| epoch: 1, batch: 100, avg loss: 1.035541, avg ranking loss: 0.016568, avg mle loss: 8.698632 | |
| learning rate: 0.000000 | |
| epoch: 1, batch: 200, avg loss: 0.939091, avg ranking loss: 0.015825, avg mle loss: 7.808369 | |
| learning rate: 0.000000 | |
| epoch: 1, batch: 300, avg loss: 0.786419, avg ranking loss: 0.015178, avg mle loss: 6.346376 | |
| learning rate: 0.000001 | |
| epoch: 1, batch: 400, avg loss: 0.653152, avg ranking loss: 0.014788, avg mle loss: 5.052757 | |
| learning rate: 0.000001 | |
| epoch: 1, batch: 500, avg loss: 0.583150, avg ranking loss: 0.014760, avg mle loss: 4.355531 | |
| learning rate: 0.000001 | |
| epoch: 1, batch: 600, avg loss: 0.531894, avg ranking loss: 0.014850, avg mle loss: 3.833926 | |
| learning rate: 0.000001 | |
| epoch: 1, batch: 700, avg loss: 0.496518, avg ranking loss: 0.014805, avg mle loss: 3.484637 | |
| learning rate: 0.000001 | |
| epoch: 1, batch: 800, avg loss: 0.470519, avg ranking loss: 0.014618, avg mle loss: 3.243440 | |
| learning rate: 0.000002 | |
| epoch: 1, batch: 900, avg loss: 0.449544, avg ranking loss: 0.014501, avg mle loss: 3.045354 | |
| learning rate: 0.000002 | |
| epoch: 1, batch: 1000, avg loss: 0.431474, avg ranking loss: 0.014440, avg mle loss: 2.870745 | |
| learning rate: 0.000002 | |
| best ranking loss - epoch: 0, batch: 999 | |
| val ranking loss: 0.729979 | |
| val ranking rouge1: 0.629180, rouge2: 0.334733, rougeLsum: 0.599455 | |
| best generation loss - epoch: 0, batch: 999 | |
| val generation loss: 0.885783 | |
| val generation rouge1: 0.321491, rouge2: 0.134002, rougeLsum: 0.299572 | |
| epoch: 1, batch: 1100, avg loss: 0.419558, avg ranking loss: 0.014534, avg mle loss: 2.742168 | |
| learning rate: 0.000002 | |
| epoch: 1, batch: 1200, avg loss: 0.405170, avg ranking loss: 0.014500, avg mle loss: 2.601675 | |
| learning rate: 0.000002 | |
| epoch: 1, batch: 1300, avg loss: 0.394226, avg ranking loss: 0.014297, avg mle loss: 2.512567 | |
| learning rate: 0.000003 | |
| epoch: 1, batch: 1400, avg loss: 0.388203, avg ranking loss: 0.014295, avg mle loss: 2.452513 | |
| learning rate: 0.000003 | |
| epoch: 1, batch: 1500, avg loss: 0.379525, avg ranking loss: 0.014269, avg mle loss: 2.368302 | |
| learning rate: 0.000003 | |
| epoch: 1, batch: 1600, avg loss: 0.375064, avg ranking loss: 0.014436, avg mle loss: 2.307035 | |
| learning rate: 0.000003 | |
| epoch: 1, batch: 1700, avg loss: 0.364328, avg ranking loss: 0.014073, avg mle loss: 2.235932 | |
| learning rate: 0.000003 | |
| epoch: 1, batch: 1800, avg loss: 0.362203, avg ranking loss: 0.014417, avg mle loss: 2.180354 | |
| learning rate: 0.000004 | |
| epoch: 1, batch: 1900, avg loss: 0.358076, avg ranking loss: 0.014373, avg mle loss: 2.143419 | |
| learning rate: 0.000004 | |
| epoch: 1, batch: 2000, avg loss: 0.351821, avg ranking loss: 0.014282, avg mle loss: 2.089963 | |
| learning rate: 0.000004 | |
| val ranking loss: 0.734460 | |
| val ranking rouge1: 0.620741, rouge2: 0.331641, rougeLsum: 0.590756 | |
| val generation loss: 0.885808 | |
| val generation rouge1: 0.314677, rouge2: 0.155852, rougeLsum: 0.293533 | |
| epoch: 1, batch: 2100, avg loss: 0.349724, avg ranking loss: 0.014392, avg mle loss: 2.058075 | |
| learning rate: 0.000004 | |
| epoch: 1, batch: 2200, avg loss: 0.343959, avg ranking loss: 0.014182, avg mle loss: 2.021375 | |
| learning rate: 0.000004 | |
| epoch: 1, batch: 2300, avg loss: 0.340133, avg ranking loss: 0.014330, avg mle loss: 1.968376 | |
| learning rate: 0.000005 | |
| epoch: 1, batch: 2400, avg loss: 0.336417, avg ranking loss: 0.014204, avg mle loss: 1.943732 | |
| learning rate: 0.000005 | |
| epoch: 1, batch: 2500, avg loss: 0.330869, avg ranking loss: 0.014164, avg mle loss: 1.892253 | |
| learning rate: 0.000005 | |
| epoch: 1, batch: 2600, avg loss: 0.329880, avg ranking loss: 0.014310, avg mle loss: 1.867807 | |
| learning rate: 0.000005 | |
| epoch: 1, batch: 2700, avg loss: 0.326743, avg ranking loss: 0.014328, avg mle loss: 1.834671 | |
| learning rate: 0.000005 | |
| epoch: 1, batch: 2800, avg loss: 0.323882, avg ranking loss: 0.014190, avg mle loss: 1.819801 | |
| learning rate: 0.000006 | |
| epoch: 1, batch: 2900, avg loss: 0.320222, avg ranking loss: 0.013906, avg mle loss: 1.811585 | |
| learning rate: 0.000006 | |
| epoch: 1, batch: 3000, avg loss: 0.317352, avg ranking loss: 0.014259, avg mle loss: 1.747636 | |
| learning rate: 0.000006 | |
| val ranking loss: 0.736030 | |
| val ranking rouge1: 0.617348, rouge2: 0.330589, rougeLsum: 0.587821 | |
| best generation loss - epoch: 0, batch: 2999 | |
| val generation loss: 0.884436 | |
| val generation rouge1: 0.314360, rouge2: 0.161066, rougeLsum: 0.296060 | |