Nightly TF / Cloned TFX - how to manage Image for Kubeflow? - tensorflow

When I access my Kubeflow endpoint to upload and run a pipeline using a cloned TFX, the process starts hanging at the first step producing this message:
"This step is in Pending state with this message: ImagePullBackOff: Back-off pulling image "tensorflow/tfx:0.14.0dev", which is the same image used in the created pipeline yaml file.
My overall goal is to build an ExampleGen for tfrecords files, just as described in the guide here. The most recent tfx version in pip is 0.13 and does not yet include the necessary functions. For this reason, I install tf-nightly and clone/build tfx (dev-version 0.14). Doing so and installing some additional modules, e.g. tensorflow_data_validation, I can now create my pipeline using the tfx components and including an ExampleGen for tfrecords files. I finally build the pipeline with the KubeflowRunner. Yet this yields the error stated above.
I now wonder about an appropriate way to address this. I guess one way would be to build an image myself with the specified versions, but maybe there is a more practical way?

TFX doesn't have a nightly image build as yet. Currently, it defaults to using the image tagged with the version of the library you use to build the pipeline, hence the reason the tag is 0.14dev0. This is the current version at HEAD, see here:
https://github.com/tensorflow/tfx/blob/a1f43af5e66f9548ae73eb64813509445843eb53/tfx/version.py#L17
You can build your own image and push it somewhere, for example gcr.io/your-gcp-project/your-image-name:tag, and specify that the pipeline use this image instead, by customizing the tfx_image argument to the pipeline:
https://github.com/tensorflow/tfx/blob/74f9b6ab26c51ebbfb5d17826c5d5288a67dcf85/tfx/orchestration/kubeflow/base_component.py#L54
See for example:
https://github.com/tensorflow/tfx/blob/b3796fc37bd4331a4e964c822502ba5096ad4bb6/tfx/examples/chicago_taxi_pipeline/taxi_pipeline_kubeflow.py#L243

Related

Unable to get image details : Environment version Autosave_(date)T(time)Z_******** provided in request doesn't match environ

On AzureML Batchendpoint, I'm recently hitting the following error:
Unable to get image details : Environment version Autosave_(date)T(time)Z_******** provided in request doesn't match environ.
when I setup the batch-endpoint with a yml config:
environment: azureml:env-name:env-version
So, AzureML creates and builds the environment with the version I specify env-version, which is just a number (in my case = 3).
and then for some weird reason, AzureML creates an extra environment version called Autosave_(date)T(time)Z_********, which is not built, but based on the previous one just created, and then it becomes the latest version of that environment.
In summary, AzureML instead of looking for the version that I specified as env-name:3 it seems to be looking for env-name:Autosave_(date)T(time)Z_******** and then throws the error message mentioned above.
I found the problem was that when creating an environment from a YAML specification file, one of my conda dependencies was cmake, which I needed to allow installation of another python module. The docker image is exactly the same as a previously created environment.
Removing the cmake dependency from the YAML file, eliminated the issue. So the workaround is to install it using a Dockerfile.
The error message was very misleading to start with, but got there in the end after understanding that AzureML reuses a cached image, based on the hash value, from the environment definition accordingly to this
So for that reason, the automatically created Autosave docker image references to that same build, which only happens once when the first job is sent.

Apache Flink to use S3 for backend state and checkpoints

Background
I was planning to use S3 to store the Flink's checkpoints using the FsStateBackend. But somehow I was getting the following error.
Error
org.apache.flink.core.fs.UnsupportedFileSystemSchemeException: Could not find a file system implementation for scheme 's3'. The scheme is not directly supported by Flink and no Hadoop file system to support this scheme could be loaded.
Flink version: I am using Flink 1.10.0 version.
I have found the solution for the above issue, so here I am listing it in steps that are required.
Steps
We need to add some configs in the flink-conf.yaml file which I have listed below.
state.backend: filesystem
state.checkpoints.dir: s3://s3-bucket/checkpoints/ #"s3://<your-bucket>/<endpoint>"
state.backend.fs.checkpointdir: s3://s3-bucket/checkpoints/ #"s3://<your-bucket>/<endpoint>"
s3.access-key: XXXXXXXXXXXXXXXXXXX #your-access-key
s3.secret-key: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx #your-secret-key
s3.endpoint: http://127.0.0.1:9000 #your-endpoint-hostname (I have used Minio)
After completing the first step we need to copy the respective(flink-s3-fs-hadoop-1.10.0.jar and flink-s3-fs-presto-1.10.0.jar) JAR files from the opt directory to the plugins directory of your Flink.
E.g:--> 1. Copy /flink-1.10.0/opt/flink-s3-fs-hadoop-1.10.0.jar to /flink-1.10.0/plugins/s3-fs-hadoop/flink-s3-fs-hadoop-1.10.0.jar // Recommended for StreamingFileSink
2. Copy /flink-1.10.0/opt/flink-s3-fs-presto-1.10.0.jar to /flink-1.10.0/plugins/s3-fs-presto/flink-s3-fs-presto-1.10.0.jar //Recommended for checkpointing
Add this in checkpointing code
env.setStateBackend(new FsStateBackend("s3://s3-bucket/checkpoints/"))
After completing all the above steps re-start the Flink if it is already running.
Note:
If you are using both(flink-s3-fs-hadoop and flink-s3-fs-presto) in Flink then please use s3p:// specificly for flink-s3-fs-presto and s3a:// for flink-s3-fs-hadoop instead of s3://.
For more details click here.

Submit a Keras training job to Google cloud

I am trying to follow this tutorial:
https://medium.com/#natu.neeraj/training-a-keras-model-on-google-cloud-ml-cb831341c196
to upload and train a Keras model on Google Cloud Platform, but I can't get it to work.
Right now I have downloaded the package from GitHub, and I have created a cloud environment with AI-Platform and a bucket for storage.
I am uploading the files (with the suggested folder structure) to my Cloud Storage bucket (basically to the root of my storage), and then trying the following command in the cloud terminal:
gcloud ai-platform jobs submit training JOB1
--module-name=trainer.cnn_with_keras
--package-path=./trainer
--job-dir=gs://mykerasstorage
--region=europe-north1
--config=gs://mykerasstorage/trainer/cloudml-gpu.yaml
But I get errors, first the cloudml-gpu.yaml file can't be found, it says "no such folder or file", and trying to just remove it, I get errors because it says the --init--.py file is missing, but it isn't, even if it is empty (which it was when I downloaded from the tutorial GitHub). I am Guessing I haven't uploaded it the right way.
Any suggestions of how I should do this? There is really no info on this in the tutorial itself.
I have read in another guide that it is possible to let gcloud package and upload the job directly, but I am not sure how to do this or where to write the commands, in my terminal with gcloud command? Or in the Cloud Shell in the browser? And how do I define the path where my python files are located?
Should mention that I am working with Mac, and pretty new to using Keras and Python.
I was able to follow the tutorial you mentioned successfully, with some modifications along the way.
I will mention all the steps although you made it halfway as you mentioned.
First of all create a Cloud Storage Bucket for the job:
gsutil mb -l europe-north1 gs://keras-cloud-tutorial
To answer your question on where you should write these commands, depends on where you want to store the files that you will download from GitHub. In the tutorial you posted, the writer is using his own computer to run the commands and that's why he initializes the gcloud command with gcloud init. However, you can submit the job from the Cloud Shell too, if you download the needed files there.
The only files we need from the repository are the trainer folder and the setup.py file. So, if we put them in a folder named keras-cloud-tutorial we will have this file structure:
keras-cloud-tutorial/
├── setup.py
└── trainer
├── __init__.py
├── cloudml-gpu.yaml
└── cnn_with_keras.py
Now, a possible reason for the ImportError: No module named eager error is that you might have changed the runtimeVersion inside the cloudml-gpu.yaml file. As we can read here, eager was introduced in Tensorflow 1.5. If you have specified an earlier version, it is expected to experience this error. So the structure of cloudml-gpu.yaml should be like this:
trainingInput:
scaleTier: CUSTOM
# standard_gpu provides 1 GPU. Change to complex_model_m_gpu for 4 GPUs
masterType: standard_gpu
runtimeVersion: "1.5"
Note: "standard_gpu" is a legacy machine type.
Also, the setup.py file should look like this:
from setuptools import setup, find_packages
setup(name='trainer',
version='0.1',
packages=find_packages(),
description='Example on how to run keras on gcloud ml-engine',
author='Username',
author_email='user#gmail.com',
install_requires=[
'keras==2.1.5',
'h5py'
],
zip_safe=False)
Attention: As you can see, I have specified that I want version 2.1.5 of keras. This is because if I don't do that, the latest version is used which has compatibility issues with versions of Tensorflow earlier than 2.0.
If everything is set, you can submit the job by running the following command inside the folder keras-cloud-tutorial:
gcloud ai-platform jobs submit training test_job --module-name=trainer.cnn_with_keras --package-path=./trainer --job-dir=gs://keras-cloud-tutorial --region=europe-west1 --config=trainer/cloudml-gpu.yaml
Note: I used gcloud ai-platform instead of gcloud ml-engine command although both will work. At some point in the future though, gcloud ml-engine will be deprecated.
Attention: Be careful when choosing the region in which the job will be submitted. Some regions do not support GPUs and will throw an error if chosen. For example, if in my command I set the region parameter to europe-north1 instead of europe-west1, I will receive the following error:
ERROR: (gcloud.ai-platform.jobs.submit.training) RESOURCE_EXHAUSTED:
Quota failure for project . The request for 1 K80
accelerators exceeds the allowed maximum of 0 K80, 0 P100, 0 P4, 0 T4,
0 TPU_V2, 0 TPU_V3, 0 V100. To read more about Cloud ML Engine quota,
see https://cloud.google.com/ml-engine/quotas.
- '#type': type.googleapis.com/google.rpc.QuotaFailure violations:
- description: The request for 1 K80 accelerators exceeds the allowed maximum of
0 K80, 0 P100, 0 P4, 0 T4, 0 TPU_V2, 0 TPU_V3, 0 V100.
subject:
You can read more about the features of each region here and here.
EDIT:
After the completion of the training job, there should be 3 folders in the bucket that you specified: logs/, model/ and packages/. The model is saved on the model/ folder a an .h5 file. Have in mind that if you set a specific folder for the destination you should include the '/' at the end. For example, you should set gs://my-bucket/output/ instead of gs://mybucket/output. If you do the latter you will end up with folders output, outputlogs and outputmodel. Inside output there should be packages. The job page link should direct to output folder so make sure to check the rest of the bucket too!
In addition, in the AI-Platform job page you should be able to see information regarding CPU, GPU and Network utilization:
Also, I would like to clarify something as I saw that you posted some related questions as an answer:
Your local environment, either it is your personal Mac or the Cloud Shell has nothing to do with the actual training job. You don't need to install any specific package or framework locally. You just need to have the Google Cloud SDK installed (in Cloud Shell is of course already installed) to run the appropriate gcloud and gsutil commands. You can read more on how exactly training jobs on the AI-Platform work here.
I hope that you will find my answer helpful.
I got it to work halfway now by not uploading the files but just running the upload commands from cloud at my local terminal... however there was an error during it running ending in "job failed"
Seems it was trying to import something from the TensorFlow backend called "from tensorflow.python.eager import context" but there was an ImportError: No module named eager
I have tried "pip install tf-nightly" which was suggested at another place, but it says I don't have permission or I am loosing the connection to cloud shell(exactly when I try to run the command).
I have also tried making a virtual environment locally to match that on gcloud (with Conda), and have made an environment with Conda with Python=3.5, Tensorflow=1.14.0 and Keras=2.2.5, which should be supported for gcloud.
The python program works fine in this environment locally, but I still get the (ImportError: No module named eager) when trying to run the job on gcloud.
I am putting the flag --python-version 3.5 when submitting the job, but when I write the command "Python -V" in the google cloud shell, it says Python=2.7. Could this be the issue? I have not fins a way to update the python version with the cloud shell prompt, but it says google cloud should support python 3.5. If this is anyway the issue, any suggestions on how to upgrade python version on google cloud?
It is also possible to manually there a new job in the google cloud web interface, doing this, I get a different error message: ERROR: Could not find a version that satisfies the requirement cnn_with_keras.py (from versions: none) and No matching distribution found for cnn_with_keras.py. Where cnn_with_keras.py is my python code from the tutorial, which runs fine locally.
Really don't know what to do next. Any suggestions or tips would be very helpful!
The issue with the GPU is solved now, it was something so simple as, my google cloud account had GPU settings disabled and needed to be upgraded.

TensorBoard without callbacks for Keras docker image in SageMaker

I'm trying to add TensorBoard functionality to this SageMaker example: https://github.com/awslabs/amazon-sagemaker-examples/blob/master/hyperparameter_tuning/keras_bring_your_own/hpo_bring_your_own_keras_container.ipynb
The issue is that SageMaker's Estimator.fit() does not seem to support Keras models compiled with callbacks.
Now from this git issue post it was described that what I need to do for TensorBoard functionality is,
"You need your code inside the container to save checkpoints to S3,
and you need to periodically sync your local Tensorboard log directory
with your S3 checkpoints."
So to sum it all up, to enable TensorBoard in SageMaker with this custom Keras docker image, it looks like I need a way of periodically uploading a file to an S3 bucket during training without using callbacks. Is this possible to do? I was considering trying to shove this code into a custom loss function, but I'm not sure if this would be the way to go about it. Any help is greatly appreciated!

Running a JSON graph locally

I've been using Flowhub.io to do my development on the nodejs device. Now that the GUI-based design is done, I'm ready to take it offline and run the code via the command line. How would do I do this? I have the JSON file corresponding to the graph I created online, but not sure how to use the noflo nodejs module.
Could someone help me by showing me an example of how to load a graph using the noflo module, please? Thanks!
f you want to run an existing graph, you can use the --graph option.
noflo-nodejs --graph graphs/MyMainGraph.json
If you also want the process to exit when the network stops, you can pass --batch.
PS: I added this to the noflo-nodejs README.