AWS S3
Set up an import job to ingest data into Arize from AWS S3
Set up an import job to log inference files to Arize. Updates to files are checked every 10 seconds. Users generally find a sweet spot around a few hundred thousand to a million rows in each file with the total file limit being 1GB.
Navigate to the 'Upload Data' page on the left navigation bar in the Arize platform. From there, select the 'AWS S3' card to begin a new file import job.
Step 1: Select Amazon S3
There are two ways to setup access permissions with Arize
Configure An Individual Bucket Policy
Configure Multiple Buckets Via Role Based Permissions
Fill in the file path where you would like Arize to pull your model's inferences. Arize will automatically infer your bucket name and prefix.
Example File Path In Arize UI
Create the path to your bucket and folder to pull your model's inferences. In this example, you might have an AWS bucket and folder named
s3://example-demo-bucket/click-thru-rate/production/v1/ that contains parquet files of your model inferences. Your bucket name is example-demo-bucket and your prefix is click-thru-rate/production/v1/.The file structure can take into consideration various model environments (training, production, etc) and locations of ground truth. In addition, S3 bucket import allows recursive operations. This means that it will include all nested subdirectories within the specified bucket prefix, regardless of the number or depth of these directories.
File Directory Example
There are multiple ways to structure your file directory. If actuals and predictions can be sent together, simply store this data in a the same file and import this data together through a single file importer job.
In the case of delayed actuals, we recommend you separate your predictions and actuals into separate folders and loading this data through two separate file importer jobs. Learn more here.
s3://bucket1/click-thru-rate/production/prediction/
├── 11-19-2022.parquet
├── 11-20-2022.parquet
├── 11-21-2022.parquet
s3://bucket1/click-thru-rate/production/actuals/
├── 12-1-2022.parquet # same prediction id column, model, and space as the corresponding prediction
├── 12-2-2022.parquet
└── 12-3-2022.parquet
In Arize UI: Copy
arize-ingestion-key value
Example Bucket Tag In Arize UI
In AWS Console: Navigate to your S3 bucket -> Properties -> Edit Bucket Policy
Navigate to your bucket properties tab
In AWS Console: Set tag Key =
arize-ingestion-key and Value as the value copied from Arize UI from the previous step
Add arize-ingestion-key as a bucket tag
In Arize UI: Copy the policy supplied by Arize in the file importer job setup
Copy AWS Policy In Arize UI
In the AWS console: Navigate to your S3 bucket -> Permission -> Edit Bucket Policy
Add/Edit bucket policy
In the AWS console: Paste the above AWS policy from Arize UI into the bucket policy
Add policy to your bucket
Ask Arize: To give Arize access to multiple buckets, ask arize to provide you with an External ID. Reach out to [email protected] for assistance.
Provide Arize: Provide Arize with the role ARN to access each AWS bucket you want to connect.
In AWS Console: For each role, Arize will assume add the following statement to the role's Trust Policy
{
"Effect": "Allow",
"Principal": {
"AWS": "arn:aws:iam::756106863523:root"
},
"Action": "sts:AssumeRole",
"Condition": {
"StringEquals": {
"sts:ExternalId": "<EXTERNAL ID PROVIDED BY ARIZE>"
}
}
}
In AWS Console: Add the following statement to the role's Permissions Policies
{
"Effect": "Allow",
"Action": [
"s3:GetBucketTagging",
"s3:GetObject",
"s3:ListBucket"
],
"Resource": [
"arn:aws:s3:::<YOUR BUCKET NAME>",
"arn:aws:s3:::<YOUR BUCKET NAME>/*"
]
}
In Arize UI: Copy
arize-ingestion-key value
In AWS Console: Navigate to your S3 bucket -> Properties -> Edit Bucket Policy
In AWS Console: Set tag Key =
arize-ingestion-key and Value as the value copied from Arize UI from the previous step
Model schema parameters are a way of organizing model inference data to ingest to Arize. When configuring your schema, be sure to match your data column headers with the model schema.
You can either use a form or a simple JSON-based schema to specify the column mapping.
Arize supports CSV, Parquet, Avro, and Apache Arrow. Refer here for a list of the expected data types by input type.
File Schema Form Inputs
Form Schema JSON Inputs
Property | Description | Required |
|---|---|---|
prediction_ID | The unique identifier of a specific prediction. Limited to 128 characters. | Required |
timestamp | The timestamp of the prediction in seconds or an RFC3339 timestamp | Optional, defaults to current timestamp at file ingestion time |
prediction_label | Column name for the prediction value | |
prediction_score | Column name for the predicted score | |
actual_label | Column name for the actual or ground truth value | Optional for production records |
actual_score | Column name for the ground truth score | |
prediction_group_id | Column name for ranking groups or lists in ranking models | |
rank | Column name for rank of each element on the its group or list | |
relevance_label | Column name for ranking actual or ground truth value | |
relevance_score | Column name for ranking ground truth score | |
features | A string prefix to describe a column feature/. Features must be sent in the same file as predictions | Arize automatically infers columns as features. Choose between feature prefixing OR inferred features. |
tags | A string prefix to describe a column tag/. Tags must be sent in the same file as predictions and features | Optional |
shap_values | A string prefix to describe a column shap/. SHAP must be sent in the same file as predictions or with a matching prediction_id | Optional |
version | A column to specify model version. version/ assigns a version to the corresponding data within a column, or configure your version within the UI | Optional, defaults to 'no_version' |
batch_id | Distinguish different batches of data under the same model_id and model_version. Must be specified as a constant during job setup or in the schema | Optional for validation records only |
exclude | A list of columns to exclude if the features property is not included in the ingestion schema | Optional |
embedding_features | A list of embedding columns, required vector column, optional raw data column, and optional link to data column. Learn more here | Optional |
Once you fill in your applicable predictions, actuals, and model inputs, click 'Validate Schema' to visualize your model schema in the Arize UI. Check that your column names and corresponding data match for a successful import job.
Once finished, your import job will be created and will start polling your bucket for files.
If your model receives delayed actuals, connect your predictions and actuals using the same prediction ID, which links your data together in the Arize platform. Arize regularly checks your data source for both predictions and actuals, and ingests them separately as they become available. Learn more here.
Arize will attempt a dry run to validate your job for any access, schema, or record-level errors. If the dry run is successful, you can proceed to create the import job. From there, you will be taken to the 'Job Status' tab.
All active jobs will regularly sync new data from your data source with Arize. You can view the job details by clicking on the job ID, which reveals more information about the job.
Job status page showing job listings
To pause, delete, or edit your file schema, click on 'Job Options'.
- Delete a job if it is no longer needed or if you made an error connecting to the wrong bucket. This will set your job status as 'deleted' in Arize.
- Pause a job if you have a set cadence to update your table. This way, you can 'start job' when you know there will be new data to reduce query costs. This will set your job status as 'inactive' in Arize.
- Edit a file schema if you have added, renamed, or missed a column in the original schema declaration.
An import job may run into a few problems. Use the dry run and job details UI to troubleshoot and quickly resolve data ingestion issues.
If there is an error validating a file against the model schema, Arize will surface an actionable error message. From there, click on the 'Fix Schema' button to adjust your model schema.
%20(1).png?alt=media&token=05143ca1-1b95-4dfa-b4da-bb2ae7a9f75e)
If your dry run is successful, but your job fails, click on the job ID to view the job details. This uncovers job details such as information about the file path or query id, the last import job, potential errors, and error locations.
.png?alt=media&token=5c5a830f-1cc5-4de2-a2d5-f091f0574e62)
Once you've identified the job failure point, fix the file errors and reupload the file to Arize with a new name.
resource "aws_s3_bucket" "arize-example-bucket" {
bucket = "my-arize-example-bucket"
tags = {
arize-ingestion-key = "value_from_arize_ui"
}
}
resource "aws_s3_bucket_policy" "grant_arize_read_only_access" {
bucket = aws_s3_bucket.arize-example-bucket.id
policy = data.aws_iam_policy_document.grant_arize_read_only_access.json
}
data "aws_iam_policy_document" "grant_arize_read_only_access" {
statement {
principals {
type = "AWS"
identifiers = ["arn:aws:iam::<REDACTED>:role/arize-importer"]
}
actions = [
"s3:GetBucketTagging",
"s3:GetObject",
"s3:ListBucket",
]
resources = [
aws_s3_bucket.arize-example-bucket.arn,
"${aws_s3_bucket.arize-example-bucket.arn}/*",
]
}
}