This is an automated email from the ASF dual-hosted git repository.
byronhsu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/submarine.git
The following commit(s) were added to refs/heads/master by this push:
new db95c95 SUBMARINE-926. [User] Other utilities
db95c95 is described below
commit db95c951dc9d0a1ac33f9860d56f82ad175a3ea3
Author: Justin Chang <[email protected]>
AuthorDate: Tue Jul 13 01:58:10 2021 +0800
SUBMARINE-926. [User] Other utilities
### What is this PR for?
Write the docs of MLflow UI and Tensorboard UI
### What type of PR is it?
Documentation
### Todos
None
### What is the Jira issue?
https://issues.apache.org/jira/browse/SUBMARINE-926
### How should this be tested?
None
### Screenshots (if appropriate)
None
### Questions:
* Do the license files need updating? No
* Are there breaking changes for older versions? No
* Does this need new documentation? No
Author: Justin Chang <[email protected]>
Signed-off-by: byronhsu <[email protected]>
Closes #665 from yuanchiachang/SUBMARINE-926 and squashes the following
commits:
a33a0e02 [Justin Chang] SUBMARINE-926. [User] Other utilities
---
website/docs/assets/mlflow-ui.png | Bin 0 -> 67866 bytes
website/docs/userDocs/others/mlflow.md | 19 +++++++++++-
website/docs/userDocs/others/tensorboard.md | 45 ++++++++++++++++++++++++++++
3 files changed, 63 insertions(+), 1 deletion(-)
diff --git a/website/docs/assets/mlflow-ui.png
b/website/docs/assets/mlflow-ui.png
new file mode 100644
index 0000000..c4b78e8
Binary files /dev/null and b/website/docs/assets/mlflow-ui.png differ
diff --git a/website/docs/userDocs/others/mlflow.md
b/website/docs/userDocs/others/mlflow.md
index 801a1b2..6d1043c 100644
--- a/website/docs/userDocs/others/mlflow.md
+++ b/website/docs/userDocs/others/mlflow.md
@@ -1,7 +1,8 @@
---
-title: Mlflow UI
+title: MLflow UI
---
+
<!--
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
@@ -20,3 +21,19 @@ KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
-->
+
+### Usage
+MLflow UI shows the tracking result of the experiments. When we
+use the log_param or log_metric in ModelClient API, we could view
+the result in MLflow UI. Below is the example of the usage of MLflow
+UI.
+
+### Example
+1. `cd ./submarine/dev-support/examples/tracking`
+2. `eval $(minikube -p minikube docker-env)`
+3. `sh build.sh`(it will take few minutes)
+4. `sh post.sh`
+5. In the MLflow UI page, you can see the log_param and the log_metric
+ result. You can also compare the training between different workers.
+
+
diff --git a/website/docs/userDocs/others/tensorboard.md
b/website/docs/userDocs/others/tensorboard.md
index 101f5cf..601dd9f 100644
--- a/website/docs/userDocs/others/tensorboard.md
+++ b/website/docs/userDocs/others/tensorboard.md
@@ -2,6 +2,7 @@
title: Tensorboard
---
+
<!--
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements. See the NOTICE file
@@ -20,3 +21,47 @@ KIND, either express or implied. See the License for the
specific language governing permissions and limitations
under the License.
-->
+
+
+## Write to LogDirs by the environment variable
+
+### Environment variable
+
+- `SUBMARINE_TENSORBOARD_LOG_DIR`: Exist in every experiment container. You
just need to direct your logs to `$(SUBMARINE_TENSORBOARD_LOG_DIR)` (**NOTICE:
it is `()` not `{}`**), and you can inspect the process on the tensorboard
webpage.
+
+### Example
+
+```
+{
+ "meta": {
+ "name": "tensorflow-tensorboard-dist-mnist",
+ "namespace": "default",
+ "framework": "TensorFlow",
+ "cmd": "python /var/tf_mnist/mnist_with_summaries.py
--log_dir=$(SUBMARINE_TENSORBOARD_LOG_DIR) --learning_rate=0.01
--batch_size=20",
+ "envVars": {
+ "ENV_1": "ENV1"
+ }
+ },
+ "environment": {
+ "image": "apache/submarine:tf-mnist-with-summaries-1.0"
+ },
+ "spec": {
+ "Worker": {
+ "replicas": 1,
+ "resources": "cpu=1,memory=512M"
+ }
+ }
+}
+
+```
+
+## Connect to the tensorboard webpage
+
+1. Open the experiment page in the workbench, and Click the `TensorBoard`
button.
+
+
+
+2. Inspect the process on tensorboard page.
+
+
+
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
