Move benchmarks out of tests and run as a separate action

Author: bkeepers

.github/workflows/test.yml

@@ -22,15 +22,8 @@ jobs:
       - name: Install modules
         run: npm install
 
-      - name: Cache test data
-        id: test-cache
-        uses: actions/cache@v5
-        with:
-          path: .test-cache
-          key: ${{ runner.os }}-test-cache
-
       - name: Test build
-        run: npm run build:all
+        run: npm run build
 
       - name: Test
         run: npm run coverage
@@ -41,8 +34,29 @@ jobs:
           fail_ci_if_error: true
           files: ./coverage/coverage-final.json
 
-      - name: Upload NOAA summary
+  benchmarks:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Install modules
+        run: npm install
+
+      - name: Build
+        run: npm run build
+
+      - name: Cache benchmark data
+        id: benchmark-cache
+        uses: actions/cache@v5
+        with:
+          path: .test-cache
+          key: ${{ runner.os }}-benchmark-cache
+
+      - name: Run benchmarks
+        run: npm run benchmark
+
+      - name: Upload benchmark results
         uses: actions/upload-artifact@v6
         with:
-          name: noaa-benchmarks
-          path: noaa-benchmarks.csv
+          name: benchmarks
+          path: benchmarks/*.csv

.gitignore

@@ -8,3 +8,4 @@ junit.xml
 dist
 *.tsbuildinfo
 package-lock.json
+benchmarks/*.csv

benchmarks/noaa.ts

@@ -0,0 +1,116 @@
+import { expect } from 'vitest'
+import fs from 'fs/promises'
+import tidePredictor from '@neaps/tide-predictor'
+import db from '@neaps/tide-database'
+import { createWriteStream } from 'fs'
+import { join } from 'path'
+
+const __dirname = new URL('.', import.meta.url).pathname
+
+const stations = db.filter(
+  (station) =>
+    // TODO: Update this to test subordinate stations too.
+    // Need to switch from `getWaterLevelAtTime` to `getExtremesPrediction` and compare time/level.
+    station.type === 'reference' &&
+    station.source.source_url.includes('noaa.gov')
+)
+
+// Create a directory for test cache
+await fs.mkdir('./.test-cache', { recursive: true })
+
+interface Stat {
+  station: string
+  count: number
+  mae: number
+  rmse: number
+  bias: number
+}
+
+const stats: Stat[] = []
+
+console.log(`Testing tide predictions against ${stations.length} NOAA stations`)
+
+for (const station of stations) {
+  // Catch error and return no levels if failing to fetch data. There is a test later to ensure enough stations are tested.
+  const { levels } = await getStation(station.source.id).catch(() => ({
+    levels: {}
+  }))
+  const tideStation = tidePredictor(station.harmonic_constituents, {
+    phaseKey: 'phase_UTC'
+  })
+
+  // No predictions available
+  if (!levels.predictions) continue
+
+  let count = 0
+  let sumError = 0
+  let sumAbsError = 0
+  let sumSqError = 0
+
+  levels.predictions.forEach((prediction: { t: string; v: string }) => {
+    const expected = parseFloat(prediction.v)
+
+    const { level: actual } = tideStation.getWaterLevelAtTime({
+      time: new Date(prediction.t)
+    })
+
+    const error = actual - expected
+
+    count += 1
+    sumError += error
+    sumAbsError += Math.abs(error)
+    sumSqError += error * error
+  })
+
+  const mae = sumAbsError / count
+  const rmse = Math.sqrt(sumSqError / count)
+  const bias = sumError / count
+
+  stats.push({ station: station.source.id, count, mae, rmse, bias })
+  process.stdout.write('.')
+}
+
+// Write stats to file for later analysis
+const summary = createWriteStream(join(__dirname, 'noaa.csv'))
+summary.write('station,count,mae_m,rmse_m,bias_m\n')
+stats.forEach(({ station, count, mae, rmse, bias }) => {
+  summary.write(
+    [station, count, mae.toFixed(4), rmse.toFixed(4), bias.toFixed(4)].join(
+      ','
+    ) + '\n'
+  )
+})
+summary.end()
+
+// Baseline expectations based on current performance. The goal should be to move these toward zero over time.
+const maeValues = stats.map((s) => s.mae).sort((a, b) => a - b)
+const medianMAE = maeValues[Math.floor(stats.length / 2)]
+const p90MAE = maeValues[Math.floor(stats.length * 0.9)]
+const p95MAE = maeValues[Math.floor(stats.length * 0.95)]
+
+expect(medianMAE, 'MAE p50').toBeLessThan(0.03) // 3 cm
+expect(p90MAE, 'MAE p90').toBeLessThan(0.06) // 6 cm
+expect(p95MAE, 'MAE p95').toBeLessThan(0.08) // 8 cm
+expect(stats.length, 'Total stations').toBeGreaterThanOrEqual(1100) // Ensure enough stations were tested
+
+async function makeRequest(url: string) {
+  const res = await fetch(url)
+  if (!res.ok) throw new Error(`Request failed with status ${res.status}`)
+  return res.json()
+}
+
+async function getStation(station: string) {
+  const filePath = `./.test-cache/${station}.json`
+
+  try {
+    return await fs.readFile(filePath, 'utf-8').then((data) => JSON.parse(data))
+  } catch {
+    const levels = await makeRequest(
+      `https://api.tidesandcurrents.noaa.gov/api/prod/datagetter?date=recent&station=${station}&product=predictions&datum=MTL&time_zone=gmt&units=metric&format=json`
+    )
+
+    const data = { levels }
+    await fs.writeFile(filePath, JSON.stringify(data))
+    return data
+  }
+}

package.json

@@ -4,9 +4,10 @@
   "scripts": {
     "test": "vitest",
     "coverage": "vitest run --coverage",
-    "build:all": "npm run build --workspaces --if-present",
+    "build": "npm run build --workspaces --if-present",
     "lint": "eslint && prettier --check .",
-    "format": "prettier --write ."
+    "format": "prettier --write .",
+    "benchmarks": "node benchmarks/noaa.ts"
   },
   "devDependencies": {
     "@eslint/js": "^9.39.2",