POST
/api/v1/scorers/calibrateCLHF — evaluator/human agreement + threshold suggestion
Continuous learning from human feedback. Quantifies how well an evaluator agrees with human labels using chance-corrected Cohen's kappa, and recommends the score threshold that maximizes agreement. Supply `pairs` (human/machine pass-fail labels) and/or `scored` (humanPass + machineScore).
Authentication
Send Authorization: Bearer YOUR_API_KEY on every request. Generate API keys at /dashboard/api-keys.
Request body required
Example
{
"projectId": "00000000-0000-0000-0000-000000000000",
"scorerId": "string",
"pairs": [
{
"human": false,
"machine": false
}
],
"scored": [
{
"humanPass": false,
"machineScore": 0
}
],
"currentThreshold": 0
}Schema
{
"application/json": {
"schema": {
"type": "object",
"properties": {
"projectId": {
"type": "string",
"format": "uuid",
"pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$"
},
"scorerId": {
"type": "string",
"maxLength": 200
},
"pairs": {
"maxItems": 100000,
"type": "array",
"items": {
"type": "object",
"properties": {
"human": {
"type": "boolean"
},
"machine": {
"type": "boolean"
}
},
"required": [
"human",
"machine"
],
"additionalProperties": false
}
},
"scored": {
"maxItems": 100000,
"type": "array",
"items": {
"type": "object",
"properties": {
"humanPass": {
"type": "boolean"
},
"machineScore": {
"type": "number",
"minimum": 0,
"maximum": 1
}
},
"required": [
"humanPass",
"machineScore"
],
"additionalProperties": false
}
},
"currentThreshold": {
"type": "number",
"minimum": 0,
"maximum": 1
}
},
"additionalProperties": false
}
}
}Response
All status codes
200Agreement report and/or threshold suggestion
400No pairs or scored data provided
Code samples
cURL
curl -X POST \
https://evalguard.ai/api/v1/scorers/calibrate \
-H "Authorization: Bearer $EVALGUARD_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "projectId": "00000000-0000-0000-0000-000000000000", "scorerId": "string", "pairs": [ { "human": false, "machine": false } ], "scored": [ { "humanPass": false, "machineScore": 0 } ], "currentThreshold": 0 }'TypeScript
import { EvalGuard } from "@evalguard/sdk";
const client = new EvalGuard({ apiKey: process.env.EVALGUARD_API_KEY });
const response = await client.request({
method: "POST",
path: "/api/v1/scorers/calibrate",
body: {
"projectId": "00000000-0000-0000-0000-000000000000",
"scorerId": "string",
"pairs": [
{
"human": false,
"machine": false
}
],
"scored": [
{
"humanPass": false,
"machineScore": 0
}
],
"currentThreshold": 0
},
});
console.log(response);Python
from evalguard import EvalGuard
import os
client = EvalGuard(api_key=os.environ["EVALGUARD_API_KEY"])
response = client.request(
method="POST",
path="/api/v1/scorers/calibrate",
body={
"projectId": "00000000-0000-0000-0000-000000000000",
"scorerId": "string",
"pairs": [
{
"human": False,
"machine": False
}
],
"scored": [
{
"humanPass": False,
"machineScore": 0
}
],
"currentThreshold": 0
},
)
print(response)Go
package main
import (
"context"
"fmt"
"net/http"
"os"
"strings"
)
func main() {
body := strings.NewReader(`{"projectId":"00000000-0000-0000-0000-000000000000","scorerId":"string","pairs":[{"human":false,"machine":false}],"scored":[{"humanPass":false,"machineScore":0}],"currentThreshold":0}`)
req, _ := http.NewRequestWithContext(context.Background(), "POST", "https://evalguard.ai/api/v1/scorers/calibrate", body)
req.Header.Set("Authorization", "Bearer "+os.Getenv("EVALGUARD_API_KEY"))
req.Header.Set("Content-Type", "application/json")
resp, err := http.DefaultClient.Do(req)
if err != nil { panic(err) }
defer resp.Body.Close()
fmt.Println(resp.Status)
}Errors
400