workdir/models/. The following table describes how each one was trained.
Model | Training Data | Resources |
---|---|---|
basic | RTE3 Dev Set | WORDNET,CATVAR,SYNTACTIC |
basic_bap | RTE3 Dev Set | WORDNET,CATVAR,SYNTACTIC,REDIS_BAP |