FATE-Flow Client Command Line Interface v2 Guide¶
[中文]
Usage¶
Before using fate flow client command line interface (CLI), please make sure that you have activated the virtual environment of FATE. For more details about how to activate virtual environment, please read the documentation of deployment.
In this version of client CLI, commands are separated into several classes, including job, data, model, component and etc. And all of these classes have a common parent (CLI entry) named ‘flow’, which means you can type ‘flow’ in your terminal window to find out all of these classes and also their sub-commands.
[IN]
flow
[OUT]
Usage: flow [OPTIONS] COMMAND [ARGS]...
Fate Flow Client
Options:
-h, --help Show this message and exit.
Commands:
component Component Operations
data Data Operations
job Job Operations
model Model Operations
queue Queue Operations
table Table Operations
task Task Operations
For more details, please check this documentation or try flow --help
for help.
Init¶
init
¶
Description: Flow CLI Init Command. Custom can choose to provide an absolute path of server conf file, or provide ip address and http port of a valid fate flow server. Notice that, if custom provides both, the server conf would be loaded in priority. In this case, ip address and http port would be ignored.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
No |
Server configuration file absolute path |
2 |
ip |
|
No |
Fate flow server ip address |
|
3 |
port |
|
No |
Fate flow server port |
|
4 |
reset |
|
No |
If specified, initialization settings of flow CLI would be reset to none. |
Examples:
flow init -c /data/projects/fate/python/conf/service_conf.yaml
flow init --ip 127.0.0.1 --port 9380
Job¶
submit
¶
Description: Submit a pipeline job.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Runtime configuration file path |
2 |
dsl_path |
|
|
Yes |
Domain-specific language(DSL) file path. If the type of job is ‘predict’, you can leave this feature blank, or you can provide a valid dsl file to replace the one that aotumatically generated by fate. |
Examples:
flow job submit -c fate_flow/examples/test_hetero_lr_job_conf.json -d fate_flow/examples/test_hetero_lr_job_dsl.json
stop
¶
Description: Cancel or stop a specified job.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
Examples:
flow job stop -j $JOB_ID
query
¶
Description: Query job information by filters.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
No |
A valid job id. |
2 |
role |
|
|
No |
Role |
3 |
party_id |
|
|
No |
Party ID |
4 |
status |
|
|
No |
Job Status |
Examples:
flow job query -r guest -p 9999 -s complete flow job query -j $JOB_ID
view
¶
Description: Query data view information by filters.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
No |
A valid job id. |
2 |
role |
|
|
No |
Role |
3 |
party_id |
|
|
No |
Party ID |
4 |
status |
|
|
No |
Job Status |
Examples:
flow job view -r guest -p 9999 flow job view -j $JOB_ID -s complete
config
¶
Description: Download the configuration of a specified job.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
output_path |
|
|
Yes |
Output Path |
Examples:
flow job config -j $JOB_ID -r host -p 10000 --output-path ./examples/
log
¶
Description: Download log files of a specified job.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
2 |
output_path |
|
|
Yes |
Output Path |
Examples:
flow job log -j JOB_ID --output-path ./examples/
list
¶
Description: List jobs.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
limit |
|
|
No |
Number of records to return. (default: 10) |
Examples:
flow job list
flow job list -l 30
dsl
¶
Description: A predict dsl generator.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
cpn_list |
|
No |
User inputs a string to specify component list. |
|
2 |
cpn_path |
|
No |
User specifies a file path which records the component list. |
|
3 |
train_dsl_path |
|
Yes |
User specifies the train dsl file path. |
|
4 |
output_path |
|
|
No |
User specifies output directory path. |
5 |
version |
|
|
No |
User specified version of dsl parser. Default: 1 |
Examples:
flow job dsl --cpn-path fate_flow/examples/component_list.txt --train-dsl-path fate_flow/examples/test_hetero_lr_job_dsl.json --version 2
flow job dsl --cpn-path fate_flow/examples/component_list.txt --train-dsl-path fate_flow/examples/test_hetero_lr_job_dsl.json -o fate_flow/examples/
flow job dsl --cpn-list "dataio_0, hetero_feature_binning_0, hetero_feature_selection_0, evaluation_0" --train-dsl-path fate_flow/examples/test_hetero_lr_job_dsl.json -o fate_flow/examples/
flow job dsl --cpn-list [dataio_0,hetero_feature_binning_0,hetero_feature_selection_0,evaluation_0] --train-dsl-path fate_flow/examples/test_hetero_lr_job_dsl.json -o fate_flow/examples/
Component (TRACKING)¶
parameters
¶
Description: Query the arguments of a specified component.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
Examples:
flow component parameters -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
metric-all
¶
Description: Query all metric data.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
Examples:
flow component metric-all -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
metrics
¶
Description: Query the list of metrics.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
A valid job id. |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
Examples:
flow component metrics -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
metric-delete
¶
Description: Delete specified metric.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
date |
|
|
No |
An 8-Digit Valid Date, Format Like ‘YYYYMMDD’ |
2 |
job_id |
|
|
No |
Job ID |
Examples:
# NOTICE: If you input both two optional arguments, the 'date' argument will be detected in priority while the 'job_id' argument would be ignored.
flow component metric-delete -d 20200101
flow component metric-delete -j $JOB_ID
output-model
¶
Description: Query a specified component model.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
Examples:
flow component output-model -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
output-data
¶
Description: Download the output data of a specified component.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
5 |
output_path |
|
|
Yes |
User specifies output directory path |
6 |
limit |
|
|
No |
Number of records to return, default -1 means return all data |
Examples:
flow component output-data -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0 --output-path ./examples/
output-data-table
¶
Description: View table name and namespace.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
Examples:
flow component output-data-table -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
list
¶
Description: List components of a specified job.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
Examples:
flow component list -j $JOB_ID
get-summary
¶
Description: Download summary of a specified component and save it as a json file.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
2 |
role |
|
|
Yes |
Role |
3 |
party_id |
|
|
Yes |
Party ID |
4 |
component_name |
|
|
Yes |
Component Name |
5 |
output_path |
|
|
No |
User specifies output directory path |
Examples:
flow component get-summary -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0
flow component get-summary -j $JOB_ID -r host -p 10000 -cpn hetero_feature_binning_0 -o ./examples/
Model¶
load
¶
Description: Load model.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
No |
Runtime configuration file path |
2 |
job_id |
|
|
No |
Job ID |
Examples:
flow model load -c fate_flow/examples/publish_load_model.json
flow model load -j $JOB_ID
bind
¶
Description: Bind model.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Runtime configuration file path |
2 |
job_id |
|
|
No |
Job ID |
Examples:
flow model bind -c fate_flow/examples/bind_model_service.json
flow model bind -c fate_flow/examples/bind_model_service.json -j $JOB_ID
import
¶
Description: Import model
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Runtime configuration file path |
2 |
from-database |
–from-database |
No |
If specified and there is a valid database environment, fate flow will import model from database which you specified in configuration file. |
Examples:
flow model import -c fate_flow/examples/import_model.json
flow model import -c fate_flow/examples/restore_model.json --from-database
export
¶
Description: Export model
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Runtime configuration file path |
2 |
to-database |
|
No |
If specified and there is a valid database environment, fate flow will export model to database which you specified in configuration file. |
Examples:
flow model export -c fate_flow/examples/export_model.json
flow model export -c fate_flow/examplse/store_model.json --to-database
migrate
¶
Description: Migrate model
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Runtime configuration file path |
Examples:
flow model migrate -c fate_flow/examples/migrate_model.json
tag-list
¶
Description: List tags of model.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
Examples:
flow model tag-list -j $JOB_ID
tag-model
¶
Description: Tag model.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
Yes |
Job ID |
2 |
tag_name |
|
|
Yes |
The name of tag |
3 |
remove |
|
No |
If specified, the name of specified model will be removed from the model name list of specified tag. |
Examples:
flow model tag-model -j $JOB_ID -t $TAG_NAME
flow model tag-model -j $JOB_ID -t $TAG_NAME --remove
Tag¶
create
¶
Description: Create tag.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
tag_name |
|
|
Yes |
The name of tag |
2 |
tag_description |
|
|
No |
The description of tag |
Examples:
flow tag create -t tag1 -d "This is the description of tag1."
flow tag create -t tag2
update
¶
Description: Update information of tag.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
tag_name |
|
|
Yes |
The name of tag |
2 |
new_tag_name |
|
No |
New name of tag |
|
3 |
new_tag_description |
|
No |
New description of tag |
Examples:
flow tag update -t tag1 --new-tag-name tag2
flow tag update -t tag1 --new-tag-desc "This is the new description."
list
¶
Description: List recorded tags.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
limit |
|
|
No |
Number of records to return. (default: 10) |
Examples:
flow tag list
flow tag list -l 3
query
¶
Description: Retrieve tag.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
tag_name |
|
|
Yes |
The name of tag |
2 |
with_model |
|
No |
If specified, the information of models which have the tag custom queried would be displayed |
Examples:
flow tag query -t $TAG_NAME
flow tag query -t $TAG_NAME --with-model
delete
¶
Description: Delete tag.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
tag_name |
|
|
Yes |
The name of tag |
Examples:
flow tag delete -t tag1
Data¶
download
¶
Description: Download Data Table.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Configuration file path |
Examples:
flow data download -c fate_flow/examples/download_host.json
upload
¶
Description: Upload Data Table.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
conf_path |
|
|
Yes |
Configuration file path |
2 |
verbose |
|
No |
If specified, verbose mode will be turn on. Users can have feedback on upload task in progress. (Default: False) |
|
3 |
drop |
|
No |
If specified, data of old version would be replaced by the current version. Otherwise, current upload task would be rejected. (Default: False) |
Examples:
flow data upload -c fate_flow/examples/upload_guest.json
flow data upload -c fate_flow/examples/upload_host.json --verbose --drop
upload-history
¶
Description: Query Upload Table History.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
limit |
|
|
No |
Number of records to return. (default: 10) |
2 |
job_id |
|
|
No |
Job ID |
Examples:
flow data upload-history -l 20
flow data upload-history --job-id $JOB_ID
Task¶
query
¶
Description: Query task information by filters.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
job_id |
|
|
No |
Job ID |
2 |
role |
|
|
No |
Role |
3 |
party_id |
|
|
No |
Party ID |
4 |
component_name |
|
|
No |
Component Name |
5 |
status |
|
|
No |
Job Status |
Examples:
flow task query -j $JOB_ID -p 9999 -r guest
flow task query -cpn hetero_feature_binning_0 -s complete
list
¶
Description: List tasks.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
limit |
|
|
No |
Number of records to return. (default: 10) |
Examples:
flow task list
flow task list -l 25
Table¶
info
¶
Description: Query Table Information.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
namespace |
|
|
Yes |
Namespace |
2 |
table_name |
|
|
Yes |
Table Name |
Examples:
flow table info -n $NAMESPACE -t $TABLE_NAME
delete
¶
Description: Delete A Specified Table.
Arguments:
No. |
Argument |
Flag_1 |
Flag_2 |
Required |
Description |
---|---|---|---|---|---|
1 |
namespace |
|
|
No |
Namespace |
2 |
table_name |
|
|
No |
Table name |
3 |
job_id |
|
|
No |
A valid job id |
4 |
role |
|
|
No |
Role |
5 |
party_id |
|
|
No |
Party ID |
6 |
component_name |
|
|
No |
Component Name |
Examples:
flow table delete -n $NAMESPACE -t $TABLE_NAME
flow table delete -j $JOB_ID -r guest -p 9999