---
id: "alpaca-to-chatml"
title: "Convert Alpaca to Conversation Format"
slug: "alpaca-to-chatml"
description: "Convert Alpaca format to ChatML Conversation format"
code: |
    -- Convert Alpaca format to Conversation format
    WITH 
    source_view AS (
      SELECT * FROM train  -- Change 'train' to your desired view name here
    )
    SELECT 
      [
        struct_pack(
          "from" := 'system',
          "value" := instruction
        ),
        struct_pack(
          "from" := 'user',
          "value" := input
        ),
        struct_pack(
          "from" := 'assistant',
          "value" := output
        )
      ] AS conversation
    FROM source_view
    WHERE instruction IS NOT NULL 
      AND input IS NOT NULL 
      AND output IS NOT NULL;
---

# Converting Alpaca to ChatML Conversation Format

```sql
-- Convert Alpaca format to Conversation format
WITH 
source_view AS (
  SELECT * FROM train  -- Change 'train' to your desired view name here
)
SELECT 
  [
    struct_pack(
      "from" := 'system',
      "value" := instruction
    ),
    struct_pack(
      "from" := 'user',
      "value" := input
    ),
    struct_pack(
      "from" := 'assistant',
      "value" := output
    )
  ] AS conversation
FROM source_view
WHERE instruction IS NOT NULL 
  AND input IS NOT NULL 
  AND output IS NOT NULL;
```

## Why?

Differences between Alpaca and ChatML Conversation Format:

1. **Alpaca Format**:
   - The Alpaca format usually has three columns: `instruction`, `input`, and `output`.

2. **ChatML Conversation Format**:
   - The ChatML Conversation format is a JSON format that contains a list of messages.
   - Each message has a `from` field, which can be either `system`, `user`, or `assistant`.
   - The `value` field contains the message content.


## Example

### `yahma/alpaca-cleaned`

<iframe
  src="https://huggingface.co/datasets/yahma/alpaca-cleaned/embed/viewer/default/train"
  frameborder="0"
  width="100%"
  height="560px"
></iframe>

You can run this query through via the `sql_console` in the Hugging Face Hub [here](https://huggingface.co/datasets/yahma/alpaca-cleaned?sql_console=true&sql=WITH+%0Asource_view+AS+%28%0A++SELECT+*+FROM+train++--+Change+%27train%27+to+your+desired+view+name+here%0A%29%0ASELECT+%0A++%5B%0A++++struct_pack%28%0A++++++%22from%22+%3A%3D+%27system%27%2C%0A++++++%22value%22+%3A%3D+instruction%0A++++%29%2C%0A++++struct_pack%28%0A++++++%22from%22+%3A%3D+%27user%27%2C%0A++++++%22value%22+%3A%3D+input%0A++++%29%2C%0A++++struct_pack%28%0A++++++%22from%22+%3A%3D+%27assistant%27%2C%0A++++++%22value%22+%3A%3D+output%0A++++%29%0A++%5D+AS+conversation%0AFROM+source_view%0AWHERE+instruction+IS+NOT+NULL+%0A++AND+input+IS+NOT+NULL+%0A++AND+output+IS+NOT+NULL%3B).

![Alpaca to ChatML](./alpaca-to-conversation.png)