Multimodal search tutorial

docs/pages/tutorials/multimodal-search.md

@@ -0,0 +1,11 @@
+---
+layout: default
+title: Multimodal Search
+parent: Tutorials
+---
+
+<!-- 
+jupyter nbconvert --to html --template basic amazon-products-multi-modal-search.ipynb --stdout > ../../docs/pages/tutorials/multimodal-search-raw.html 
+-->
+
+{% include_relative multimodal-search-raw.html %}

docs/pages/tutorials/tutorials.md

@@ -0,0 +1,10 @@
+---
+layout: default
+title: Tutorials
+nav_order: 6
+has_children: true
+permalink: tutorials
+---
+
+# Tutorials
+{: .no_toc }

examples/tutorial-notebooks/.gitignore

@@ -0,0 +1,134 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# Data
+*.json
+*.json.gz
+*.b

examples/tutorial-notebooks/Dockerfile

@@ -0,0 +1,2 @@
+FROM docker.elastic.co/elasticsearch/elasticsearch:7.9.2-amd64
+RUN elasticsearch-plugin install -b https://github.com/alexklibisz/elastiknn/releases/download/0.1.0-PRE49/elastiknn-0.1.0-PRE49_es7.9.2.zip

examples/tutorial-notebooks/amazonutils.py

@@ -0,0 +1,49 @@
+from IPython.display import display, HTML, Image
+from itertools import islice
+import numpy as np
+import json
+import gzip 
+import array
+
+def iter_products(fname):
+  with gzip.open(fname, 'r') as g:
+    for l in g:
+      yield eval(l)
+
+def iter_vectors(fname):
+  with open(fname, 'rb') as f:
+    while True:
+      try:
+        asin = f.read(10)
+        a = array.array('f')
+        a.fromfile(f, 4096)
+        yield (asin.decode(), a.tolist())
+      except EOFError:
+        break
+
+def iter_vectors_reduced(fname, dims=1024, samples=10000):
+  sumarr = np.zeros(4096) * 1.0
+  for (_, v) in islice(iter_vectors(fname), samples):
+    sumarr -= np.array(v)
+  ii = np.argsort(sumarr)[:dims]
+
+  def f(fname):
+    for (asin, vec) in iter_vectors(fname):
+      yield (asin, np.array(vec)[ii].tolist())
+
+  return f
+
+def display_hits(res):
+    print(f"Found {res['hits']['total']['value']} hits in {res['took']} ms. Showing top {len(res['hits']['hits'])}.")
+    print("")
+    for hit in res['hits']['hits']:
+        s = hit['_source']
+        print(f"Title   {s.get('title', None)}"[:80] + "...")
+        if 'description' in s:
+          print(f"Desc    {s['description']}"[:80] + "...")
+        if 'price' in s:
+          print(f"Price   {s['price']}")
+        print(f"ID      {s.get('asin', None)}")
+        print(f"Score   {hit.get('_score', None)}")
+        display(Image(s.get("imUrl"), width=128))
+        print("")

examples/tutorial-notebooks/docker-compose.yml

@@ -0,0 +1,31 @@
+version: "3.8"
+
+services:
+  elasticsearch_master:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    container_name: elasticsearch_master
+    environment:
+      - node.name=elasticsearch_master
+      - cluster.name=docker-cluster
+      - cluster.initial_master_nodes=elasticsearch_master
+      - node.master=true
+      - bootstrap.memory_lock=true
+      - http.cors.enabled=true
+      - http.cors.allow-origin=*
+      - ES_JAVA_OPTS=-Xms4G -Xmx4G
+    ports:
+      - "9200:9200"
+    volumes:
+      - data01:/usr/share/elasticsearch/data
+    ulimits:
+      nofile:
+        soft: 65536
+        hard: 65536
+      memlock:
+        soft: -1
+        hard: -1
+volumes:
+  data01:
+    driver: local

examples/tutorial-notebooks/requirements.txt

@@ -0,0 +1,4 @@
+matplotlib
+jupyter
+tqdm
+elasticsearch