test/fastNLP
1
0
Fork 0
mirror of https://gitee.com/fastnlp/fastNLP.git synced 2024-11-29 18:59:01 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

.Jenkinsfile 添加 shm-size 设置；test_seq2seq_model 调整训练参数

Browse Source
This commit is contained in:
x54-729 2022-09-28 13:47:26 +08:00
parent 4775581084
commit f090e4a99c
2 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
.Jenkinsfile
View File

@ -46,7 +46,7 @@ pipeline {
agent { agent {
docker { docker {
image 'fnlp:torch-1.6' image 'fnlp:torch-1.6'
args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all' args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all --shm-size 1G'
} }
} }
steps { steps {
@ -62,7 +62,7 @@ pipeline {
agent { agent {
docker { docker {
image 'fnlp:paddle' image 'fnlp:paddle'
args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all' args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all --shm-size 1G'
} }
} }
steps { steps {
@ -82,7 +82,7 @@ pipeline {
// agent { // agent {
// docker { // docker {
// image 'fnlp:jittor' // image 'fnlp:jittor'
// args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all' // args '-u root:root -v ${JENKINS_HOME}/html/docs:/docs -v ${JENKINS_HOME}/html/_ci:/ci --gpus all --shm-size 1G'
// } // }
// } // }
// steps { // steps {

4
tests/models/torch/test_seq2seq_model.py
View File

@ -25,11 +25,11 @@ def prepare_env():
def train_model(model, src_words_idx, tgt_words_idx, tgt_seq_len, src_seq_len): def train_model(model, src_words_idx, tgt_words_idx, tgt_seq_len, src_seq_len):
optimizer = optim.Adam(model.parameters(), lr=1e-2) optimizer = optim.Adam(model.parameters(), lr=5e-3)
mask = seq_len_to_mask(tgt_seq_len).eq(0) mask = seq_len_to_mask(tgt_seq_len).eq(0)
target = tgt_words_idx.masked_fill(mask, -100) target = tgt_words_idx.masked_fill(mask, -100)
for i in range(100): for i in range(50):
optimizer.zero_grad() optimizer.zero_grad()
pred = model(src_words_idx, tgt_words_idx, src_seq_len)['pred'] # bsz x max_len x vocab_size pred = model(src_words_idx, tgt_words_idx, src_seq_len)['pred'] # bsz x max_len x vocab_size
loss = F.cross_entropy(pred.transpose(1, 2), target) loss = F.cross_entropy(pred.transpose(1, 2), target)