I want to run batch predictions inside Google Cloud's vertex.ai using a custom trained model. I was able to find documentation to get online prediction working with a custom built docker image by setting up an endpoint, but I can't seem to find any documentation on what the Dockerfile should be for batch prediction. Specifically how does my custom code get fed the input and where does it put the output?
The documentation I've found is here, it certainly looks possible to use a custom model and when I tried it didn't complain, but eventually it did throw an error. According to the documentation no endpoint is required for running batch jobs.
Related
I'm running a training job using AWS SageMaker and i'm using a custom Estimator based on an available docker image from AWS. I wanted to get some feedback on whether my process is correct or not prior to deployment.
I'm running the training job in a docker container using 'local' in a SageMaker notebook instance and the training job runs successfully. However, after the job completes and saves the model to opt/model/models within the docker image, once the docker container exits, the model saved from training is lost. Ideally, i'd like to use the model for inference, however, I'm not sure about the best way of doing it. I have also tried the training job after pushing the image to ECR, but the same thing happens.
It is my understanding that the docker state is lost, once the image exits, as such, is it possible to persist the model that was produced in training in the image? One option I have thought about is saving the model output to an S3 bucket once the training job is complete, then pulling that model into another docker image for inference. Is this expected behaviour and the correct way of doing it?
I am fairly new to using SageMaker but i'd like to do it according to best practices. I've looked at a lot of the AWS documents and followed the tutorials but it doesn't seem to mention explicitly if this is how it should be done.
Thanks for any feedback on this.
You can refer to Rok's comment on saving a model file when you're using a custom estimator. That said, SageMaker built-in estimators save the model artifacts to S3. To make inferences using that model, you can either use a real-time inference endpoint for real time predictions, or a batch transformer to run inferences in batch mode. In both cases, you'll have to point the configuration to the container for inference and the model artifacts. the amazon-sagemaker-examples repository has examples for common frameworks, especially, the scikit-learn example has detailed explanations.
Also, make sure the model is being saved to /opt/ml/model/, not opt/model/models as mentioned in your question.
I am aware that it is possible to deploy custom containers for training jobs on google cloud and I have been able to get the same running using command.
gcloud ai-platform jobs submit training infer name --region some_region --master-image-uri=path/to/docker/image --config config.yaml
The training job was completed successfully and the model was successfully obtained, Now I want to use this model for inference, but the issue is a part of my code has system level dependencies, so I have to make some modification into the architecture in order to get it running all the time. This was the reason to have a custom container for the training job in the first place.
The documentation is only available for the training part and the inference part, (if possible) with custom containers has not been explored to the best of my knowledge.
The training part documentation is available on this link
My question is, is it possible to deploy custom containers for inference purposes on google cloud-ml?
This response refers to using Vertex AI Prediction, the newest platform for ML on GCP.
Suppose you wrote the model artifacts out to cloud storage from your training job.
The next step is to create the custom container and push to a registry, by following something like what is described here:
https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements
This section describes how you pass the model artifact directory to the custom container to be used for interence:
https://cloud.google.com/vertex-ai/docs/predictions/custom-container-requirements#artifacts
You will also need to create an endpoint in order to deploy the model:
https://cloud.google.com/vertex-ai/docs/predictions/deploy-model-api#aiplatform_deploy_model_custom_trained_model_sample-gcloud
Finally, you would use gcloud ai endpoints deploy-model ... to deploy the model to the endpoint:
https://cloud.google.com/sdk/gcloud/reference/ai/endpoints/deploy-model
I want to use a model that I have trained for inference on Google Cloud ML. It is a NLP model, and I want my node.js server to interact with the model to get predictions at train time.
I have a process for running inference on the model manually, that I would like to duplicate in the cloud:
Use Stanford Core NLP to tokenize my text and generate data files that store my tokenized text.
Have the model use those data files, create Tensorflow Examples out of it, and run the model.
Have the model print out the predictions.
Here is how I think I can replicate it in the Cloud:
Send the text to the cloud using my node.js server.
Run my python script to generate the data file. It seems like I will have to do this inside of a custom prediction routine. I'm not sure how I can use Stanford Core NLP here.
Save the data file in a bucket in Google Cloud.
In the custom prediction routine, load the saved data file and execute the model.
Can anyone tell me if this process is correct? Also, how can I run Stanford CoreNLP on Google Cloud custom prediction routine? Also, is there a way for me to just run command line scripts (for example for creating the data files I have a simple command that I normally just run to create them)?
You can implement a custom preprocessing method in Python and invoke the Stanford toolkit from there. See this blog and associated sample code for details: https://cloud.google.com/blog/products/ai-machine-learning/ai-in-depth-creating-preprocessing-model-serving-affinity-with-custom-online-prediction-on-ai-platform-serving
I am trying to deploy a locally trained model. I followed all of the instructions here for model preparation and I managed to deploy it.
However when I try to get the predictions, the online prediction responds with 502 Server error and the batch prediction returns ('Failed to run the inference graph', 1)
Is there a way to get a better error message to narrow down what's wrong?
Thanks
The error message indicated it occurred when running the session for the inference graph. It might be possible to uncover what is be happening with some code to use the model locally. One way to test it is to create a small input dataset and feed it to the inference graph to check if you can run the session locally.
You may refer the local_predict.py in the samples/mnist/deployable/ in SDK about how to do that. Here is an example use:
python local_predict.py --input=/path/to/my/local/files --model_dir=/path/to/modeldir.
Note that the model_dir points to where the tensorflow meta graph proto and checkpoint files are saved. They are generated by training. Here is the doc link about how to train a model. https://cloud.google.com/ml/docs/how-tos/training-models. The model dir can be on GCS as well.
Thanks for bringing this up. We're continually working to improve the overall experience of the service including error reporting.
I've been trying to use custom made images to run my google data flow pipeline. Given the information from https://cloud.google.com/compute/docs/reference/latest/images I've tested the following code snippets:
DataflowPipelineOptions options = PipelineOptionsFactory.create().as(DataflowPipelineOptions.class);
...
options.setDiskSourceImage("ubuntu-1504-vivid-v20150911");
options.setDiskSourceImage("projects/ubuntu-os-cloud/global/images/ubuntu-1504-vivid-v20150911");
options.setDiskSourceImage("https://www.googleapis.com/compute/beta/projects/ubuntu-os-cloud/global/images/ubuntu-1504-vivid-v20150911");
all of the above tries led to the following error in my pipeline:
(b9c7b66a676906f4): Unable to create VMs. Causes: (b9c7b66a67690aef): Error: Message: Invalid value for field 'resource.disks[0].initializeParams.sourceImage': '[edited]'. Must be the URL to a Compute resource of the correct type HTTP Code: 400
Using a custom disk image with Dataflow is not a viable option. The flag diskSourceImage is deprecated and will be removed in a future SDK release. The reason it is no longer supported is because the Dataflow service relies on versioned resources in the VM image. So Dataflow needs control of the VM image so that we can upgrade it as necessary. If users supply their own custom images we have no way of keeping them in sync with the requirements of the Dataflow service.
If your custom VM image is based off a Dataflow image then you would be able to execute jobs using that custom image until the next release of a Dataflow VM image. There is no reasonable way in which you would be able to keep your custom images in sync with Dataflow's VM images so that you would be able to keep this working.
If you would like to customize the VM image please let us know why (e.g. send us an email at dataflow-feedback#google.com) so we can either suggest an alternative solution or else consider supporting your use case in the future.
There's a subtle issue with setDiskSourceImage -- it uses 'beta' instead of the current 'v1' version for Compute Engine. If you try the following, it should work:
options.setDiskSourceImage("https://www.googleapis.com/compute/v1/projects/ubuntu-os-cloud/global/images/ubuntu-1504-vivid-v20150911");