#With SplitCrossEntropyLoss parameters being updated 
CUDA_VISIBLE_DEVICES=3 python -u main.py --epochs 14 --nlayers 4 --emsize 400 --nhid 2500 --alpha 0 --beta 0 --dropoute 0 --dropouth 0.1 --dropouti 0.1 --dropout 0.1 --wdrop 0.5 --wdecay 0 --bptt 140 --batch_size 32 --optimizer adam --lr 1e-3 --data data/wikitext-103 --save WT103.12hr.QRNN.pt --when 12 --log-interval 10
Loading cached dataset...
Applying weight drop of 0.5 to weight_hh_l0
Applying weight drop of 0.5 to weight_hh_l0
Applying weight drop of 0.5 to weight_hh_l0
Applying weight drop of 0.5 to weight_hh_l0
[WeightDrop (
  (module): LSTM(400, 2500)
), WeightDrop (
  (module): LSTM(2500, 2500)
), WeightDrop (
  (module): LSTM(2500, 2500)
), WeightDrop (
  (module): LSTM(2500, 400)
)]
Using [2800, 20000, 76000]
Args: Namespace(alpha=0.0, batch_size=32, beta=0.0, bptt=140, clip=0.25, cuda=True, data='data/wikitext-103', dropout=0.1, dropoute=0.0, dropouth=0.1, dropouti=0.1, emsize=400, epochs=14, log_interval=10, lr=0.001, model='LSTM', nhid=2500, nlayers=4, nonmono=5, optimizer='adam', resume='', save='WT103.12hr.QRNN.pt', seed=1111, tied=True, wdecay=0.0, wdrop=0.5, when=[12])
| epoch   1 |    10/23041 batches | lr 0.00100 | ms/batch 1426.49 | loss 10.69 | ppl 43880.92 | bpc   15.421
| epoch   1 |    20/23041 batches | lr 0.00100 | ms/batch 936.87 | loss  8.09 | ppl  3266.35 | bpc   11.673
| epoch   1 |    30/23041 batches | lr 0.00100 | ms/batch 843.49 | loss  7.62 | ppl  2034.29 | bpc   10.990
| epoch   1 |    40/23041 batches | lr 0.00100 | ms/batch 794.54 | loss  7.63 | ppl  2067.34 | bpc   11.014
| epoch   1 |    50/23041 batches | lr 0.00100 | ms/batch 837.59 | loss  7.63 | ppl  2064.16 | bpc   11.011
| epoch   1 |    60/23041 batches | lr 0.00100 | ms/batch 825.59 | loss  7.60 | ppl  2007.57 | bpc   10.971
| epoch   1 |    70/23041 batches | lr 0.00100 | ms/batch 864.99 | loss  7.65 | ppl  2091.11 | bpc   11.030
| epoch   1 |    80/23041 batches | lr 0.00100 | ms/batch 870.91 | loss  7.59 | ppl  1983.57 | bpc   10.954
| epoch   1 |    90/23041 batches | lr 0.00100 | ms/batch 799.84 | loss  7.56 | ppl  1917.90 | bpc   10.905
| epoch   1 |   100/23041 batches | lr 0.00100 | ms/batch 845.25 | loss  7.60 | ppl  2001.65 | bpc   10.967
| epoch   1 |   110/23041 batches | lr 0.00100 | ms/batch 935.67 | loss  7.54 | ppl  1877.04 | bpc   10.874
| epoch   1 |   120/23041 batches | lr 0.00100 | ms/batch 825.33 | loss  7.54 | ppl  1879.97 | bpc   10.876
| epoch   1 |   130/23041 batches | lr 0.00100 | ms/batch 847.91 | loss  7.56 | ppl  1912.42 | bpc   10.901
| epoch   1 |   140/23041 batches | lr 0.00100 | ms/batch 841.25 | loss  7.53 | ppl  1853.87 | bpc   10.856