未验证 提交 26dbcf52 编写于 作者: Q Qinghe JING 提交者: GitHub

Merge pull request #67 from qjing666/v0.2.0

update submitter demo
...@@ -38,7 +38,7 @@ In PaddleFL, components for defining a federated learning task and training a fe ...@@ -38,7 +38,7 @@ In PaddleFL, components for defining a federated learning task and training a fe
#### Compile Time #### Compile Time
- **FL-Strategy**: a user can define federated learning strategies with FL-Strategy such as Fed-Avg[1] - **FL-Strategy**: a user can define federated learning strategies with FL-Strategy such as Fed-Avg[2]
- **User-Defined-Program**: PaddlePaddle's program that defines the machine learning model structure and training strategies such as multi-task learning. - **User-Defined-Program**: PaddlePaddle's program that defines the machine learning model structure and training strategies such as multi-task learning.
......
...@@ -35,7 +35,7 @@ PaddleFL是一个基于PaddlePaddle的开源联邦学习框架。研究人员可 ...@@ -35,7 +35,7 @@ PaddleFL是一个基于PaddlePaddle的开源联邦学习框架。研究人员可
#### 编译时 #### 编译时
- **FL-Strategy**: 用户可以使用FL-Strategy定义联邦学习策略,例如Fed-Avg[1]。 - **FL-Strategy**: 用户可以使用FL-Strategy定义联邦学习策略,例如Fed-Avg[2]。
- **User-Defined-Program**: PaddlePaddle的程序定义了机器学习模型结构和训练策略,如多任务学习。 - **User-Defined-Program**: PaddlePaddle的程序定义了机器学习模型结构和训练策略,如多任务学习。
......
# commonly configured # commonly configured
task_name=test_fl_job_submit_jingqinghe task_name=test_name
hdfs_output=/user/feed/mlarch/sequence_generator/dongdaxiang/job_44 hdfs_output=/path/to/your/output/file
train_cmd=python dist_trainer.py train_cmd=python dist_trainer.py
#monitor_cmd=python system_monitor_app.py 10 100 #monitor_cmd=python system_monitor_app.py 10 100
monitor_cmd= monitor_cmd=
#train_cmd=python test_hadoop.py #train_cmd=python test_hadoop.py
hdfs_path=afs://xingtian.afs.baidu.com:9902 hdfs_path=/site/of/hdfs
ugi=your_fs_name, your_ugi ugi=your_fs_name, your_ugi
hdfs_output=/user/feed/mlarch/sequence_generator/dongdaxiang/job_44
worker_nodes=2 worker_nodes=2
server_nodes=1 server_nodes=1
hadoop_home=/home/jingqinghe/hadoop-xingtian/hadoop hadoop_home=/local/path/to/your/hadoop
hpc_home=/home/jingqinghe/mpi_feed4/smart_client hpc_home=/local/path/to/your/hpc
package_path=./package package_path=./package
priority=high priority=high
#queue name #queue name
queue=paddle-dev-amd queue=queue_name
server=yq01-hpc-lvliang01-smart-master.dmop.baidu.com server=/site/of/mpi/server
python_tar=./python.tar.gz python_tar=/path/to/your/executable/python
wheel=./paddlepaddle-0.0.0-cp27-cp27mu-linux_x86_64.whl wheel=/path/to/paddle/installation/package
tar -xf python.tar.gz tar -xf python.tar.gz
python/bin/python scheduler_client.py conf.txt python/bin/python scheduler_client.py config.txt
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册