arize-experiment

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Arize Experiment Skill

Arize实验Skill

Concepts

核心概念

Experiment = a named evaluation run against a specific dataset version, containing one run per example
Experiment Run = the result of processing one dataset example -- includes the model output, optional evaluations, and optional metadata
Dataset = a versioned collection of examples; every experiment is tied to a dataset and a specific dataset version
Evaluation = a named metric attached to a run (e.g.,
```
correctness
```
,
```
relevance
```
), with optional label, score, and explanation

The typical flow: export a dataset → process each example → collect outputs and evaluations → create an experiment with the runs.

Experiment = 针对特定数据集版本的命名评估运行，每个示例对应一个运行记录
Experiment Run = 处理单个数据集示例的结果——包含模型输出、可选的评估结果和可选元数据
Dataset = 带版本的示例集合；每个实验都关联一个数据集和特定的数据集版本
Evaluation = 附加到运行记录的命名指标（例如
```
correctness
```
、
```
relevance
```
），包含可选的标签、分数和解释

典型流程：导出数据集 → 处理每个示例 → 收集输出和评估结果 → 使用运行记录创建实验。

Prerequisites

前提条件

Three things are needed:

ax

CLI, an API key (env var or profile), and a space ID. A project name is also needed but usually comes from the user's message.

需要准备三样东西：

ax

CLI、API密钥（环境变量或配置文件），以及空间ID。还需要项目名称，通常可从用户的消息中获取。

Install ax

安装ax

Verify

ax

is installed and working before proceeding:

Check if
```
ax
```
is on PATH:
```
command -v ax
```
(Unix) or
```
where ax
```
(Windows)

If not found, check common install locations:

macOS/Linux:

test -x ~/.local/bin/ax && export PATH="$HOME/.local/bin:$PATH"

Windows: check

%APPDATA%\Python\Scripts\ax.exe

%LOCALAPPDATA%\Programs\Python\Scripts\ax.exe

If still not found, install it (requires shell access to install packages):
- Preferred:
```
uv tool install arize-ax-cli
```
- Alternative:
```
pipx install arize-ax-cli
```
- Fallback:
```
pip install arize-ax-cli
```

After install, if

ax

is not on PATH:

macOS/Linux:
```
export PATH="$HOME/.local/bin:$PATH"
```

Windows (PowerShell):

$env:PATH = "$env:APPDATA\Python\Scripts;$env:PATH"

ax --version

fails with an SSL/certificate error:

macOS:
```
export SSL_CERT_FILE=/etc/ssl/cert.pem
```

Linux:

export SSL_CERT_FILE=/etc/ssl/certs/ca-certificates.crt

Windows (PowerShell):

$env:SSL_CERT_FILE = "C:\Program Files\Common Files\SSL\cert.pem"

(or use

python -c "import certifi; print(certifi.where())"

to find the cert bundle)

```
ax --version
```
must succeed before proceeding. If it doesn't, stop and ask the user for help.

在继续操作前，请先验证

ax

已安装且可正常使用：

检查
```
ax
```
是否在PATH中：
```
command -v ax
```
（Unix系统）或
```
where ax
```
（Windows系统）

如果未找到，检查常见安装位置：

macOS/Linux：

test -x ~/.local/bin/ax && export PATH="$HOME/.local/bin:$PATH"

Windows：检查

%APPDATA%\Python\Scripts\ax.exe

或

%LOCALAPPDATA%\Programs\Python\Scripts\ax.exe

如果仍未找到，请安装它（需要Shell权限来安装软件包）：
- 推荐方式：
```
uv tool install arize-ax-cli
```
- 替代方式：
```
pipx install arize-ax-cli
```
- 备用方式：
```
pip install arize-ax-cli
```

安装后，如果

ax

不在PATH中：

macOS/Linux：
```
export PATH="$HOME/.local/bin:$PATH"
```

Windows（PowerShell）：

$env:PATH = "$env:APPDATA\Python\Scripts;$env:PATH"

如果

ax --version

因SSL/证书错误失败：

macOS：
```
export SSL_CERT_FILE=/etc/ssl/cert.pem
```

Linux：

export SSL_CERT_FILE=/etc/ssl/certs/ca-certificates.crt

Windows（PowerShell）：

$env:SSL_CERT_FILE = "C:\Program Files\Common Files\SSL\cert.pem"

（或使用

python -c "import certifi; print(certifi.where())"

查找证书包路径）

必须确保
```
ax --version
```
执行成功才能继续。如果失败，请停止操作并向用户寻求帮助。

Verify environment

验证环境

Run a quick check for credentials:

macOS/Linux (bash):

bash

ax --version && echo "--- env ---" && echo "ARIZE_API_KEY: ${ARIZE_API_KEY:-(not set)}" && echo "ARIZE_SPACE_ID: ${ARIZE_SPACE_ID:-(not set)}" && echo "--- profiles ---" && ax profiles show 2>&1

Windows (PowerShell):

powershell

ax --version; Write-Host "--- env ---"; Write-Host "ARIZE_API_KEY: $env:ARIZE_API_KEY"; Write-Host "ARIZE_SPACE_ID: $env:ARIZE_SPACE_ID"; Write-Host "--- profiles ---"; ax profiles show 2>&1

Read the output and proceed immediately if either the env var or the profile has an API key. Only ask the user if both are missing. Resolve failures:

No API key in env and no profile → AskQuestion: "Arize API key (https://app.arize.com/admin > API Keys)"
Space ID unknown → AskQuestion, or run
```
ax projects list -o json --limit 100
```
and search for a match
Project unclear → ask, or run
```
ax projects list -o json --limit 100
```
and present as selectable options

快速检查凭证是否配置正确：

macOS/Linux（bash）：

bash

ax --version && echo "--- 环境变量 ---" && echo "ARIZE_API_KEY: ${ARIZE_API_KEY:-(未设置)}" && echo "ARIZE_SPACE_ID: ${ARIZE_SPACE_ID:-(未设置)}" && echo "--- 配置文件 ---" && ax profiles show 2>&1

Windows（PowerShell）：

powershell

ax --version; Write-Host "--- 环境变量 ---"; Write-Host "ARIZE_API_KEY: $env:ARIZE_API_KEY"; Write-Host "ARIZE_SPACE_ID: $env:ARIZE_SPACE_ID"; Write-Host "--- 配置文件 ---"; ax profiles show 2>&1

立即根据输出结果操作：如果环境变量或配置文件中存在API密钥，可直接继续。只有当两者都缺失时，才需要询问用户。解决失败情况：

环境变量中无API密钥且无配置文件 → 询问用户："请提供Arize API密钥（获取地址：https://app.arize.com/admin > API Keys）"
空间ID未知 → 询问用户，或运行
```
ax projects list -o json --limit 100
```
并搜索匹配项
项目不明确 → 询问用户，或运行
```
ax projects list -o json --limit 100
```
并提供可选选项

Space ID and Project

空间ID与项目

Both are needed for most commands. Resolve each:

User provides it in the conversation -- use directly via
```
--space-id
```
/
```
--project
```
flags.
Env var is set (
```
ARIZE_SPACE_ID
```
,
```
ARIZE_DEFAULT_PROJECT
```
) -- use silently.
If missing, AskQuestion once. Tell the user:
- Space ID is in the Arize URL:
```
/spaces/{SPACE_ID}/...
```
- Project is the project name as shown in the Arize UI.
- For convenience, recommend setting env vars so they don't get asked again:
```
export ARIZE_SPACE_ID="U3BhY2U6..."
```
  and
```
export ARIZE_DEFAULT_PROJECT="my-project"
```

Prefer asking the user over searching or iterating through projects and API keys. If you get a

401 Unauthorized

, tell the user their API key may not have access to that space and ask them to verify.

大多数命令都需要这两个参数。获取方式如下：

用户在对话中提供——直接通过
```
--space-id
```
/
```
--project
```
参数使用。
已设置环境变量（
```
ARIZE_SPACE_ID
```
、
```
ARIZE_DEFAULT_PROJECT
```
）——静默使用。
如果缺失，询问用户一次。告知用户：
- 空间ID可在Arize的URL中找到：
```
/spaces/{SPACE_ID}/...
```
- 项目是Arize UI中显示的项目名称。
- 为方便后续使用，建议设置环境变量：
```
export ARIZE_SPACE_ID="U3BhY2U6..."
```
  和
```
export ARIZE_DEFAULT_PROJECT="my-project"
```

优先询问用户，而非遍历搜索项目和API密钥。如果收到

401 Unauthorized

错误，告知用户其API密钥可能无权访问该空间，请他们验证。

List Experiments:

ax experiments list

列出实验：

ax experiments list

Browse experiments, optionally filtered by dataset. Output goes to stdout.

bash

ax experiments list
ax experiments list --dataset-id DATASET_ID --limit 20
ax experiments list --cursor CURSOR_TOKEN
ax experiments list -o json

浏览实验，可按数据集筛选。输出结果将打印到标准输出。

bash

ax experiments list
ax experiments list --dataset-id DATASET_ID --limit 20
ax experiments list --cursor CURSOR_TOKEN
ax experiments list -o json

Flags

参数说明

Flag	Type	Default	Description
`--dataset-id`	string	none	Filter by dataset
`--limit, -l`	int	15	Max results (1-100)
`--cursor`	string	none	Pagination cursor from previous response
`-o, --output`	string	table	Output format: table, json, csv, parquet, or file path
`-p, --profile`	string	default	Configuration profile

参数	类型	默认值	描述
`--dataset-id`	字符串	无	按数据集筛选
`--limit, -l`	整数	15	最大结果数（1-100）
`--cursor`	字符串	无	上一次响应返回的分页游标
`-o, --output`	字符串	table	输出格式：table、json、csv、parquet或文件路径
`-p, --profile`	字符串	default	配置文件

Get Experiment:

ax experiments get

获取实验详情：

ax experiments get

Quick metadata lookup -- returns experiment name, linked dataset/version, and timestamps.

bash

ax experiments get EXPERIMENT_ID
ax experiments get EXPERIMENT_ID -o json

快速查询元数据——返回实验名称、关联的数据集/版本以及时间戳。

bash

ax experiments get EXPERIMENT_ID
ax experiments get EXPERIMENT_ID -o json

Flags

参数说明

Flag	Type	Default	Description
`EXPERIMENT_ID`	string	required	Positional argument
`-o, --output`	string	table	Output format
`-p, --profile`	string	default	Configuration profile

参数	类型	默认值	描述
`EXPERIMENT_ID`	字符串	必填	位置参数
`-o, --output`	字符串	table	输出格式
`-p, --profile`	字符串	default	配置文件

Response fields

响应字段

Field	Type	Description
`id`	string	Experiment ID
`name`	string	Experiment name
`dataset_id`	string	Linked dataset ID
`dataset_version_id`	string	Specific dataset version used
`experiment_traces_project_id`	string	Project where experiment traces are stored
`created_at`	datetime	When the experiment was created
`updated_at`	datetime	Last modification time

字段	类型	描述
`id`	字符串	实验ID
`name`	字符串	实验名称
`dataset_id`	字符串	关联的数据集ID
`dataset_version_id`	字符串	使用的特定数据集版本ID
`experiment_traces_project_id`	字符串	存储实验追踪数据的项目ID
`created_at`	日期时间	实验创建时间
`updated_at`	日期时间	最后修改时间

Export Experiment:

ax experiments export

导出实验：

ax experiments export

Download all runs to a file. By default uses the REST API; pass

--all

to use Arrow Flight for bulk transfer.

bash

ax experiments export EXPERIMENT_ID

将所有运行记录下载到文件中。默认使用REST API；添加

--all

参数可使用Arrow Flight进行批量传输。

bash

ax experiments export EXPERIMENT_ID

-> experiment_abc123_20260305_141500/runs.json

ax experiments export EXPERIMENT_ID --all ax experiments export EXPERIMENT_ID --output-dir ./results ax experiments export EXPERIMENT_ID --stdout ax experiments export EXPERIMENT_ID --stdout | jq '.[0]'

undefined

undefined

Flags

参数说明

Flag	Type	Default	Description
`EXPERIMENT_ID`	string	required	Positional argument
`--all`	bool	false	Use Arrow Flight for bulk export (see below)
`--output-dir`	string	`.`	Output directory
`--stdout`	bool	false	Print JSON to stdout instead of file
`-p, --profile`	string	default	Configuration profile

参数	类型	默认值	描述
`EXPERIMENT_ID`	字符串	必填	位置参数
`--all`	布尔值	false	使用Arrow Flight进行批量导出（见下文）
`--output-dir`	字符串	`.`	输出目录
`--stdout`	布尔值	false	将JSON打印到标准输出而非保存到文件
`-p, --profile`	字符串	default	配置文件

REST vs Flight (

--all

)

REST与Flight（

--all

）对比

REST (default): Lower friction -- no Arrow/Flight dependency, standard HTTPS ports, works through any corporate proxy or firewall. Limited to 500 runs per page.
Flight (
```
--all
```
): Required for experiments with more than 500 runs. Uses gRPC+TLS on a separate host/port (
```
flight.arize.com:443
```
) which some corporate networks may block.

Agent auto-escalation rule: If a REST export returns exactly 500 runs, the result is likely truncated. Re-run with

--all

to get the full dataset.

Output is a JSON array of run objects:

json

[
  {
    "id": "run_001",
    "example_id": "ex_001",
    "output": "The answer is 4.",
    "evaluations": {
      "correctness": { "label": "correct", "score": 1.0 },
      "relevance": { "score": 0.95, "explanation": "Directly answers the question" }
    },
    "metadata": { "model": "gpt-4o", "latency_ms": 1234 }
  }
]

REST（默认）：门槛更低——无需依赖Arrow/Flight，使用标准HTTPS端口，可通过任何企业代理或防火墙。每页最多返回500条运行记录。
Flight（
```
--all
```
）：实验运行记录超过500条时必须使用。使用gRPC+TLS协议，独立的主机/端口（
```
flight.arize.com:443
```
），部分企业网络可能会阻止此连接。

Agent自动升级规则：如果REST导出恰好返回500条运行记录，结果可能被截断。需重新运行并添加

--all

参数以获取完整数据集。

输出为运行记录对象的JSON数组：

json

[
  {
    "id": "run_001",
    "example_id": "ex_001",
    "output": "答案是4。",
    "evaluations": {
      "correctness": { "label": "correct", "score": 1.0 },
      "relevance": { "score": 0.95, "explanation": "直接回答了问题" }
    },
    "metadata": { "model": "gpt-4o", "latency_ms": 1234 }
  }
]

Create Experiment:

ax experiments create

创建实验：

ax experiments create

Create a new experiment with runs from a data file.

bash

ax experiments create --name "gpt-4o-baseline" --dataset-id DATASET_ID --file runs.json
ax experiments create --name "claude-test" --dataset-id DATASET_ID --file runs.csv

使用数据文件中的运行记录创建新实验。

bash

ax experiments create --name "gpt-4o-baseline" --dataset-id DATASET_ID --file runs.json
ax experiments create --name "claude-test" --dataset-id DATASET_ID --file runs.csv

Flags

参数说明

Flag	Type	Required	Description
`--name, -n`	string	yes (prompted)	Experiment name
`--dataset-id`	string	yes (prompted)	Dataset to run the experiment against
`--file, -f`	path	yes (prompted)	Data file with runs: CSV, JSON, JSONL, or Parquet
`-o, --output`	string	no	Output format
`-p, --profile`	string	no	Configuration profile

参数	类型	是否必填	描述
`--name, -n`	字符串	是（会提示输入）	实验名称
`--dataset-id`	字符串	是（会提示输入）	实验关联的数据集ID
`--file, -f`	文件路径	是（会提示输入）	包含运行记录的数据文件：CSV、JSON、JSONL或Parquet格式
`-o, --output`	字符串	否	输出格式
`-p, --profile`	字符串	否	配置文件

Required columns in the runs file

运行记录文件的必填列

Column	Type	Required	Description
`example_id`	string	yes	ID of the dataset example this run corresponds to
`output`	string	yes	The model/system output for this example

Additional columns are passed through as

additionalProperties

on the run.

列	类型	是否必填	描述
`example_id`	字符串	是	此运行记录对应的数据集示例ID
`output`	字符串	是	此示例对应的模型/系统输出

其他列将作为运行记录的

additionalProperties

传递。

Delete Experiment:

ax experiments delete

删除实验：

ax experiments delete

bash

ax experiments delete EXPERIMENT_ID
ax experiments delete EXPERIMENT_ID --force   # skip confirmation prompt

bash

ax experiments delete EXPERIMENT_ID
ax experiments delete EXPERIMENT_ID --force   # 跳过确认提示

Flags

参数说明

Flag	Type	Default	Description
`EXPERIMENT_ID`	string	required	Positional argument
`--force, -f`	bool	false	Skip confirmation prompt
`-p, --profile`	string	default	Configuration profile

参数	类型	默认值	描述
`EXPERIMENT_ID`	字符串	必填	位置参数
`--force, -f`	布尔值	false	跳过确认提示
`-p, --profile`	字符串	default	配置文件

Experiment Run Schema

实验运行记录 Schema

Each run corresponds to one dataset example:

json

{
  "example_id": "required -- links to dataset example",
  "output": "required -- the model/system output for this example",
  "evaluations": {
    "metric_name": {
      "label": "optional string label (e.g., 'correct', 'incorrect')",
      "score": "optional numeric score (e.g., 0.95)",
      "explanation": "optional freeform text"
    }
  },
  "metadata": {
    "model": "gpt-4o",
    "temperature": 0.7,
    "latency_ms": 1234
  }
}

每个运行记录对应一个数据集示例：

json

{
  "example_id": "必填——关联到数据集示例",
  "output": "必填——此示例对应的模型/系统输出",
  "evaluations": {
    "metric_name": {
      "label": "可选字符串标签（例如'correct'、'incorrect'）",
      "score": "可选数值分数（例如0.95）",
      "explanation": "可选自由文本"
    }
  },
  "metadata": {
    "model": "gpt-4o",
    "temperature": 0.7,
    "latency_ms": 1234
  }
}

Evaluation fields

评估字段

Field	Type	Required	Description
`label`	string	no	Categorical classification (e.g., `correct` , `incorrect` , `partial` )
`score`	number	no	Numeric quality score (e.g., 0.0 - 1.0)
`explanation`	string	no	Freeform reasoning for the evaluation

At least one of

label

score

, or

explanation

should be present per evaluation.

字段	类型	是否必填	描述
`label`	字符串	否	分类标签（例如 `correct` 、 `incorrect` 、 `partial` ）
`score`	数字	否	数值质量分数（例如0.0 - 1.0）
`explanation`	字符串	否	评估的自由文本推理说明

每个评估至少应包含

label

、

score

或

explanation

中的一个。

Workflows

工作流

Run an experiment against a dataset

针对数据集运行实验

Find or create a dataset:

bash

ax datasets list
ax datasets export DATASET_ID --stdout | jq 'length'

Export the dataset examples:
bash
```
ax datasets export DATASET_ID
```
Process each example through your system, collecting outputs and evaluations

Build a runs file (JSON array) with

example_id

output

, and optional

evaluations

json

[
  {"example_id": "ex_001", "output": "4", "evaluations": {"correctness": {"label": "correct", "score": 1.0}}},
  {"example_id": "ex_002", "output": "Paris", "evaluations": {"correctness": {"label": "correct", "score": 1.0}}}
]

Create the experiment:

bash

ax experiments create --name "gpt-4o-baseline" --dataset-id DATASET_ID --file runs.json

Verify:
```
ax experiments get EXPERIMENT_ID
```

查找或创建数据集：

bash

ax datasets list
ax datasets export DATASET_ID --stdout | jq 'length'

导出数据集示例：
bash
```
ax datasets export DATASET_ID
```
通过你的系统处理每个示例，收集输出和评估结果

构建运行记录文件（JSON数组），包含

example_id

、

output

和可选的

evaluations

：

json

[
  {"example_id": "ex_001", "output": "4", "evaluations": {"correctness": {"label": "correct", "score": 1.0}}},
  {"example_id": "ex_002", "output": "Paris", "evaluations": {"correctness": {"label": "correct", "score": 1.0}}}
]

创建实验：

bash

ax experiments create --name "gpt-4o-baseline" --dataset-id DATASET_ID --file runs.json

验证：
```
ax experiments get EXPERIMENT_ID
```

Compare two experiments

对比两个实验

Export both experiments:

bash

ax experiments export EXPERIMENT_ID_A --stdout > a.json
ax experiments export EXPERIMENT_ID_B --stdout > b.json

Compare evaluation scores by

example_id

bash

# Average correctness score for experiment A
jq '[.[] | .evaluations.correctness.score] | add / length' a.json

# Same for experiment B
jq '[.[] | .evaluations.correctness.score] | add / length' b.json

Find examples where results differ:

bash

jq -s '.[0] as $a | .[1][] | {example_id, b_score: .evaluations.correctness.score, a_score: ($a[] | select(.example_id == .example_id) | .evaluations.correctness.score)}' a.json b.json

导出两个实验的运行记录：

bash

ax experiments export EXPERIMENT_ID_A --stdout > a.json
ax experiments export EXPERIMENT_ID_B --stdout > b.json

按

example_id

对比评估分数：

bash

# 实验A的平均正确性分数
jq '[.[] | .evaluations.correctness.score] | add / length' a.json

# 实验B的平均正确性分数
jq '[.[] | .evaluations.correctness.score] | add / length' b.json

找出结果不同的示例：

bash

jq -s '.[0] as $a | .[1][] | {example_id, b_score: .evaluations.correctness.score, a_score: ($a[] | select(.example_id == .example_id) | .evaluations.correctness.score)}' a.json b.json

Download experiment results for analysis

下载实验结果用于分析

ax experiments list --dataset-id DATASET_ID

-- find experiments

```
ax experiments export EXPERIMENT_ID
```
-- download to file

Parse:

jq '.[] | {example_id, score: .evaluations.correctness.score}' experiment_*/runs.json

ax experiments list --dataset-id DATASET_ID

—— 查找实验

```
ax experiments export EXPERIMENT_ID
```
—— 下载到文件

解析：

jq '.[] | {example_id, score: .evaluations.correctness.score}' experiment_*/runs.json

Pipe export to other tools

将导出结果管道到其他工具

bash

undefined

bash

undefined

Count runs

统计运行记录数量

ax experiments export EXPERIMENT_ID --stdout | jq 'length'

Extract all outputs

提取所有输出内容

ax experiments export EXPERIMENT_ID --stdout | jq '.[].output'

Get runs with low scores

获取分数较低的运行记录

ax experiments export EXPERIMENT_ID --stdout | jq '[.[] | select(.evaluations.correctness.score < 0.5)]'

Convert to CSV

转换为CSV格式

ax experiments export EXPERIMENT_ID --stdout | jq -r '.[] | [.example_id, .output, .evaluations.correctness.score] | @csv'

undefined

ax experiments export EXPERIMENT_ID --stdout | jq -r '.[] | [.example_id, .output, .evaluations.correctness.score] | @csv'

undefined

Troubleshooting

故障排除

Problem	Solution
`ax: command not found`	Check `~/.local/bin/ax` ; if missing: `uv tool install arize-ax-cli` (requires shell access to install packages)
`401 Unauthorized`	API key may not have access to this space. Verify the key and space ID are correct. Keys are scoped per space -- get the right one from https://app.arize.com/admin > API Keys.
`No profile found`	Run `ax profiles show --expand` to check; set `ARIZE_API_KEY` env var or write `~/.arize/config.toml`
`Experiment not found`	Verify experiment ID with `ax experiments list`
`Invalid runs file`	Each run must have `example_id` and `output` fields
`example_id mismatch`	Ensure `example_id` values match IDs from the dataset (export dataset to verify)
`No runs found`	Export returned empty -- verify experiment has runs via `ax experiments get`
`Dataset not found`	The linked dataset may have been deleted; check with `ax datasets list`

问题	解决方案
`ax: command not found`	检查 `~/.local/bin/ax` 是否存在；如果缺失，运行 `uv tool install arize-ax-cli` （需要Shell权限安装软件包）
`401 Unauthorized`	API密钥可能无权访问该空间。验证密钥和空间ID是否正确。密钥按空间划分权限——请从https://app.arize.com/admin > API Keys获取正确的密钥。
`No profile found`	运行 `ax profiles show --expand` 检查；设置 `ARIZE_API_KEY` 环境变量或编写 `~/.arize/config.toml` 配置文件
`Experiment not found`	使用 `ax experiments list` 验证实验ID是否正确
`Invalid runs file`	每个运行记录必须包含 `example_id` 和 `output` 字段
`example_id mismatch`	确保 `example_id` 的值与数据集中的ID匹配（可导出数据集进行验证）
`No runs found`	导出结果为空——通过 `ax experiments get` 验证实验是否包含运行记录
`Dataset not found`	关联的数据集可能已被删除；使用 `ax datasets list` 检查

Save Credentials for Future Use

保存凭证供后续使用

At the end of the session, if the user manually provided any of the following during this conversation (via AskQuestion response, pasted text, or inline values) and those values were NOT already loaded from a saved profile or environment variable, offer to save them for future use.

Credential	Where it gets saved
API key	`ax` profile at `~/.arize/config.toml`
Space ID	macOS/Linux: shell config ( `~/.zshrc` or `~/.bashrc` ) as `export ARIZE_SPACE_ID="..."` . Windows: user environment variable via `[System.Environment]::SetEnvironmentVariable('ARIZE_SPACE_ID', '...', 'User')`

Skip this entirely if:

The API key was already loaded from an existing profile or
```
ARIZE_API_KEY
```
env var
The space ID was already set via
```
ARIZE_SPACE_ID
```
env var
The user only used base64 project IDs (no space ID was needed)

How to offer: Use AskQuestion: "Would you like to save your Arize credentials so you don't have to enter them next time?" with options

"Yes, save them"

"No thanks"

If the user says yes:

API key — Check if

~/.arize/config.toml

exists. If it does, read it and update the

[auth]

section. If not, create it with this minimal content:

toml

[profile]
name = "default"

[auth]
api_key = "THE_API_KEY"

[output]
format = "table"

Verify with:

ax profiles show

Space ID — Persist the space ID as an environment variable:
macOS/Linux — Detect the user's shell config file (
```
~/.zshrc
```
for zsh,
```
~/.bashrc
```
for bash). Append:
bash
```
export ARIZE_SPACE_ID="THE_SPACE_ID"
```
Tell the user to run
```
source ~/.zshrc
```
(or restart their terminal) for it to take effect.
Windows (PowerShell) — Set a persistent user environment variable:
powershell
```
[System.Environment]::SetEnvironmentVariable('ARIZE_SPACE_ID', 'THE_SPACE_ID', 'User')
```
Tell the user to restart their terminal for it to take effect.

在会话结束时，如果用户在此期间手动提供了以下任何信息（通过询问响应、粘贴文本或内联值）且这些值并非从已保存的配置文件或环境变量中加载，请主动提供保存选项。

凭证	保存位置
API密钥	`ax` 配置文件 `~/.arize/config.toml`
空间ID	macOS/Linux：Shell配置文件（ `~/.zshrc` 或 `~/.bashrc` ），格式为 `export ARIZE_SPACE_ID="..."` 。Windows：用户环境变量，通过 `[System.Environment]::SetEnvironmentVariable('ARIZE_SPACE_ID', '...', 'User')` 设置

完全跳过此步骤的情况：

API密钥已从现有配置文件或
```
ARIZE_API_KEY
```
环境变量加载
空间ID已通过
```
ARIZE_SPACE_ID
```
环境变量设置
用户仅使用base64格式的项目ID（无需空间ID）

如何询问：使用询问用户："是否需要保存你的Arize凭证，以便下次无需重复输入？" 提供选项

"是，保存"

"不用了，谢谢"

。

如果用户选择是：

API密钥 —— 检查
```
~/.arize/config.toml
```
是否存在。如果存在，读取并更新
```
[auth]
```
部分。如果不存在，创建包含以下内容的最小配置文件：
toml
```
[profile]
name = "default"

[auth]
api_key = "你的API密钥"

[output]
format = "table"
```
验证：
```
ax profiles show
```
空间ID —— 将空间ID持久化为环境变量：
macOS/Linux —— 检测用户的Shell配置文件（zsh使用
```
~/.zshrc
```
，bash使用
```
~/.bashrc
```
）。添加：
bash
```
export ARIZE_SPACE_ID="你的空间ID"
```
告知用户运行
```
source ~/.zshrc
```
（或重启终端）以使设置生效。
Windows（PowerShell） —— 设置持久化用户环境变量：
powershell
```
[System.Environment]::SetEnvironmentVariable('ARIZE_SPACE_ID', '你的空间ID', 'User')
```
告知用户重启终端以使设置生效。

arize-experiment

Original

Translation

Arize Experiment Skill

Arize实验Skill

Concepts

核心概念

Prerequisites

前提条件

Install ax

安装ax

Verify environment

验证环境

Space ID and Project

空间ID与项目

List Experiments: ax experiments list

列出实验：ax experiments list

Flags

参数说明

Get Experiment: ax experiments get

获取实验详情：ax experiments get

Flags

参数说明

Response fields

响应字段

Export Experiment: ax experiments export

导出实验：ax experiments export

-> experiment_abc123_20260305_141500/runs.json

-> experiment_abc123_20260305_141500/runs.json

Flags

参数说明

REST vs Flight (--all)

REST与Flight（--all）对比

Create Experiment: ax experiments create

创建实验：ax experiments create

Flags

参数说明

Required columns in the runs file

运行记录文件的必填列

Delete Experiment: ax experiments delete

删除实验：ax experiments delete

Flags

参数说明

Experiment Run Schema

实验运行记录 Schema

Evaluation fields

评估字段

Workflows

工作流

Run an experiment against a dataset

针对数据集运行实验

Compare two experiments

对比两个实验

Download experiment results for analysis

下载实验结果用于分析

Pipe export to other tools

将导出结果管道到其他工具

Count runs

统计运行记录数量

Extract all outputs

提取所有输出内容

Get runs with low scores

获取分数较低的运行记录

Convert to CSV

转换为CSV格式

Troubleshooting

故障排除

Save Credentials for Future Use

保存凭证供后续使用

List Experiments:
`ax experiments list`

列出实验：
`ax experiments list`

Get Experiment:
`ax experiments get`

获取实验详情：
`ax experiments get`

Export Experiment:
`ax experiments export`

导出实验：
`ax experiments export`

REST vs Flight (
`--all`
)

REST与Flight（
`--all`
）对比

Create Experiment:
`ax experiments create`

创建实验：
`ax experiments create`

Delete Experiment:
`ax experiments delete`

删除实验：
`ax experiments delete`