This is the replication package of our work SemParser: A semantic parser for log analysis published in Proceedings of the 45th International Conference on Software and Engineering (ICSE 2023).
- Data. We use
/data
to evaluate the semantics mining ability. For downstream tasks evaluation, we use existing datasets: HDFS and OpenStack. - Model. The implementation of SemParser is described in
/model
.
- Download the word embedding file and put in under
/model
folder. - Execute
python main.py
We put our semantics mining experiment results under the folder /data/prediction
and checkpoints under model/ckpt
.