add int4

.github/workflows/run-elasticsearch-int4-linux.yml

@@ -0,0 +1,56 @@
+name: Run Elasticsearch (int4) on Linux
+
+on:
+  push:
+    branches:
+    - main
+    paths:
+    - run-elasticsearch.ipynb
+  pull_request:
+    branches:
+    - main
+  workflow_dispatch:
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+    - name: Check Hardware Spec
+      run: /bin/bash ./scripts/get_hardware_info.sh
+    - name: Cache dataset
+      uses: actions/cache@v4
+      with:
+        path: |
+          dataset
+        key: ${{ runner.os }}-dataset
+    - name: Download data
+      run: /bin/bash ./scripts/setup.sh gha
+    - name: Set up Python 3.10
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.10'
+    - name: Cache pip modules
+      uses: actions/cache@v4
+      with:
+        path: ~/.cache/pip
+        key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements.txt') }}
+        restore-keys: |
+          ${{ runner.os }}-pip-
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
+    - name: Run ANN search
+      run: papermill run-elasticsearch.ipynb output-elasticsearch-linux.ipynb --log-output
+      env:
+        PRODUCT_VARIANT: "int4"
+        SETTING_QUANTIZATION: "int4"
+    - name: 'Upload Results'
+      uses: actions/upload-artifact@v4
+      with:
+        name: elasticsearch-int4-results
+        path: results.json
+

run-elasticsearch.ipynb

@@ -260,6 +260,8 @@
     "        knn_type = \"flat\"\n",
     "    if config.quantization == \"int8\":\n",
     "        knn_type = \"int8_hnsw\"\n",
+    "    elif config.quantization == \"int4\":\n",
+    "        knn_type = \"int4_hnsw\"\n",
     "    else:\n",
     "        knn_type = \"hnsw\"\n",
     "    print(F\"Creating {config.index_name} with {knn_type}... \", end=\"\")\n",