-
Notifications
You must be signed in to change notification settings - Fork 538
[Enhancement] BERT pre-training data generation from sentencepiece vocab #743
Conversation
Codecov Report
|
Codecov Report
@@ Coverage Diff @@
## master #743 +/- ##
=========================================
- Coverage 90.58% 90.49% -0.1%
=========================================
Files 66 66
Lines 6121 6120 -1
=========================================
- Hits 5545 5538 -7
- Misses 576 582 +6
|
Job PR-743/1 is complete. |
@davisliang FYI |
Job PR-743/3 is complete. |
Job PR-743/4 is complete. |
Job PR-743/7 is complete. |
Job PR-743/8 is complete. |
Job PR-743/9 is complete. |
Job PR-743/10 is complete. |
Job PR-743/11 is complete. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice work! Some comments.
Job PR-743/12 is complete. |
Job PR-743/13 is complete. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice work! Looks good to me!
|
||
Run pre-training with horovod on node0 and node1, with 8 GPUs each: | ||
|
||
$ mpirun -np 16 -H node0:8,node1:8 -mca pml ob1 -mca btl ^openib -mca btl_tcp_if_exclude docker0,lo --map-by ppr:4:socket -x NCCL_MIN_NRINGS=8 -x NCCL_DEBUG=WARNING -x HOROVOD_HIERARCHICAL_ALLREDUCE=1 --tag-output python run_pretraining_hvd.py --batch_size 8192 --accumulate 1 --lr 1e-4 --data "/path/to/generated/samples/train/*.npz" --warmup_ratio 0.01 --num_steps 1000000 --log_interval=250 --ckpt_dir './ckpt' --ckpt_interval 25000 --num_buckets 10 --dtype float16 --use_avg_len --verbose |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This line is raising warnings when generating the documentations: Inline emphasis start-string without end-string.
(warnings are treated as fatal in our CI setup)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks. I split them into two separate blocks with code-block
annotations
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
Job PR-743/15 is complete. |
…cab (dmlc#743) * enable fp16 for ln. enable gelu * manage processes * manual killl * support sentencepiece * dont register signal handler * add unigram sampling * support comma for npz format * fix file race condition * switch to thread prefetcher * fix pool.apply * revert gelu support * update documentation * code cleanup * avoid file download conflcit * update doc * fix bug * bug fix * more multi-processing * fix lint * update doc and fix lint * fix test argument * remove -test_bert_sentencepiece_sentences_transform() * bug fix * fix lint * fix doc build
Description
Checklist
Essentials
Changes
Comments