wassname/multifit
1
0
Fork 0
mirror of https://github.com/wassname/multifit.git synced 2026-09-09 11:27:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add experiment where wd_true is set to True.

Browse Source
This commit is contained in:
Piotr Czapla
2018-11-19 15:30:28 +01:00
parent 7f1f8efcc3
commit 9f526a9849
1 changed files with 24 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
results/lm_qrnn-wt-103-unk-wd1
+24
View File
@@ -0,0 +1,24 @@
ime python -m ulmfit.pretrain_lm data/wiki/wikitext-103-unk --qrnn=True --cuda-id=0 --name=wt-103-wd1 --num_epochs=10 --bs=32
Batch size: 32
Max vocab: 60000
Using QRNNs...
Saving vocabulary as data/wiki/wikitext-103-unk/models
Size of vocabulary: 60001
First 10 words in vocab: the, <pad>, ,, ., of, and, to, in, <eos>, a
true_wd: True
Starting from random weights
epoch train_loss valid_loss accuracy
1 4.355600 4.271729 0.306827
2 4.384312 4.288582 0.299482
3 4.527577 4.366407 0.289836
4 4.519645 4.387311 0.288076
5 4.514816 4.371242 0.289862
6 4.506230 4.337718 0.294208
7 4.444272 4.295349 0.298786
8 4.432104 4.257432 0.301785
9 4.416849 4.242803 0.304147
10 4.389035 4.239752 0.304871
Saving models at data/wiki/wikitext-103-unk/models
Saving optimiser state at data/wiki/wikitext-103-unk/models/qrnn3_wt-103-wd1_state.pth
itos_fname: data/wiki/wikitext-103-unk/models/itos_wt-103-wd1.pkl
accuracy: tensor(0.2829)