Skip to Main Content
Table 5: 
Running times on development sets of the Tiger and the DPTB, reported in tokens per second (tok/s) and sentences per second (sent/s). Runtimes are only indicative; they are not comparable with those reported by other authors, since they use different hardware.
TigerDPTB
Modeltok/ssent/stok/ssent/s
This work, ml-gap, base 1,454 95 1,450 61 
This work, sr-gap-unlex, base 1,934 126 1,887 80 
Maier (2015), beam=8  80   
Coavoux and Crabbé (2017a), beam=4 4,700 260   
Corro et al. (2017)    ≈7.3 
TigerDPTB
Modeltok/ssent/stok/ssent/s
This work, ml-gap, base 1,454 95 1,450 61 
This work, sr-gap-unlex, base 1,934 126 1,887 80 
Maier (2015), beam=8  80   
Coavoux and Crabbé (2017a), beam=4 4,700 260   
Corro et al. (2017)    ≈7.3 
Close Modal

or Create an Account

Close Modal
Close Modal