Skip to content

Commit a694951

Browse files
prmerger-automator[bot]prmerger-automator[bot]
authored
Merge pull request #279965 from fkriti/update-nixtla-token-information
Update token and pricing information for Nixtla
2 parents 50a5229 + a8067bf commit a694951

File tree

2 files changed

+71
-0
lines changed

2 files changed

+71
-0
lines changed

‎articles/ai-studio/how-to/deploy-models-timegen-1.md

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -47,6 +47,42 @@ You can deploy TimeGEN-1 as a serverless API with pay-as-you-go billing. Nixtla
4747
- An [Azure AI Studio project](../how-to/create-projects.md).
4848
- Azure role-based access controls (Azure RBAC) are used to grant access to operations in Azure AI Studio. To perform the steps in this article, your user account must be assigned the __Azure AI Developer role__ on the resource group. For more information on permissions, visit [Role-based access control in Azure AI Studio](../concepts/rbac-ai-studio.md).
4949

50+
### Pricing information
51+
52+
#### Estimate the number of tokens needed
53+
54+
Before you create a deployment, it's useful to estimate the number of tokens that you plan to use and be billed for.
55+
One token corresponds to one data point in your input dataset or output dataset.
56+
57+
Suppose you have the following input time series dataset:
58+
59+
| Unique_id | Timestamp | Target Variable | Exogenous Variable 1 | Exogenous Variable 2 |
60+
|:---------:|:-------------------:|:---------------:|:--------------------:|:--------------------:|
61+
| BE | 2016-10-22 00:00:00 | 70.00 | 49593.0 | 57253.0 |
62+
| BE | 2016-10-22 01:00:00 | 37.10 | 46073.0 | 51887.0 |
63+
64+
To determine the number of tokens, multiply the number of rows (in this example, two) and the number of columns used for forecasting—not counting the unique_id and timestamp columns (in this example, three) to get a total of six tokens.
65+
66+
Given the following output dataset:
67+
68+
| Unique_id | Timestamp | Forecasted Target Variable |
69+
|:---------:|:-------------------:|:--------------------------:|
70+
| BE | 2016-10-22 02:00:00 | 46.57 |
71+
| BE | 2016-10-22 03:00:00 | 48.57 |
72+
73+
You can also determine the number of tokens by counting the number of data points returned after data forecasting. In this example, the number of tokens is two.
74+
75+
#### Estimate the pricing
76+
77+
There are four pricing meters, as described in the following table:
78+
79+
| Pricing Meter | Description |
80+
|-----------------------------------------------|-----------------------------------------------------------------------------------------|
81+
| paygo-inference-input-tokens | Costs associated with the tokens used as input for inference when *finetune_steps* = 0 |
82+
| paygo-inference-output-tokens | Costs associated with the tokens used as output for inference when *finetune_steps* = 0 |
83+
| paygo-finetuned-model-inference-input-tokens | Costs associated with the tokens used as input for inference when *finetune_steps* > 0 |
84+
| paygo-finetuned-model-inference-output-tokens | Costs associated with the tokens used as output for inference when *finetune_steps* > 0 |
85+
5086
### Create a new deployment
5187

5288
These steps demonstrate the deployment of TimeGEN-1. To create a deployment:

‎articles/machine-learning/how-to-deploy-models-timegen-1.md

Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -47,6 +47,41 @@ You can deploy TimeGEN-1 as a serverless API with pay-as-you-go billing. Nixtla
4747

4848
- Azure role-based access controls (Azure RBAC) are used to grant access to operations in Azure Machine Learning. To perform the steps in this article, your user account must be assigned the __Azure AI Developer role__ on the resource group. For more information on permissions, see [Manage access to an Azure Machine Learning workspace](how-to-assign-roles.md).
4949

50+
### Pricing information
51+
52+
#### Estimate the number of tokens needed
53+
54+
Before you create a deployment, it's useful to estimate the number of tokens that you plan to use and be billed for.
55+
One token corresponds to one data point in your input dataset or output dataset.
56+
57+
Suppose you have the following input time series dataset:
58+
59+
| Unique_id | Timestamp | Target Variable | Exogenous Variable 1 | Exogenous Variable 2 |
60+
|:---------:|:-------------------:|:---------------:|:--------------------:|:--------------------:|
61+
| BE | 2016-10-22 00:00:00 | 70.00 | 49593.0 | 57253.0 |
62+
| BE | 2016-10-22 01:00:00 | 37.10 | 46073.0 | 51887.0 |
63+
64+
To determine the number of tokens, multiply the number of rows (in this example, two) and the number of columns used for forecasting—not counting the unique_id and timestamp columns (in this example, three) to get a total of six tokens.
65+
66+
Given the following output dataset:
67+
68+
| Unique_id | Timestamp | Forecasted Target Variable |
69+
|:---------:|:-------------------:|:--------------------------:|
70+
| BE | 2016-10-22 02:00:00 | 46.57 |
71+
| BE | 2016-10-22 03:00:00 | 48.57 |
72+
73+
You can also determine the number of tokens by counting the number of data points returned after data forecasting. In this example, the number of tokens is two.
74+
75+
#### Estimate the pricing
76+
77+
There are four pricing meters, as described in the following table:
78+
79+
| Pricing Meter | Description |
80+
|-----------------------------------------------|-----------------------------------------------------------------------------------------|
81+
| paygo-inference-input-tokens | Costs associated with the tokens used as input for inference when *finetune_steps* = 0 |
82+
| paygo-inference-output-tokens | Costs associated with the tokens used as output for inference when *finetune_steps* = 0 |
83+
| paygo-finetuned-model-inference-input-tokens | Costs associated with the tokens used as input for inference when *finetune_steps* > 0 |
84+
| paygo-finetuned-model-inference-output-tokens | Costs associated with the tokens used as output for inference when *finetune_steps* > 0 |
5085

5186
### Create a new deployment
5287

0 commit comments

Comments
 (0)