@mastra/evals 0.1.0-alpha.22 → 0.1.0-alpha.24

package/CHANGELOG.md

@@ -1,5 +1,19 @@
 # @mastra/evals
 
+## 0.1.0-alpha.24
+
+### Patch Changes
+
+- Updated dependencies [73d112c]
+  - @mastra/core@0.1.27-alpha.82
+
+## 0.1.0-alpha.23
+
+### Patch Changes
+
+- Updated dependencies [9fb3039]
+  - @mastra/core@0.1.27-alpha.81
+
 ## 0.1.0-alpha.22
 
 ### Patch Changes

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mastra/evals",
-  "version": "0.1.0-alpha.22",
+  "version": "0.1.0-alpha.24",
   "description": "",
   "type": "module",
   "main": "dist/index.js",
@@ -38,7 +38,7 @@
     "sentiment": "^5.0.2",
     "string-similarity": "^4.0.4",
     "zod": "^3.24.1",
-    "@mastra/core": "0.1.27-alpha.80"
+    "@mastra/core": "0.1.27-alpha.82"
   },
   "devDependencies": {
     "@babel/preset-env": "^7.26.0",

package/src/metrics/llm/answer-relevancy/index.test.ts

@@ -109,13 +109,13 @@ describe(
       expect(result.score).toBeCloseTo(testCases[0].expectedResult.score, 1);
     });
 
-  it('should be able to measure a prompt with mostly relevant information', async () => {
-    const result = await metric.measure(testCases[1].input, testCases[1].output);
-    const expectedScore = testCases[1].expectedResult.score;
-    const difference = Math.abs(result.score - expectedScore);
+    it('should be able to measure a prompt with mostly relevant information', async () => {
+      const result = await metric.measure(testCases[1].input, testCases[1].output);
+      const expectedScore = testCases[1].expectedResult.score;
+      const difference = Math.abs(result.score - expectedScore);
 
-    expect(Math.round(difference * 10) / 10).toBeLessThanOrEqual(0.1);
-  });
+      expect(Math.round(difference * 10) / 10).toBeLessThanOrEqual(0.1);
+    });
 
     it('should be able to measure a prompt with partial relevance', async () => {
       const result = await metric.measure(testCases[2].input, testCases[2].output);

package/src/metrics/llm/bias/index.test.ts

@@ -50,28 +50,26 @@ const modelConfig: ModelConfig = {
   apiKey: process.env.OPENAI_API_KEY,
 };
 
-describe(
-  'BiasMetric',
-  () => {
-    const metric = new BiasMetric(modelConfig);
+describe('BiasMetric', () => {
+  const metric = new BiasMetric(modelConfig);
 
-    it('should be able to measure a prompt that is biased', async () => {
-      const result = await metric.measure(testCases[0].input, testCases[0].output);
-      expect(result.score).toBeCloseTo(testCases[0].expectedResult.score, 1);
-    });
+  it('should be able to measure a prompt that is biased', async () => {
+    const result = await metric.measure(testCases[0].input, testCases[0].output);
+    expect(result.score).toBeCloseTo(testCases[0].expectedResult.score, 1);
+  }, 10000);
 
-    it('should be able to measure a prompt that is almost not biased', async () => {
-      const result = await metric.measure(testCases[1].input, testCases[1].output);
-      expect(result.score).toBeLessThan(0.5);
-    });
+  it('should be able to measure a prompt that is almost not biased', async () => {
+    const result = await metric.measure(testCases[1].input, testCases[1].output);
+    expect(result.score).toBeLessThan(0.5);
+  }, 10000);
 
   it('should be able to measure a prompt that is mildly biased but actually not', async () => {
     const result = await metric.measure(testCases[2].input, testCases[2].output);
     expect(result.score).toBe(0);
-  });
+  }, 10000);
 
   it('should be able to measure a prompt that is mildly biased', async () => {
     const result = await metric.measure(testCases[3].input, testCases[3].output);
     expect(result.score).toBeLessThan(0.8);
-  });
+  }, 10000);
 });