Mar 4, 2024 · Mar 4, 2024
diff --git a/.../announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md b/.../announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md
  GPTQ & GGML allow PostgresML to fit larger models in less RAM. These
  algorithms perform inference significantly faster on NVIDIA, Apple and Intel
  hardware.
 featured:false
 tags:[engineering]
 image:".gitbook/assets/image (14).png"
 ---

 #Announcing GPTQ & GGML Quantized LLM support for Huggingface Transformers
diff --git a/pgml-cms/blog/announcing-support-for-aws-us-east-1-region.md b/pgml-cms/blog/announcing-support-for-aws-us-east-1-region.md
 ---
 description: >-
  We added aws us east 1 to our list of support aws regions.
 featured: false
 tags: [product]
 ---

 # Announcing Support for AWS us-east-1 Region

 <div align="left">
diff --git a/pgml-cms/blog/data-is-living-and-relational.md b/pgml-cms/blog/data-is-living-and-relational.md
  A common problem with data science and machine learning tutorials is the
  published and studied datasets are often nothing like what you’ll find in
  industry.
 featured: false
 tags: [engineering]
 ---

 # Data is Living and Relational
diff --git a/pgml-cms/blog/generating-llm-embeddings-with-open-source-models-in-postgresml.md b/pgml-cms/blog/generating-llm-embeddings-with-open-source-models-in-postgresml.md
 description:>-
  How to use the pgml.embed(...) function to generate embeddings with free and
  open source models in your own database.
 image:".gitbook/assets/blog_image_generating_llm_embeddings.png"
 features:true
 ---

 #Generating LLM embeddings with open source models in PostgresML
diff --git a/pgml-cms/blog/how-to-improve-search-results-with-machine-learning.md b/pgml-cms/blog/how-to-improve-search-results-with-machine-learning.md
  PostgresML makes it easy to use machine learning on your data and scale
  workloads horizontally in our cloud. One of the most common use cases is to
  improve search results.
 featured: true
 image: ".gitbook/assets/image (2) (2).png"
 tags: ["Engineering"]
 ---

 # How-to Improve Search Results with Machine Learning
diff --git a/...roducing-the-openai-switch-kit-move-from-closed-to-open-source-ai-in-minutes.md b/...roducing-the-openai-switch-kit-move-from-closed-to-open-source-ai-in-minutes.md
 ---
 featured: true
 tags: [engineering, product]
 image: https://postgresml.org/dashboard/static/images/open_source_ai_social_share.png
 description: >-
  Quickly and easily transition from the confines of the OpenAI APIs to higher
  quality embeddings and unrestricted text generation models.
 image: ".gitbook/assets/blog_image_switch_kit.png"
 ---

 # Introducing the OpenAI Switch Kit: Move from closed to open-source AI in minutes
diff --git a/pgml-cms/blog/postgres-full-text-search-is-awesome.md b/pgml-cms/blog/postgres-full-text-search-is-awesome.md
 description:>-
  If you want to improve your search results, don't rely on expensive O(n*m)
  word frequency statistics. Get new sources of data instead.
 image:".gitbook/assets/image (53).png"
 ---

 #Postgres Full Text Search is Awesome!
diff --git a/pgml-cms/blog/postgresml-is-going-multicloud.md b/pgml-cms/blog/postgresml-is-going-multicloud.md
 #PostgresML is going multicloud

 <divalign="left">

 <figure><imgsrc=".gitbook/assets/lev.jpg"alt="Author"width="100"><figcaption></figcaption></figure>

 </div>

 Lev Kokotov

 Jan 18, 2024


 We started PostgresML two years ago with the goal of making machine learning and AI accessible and easy for everyone.  To make this a reality, we needed to deploy PostgresML as closely as possible to our end users. With that goal mind, today we're proud to announce support for a new cloud provider: Azure.

 ###How we got here
diff --git a/pgml-cms/blog/speeding-up-vector-recall-5x-with-hnsw.md b/pgml-cms/blog/speeding-up-vector-recall-5x-with-hnsw.md
  HNSW indexing is the latest upgrade in vector recall performance. In this post
  we announce our updated SDK that utilizes HNSW indexing to give world class
  performance in vector search.
 tags: [engineering]
 featured: true
 image: ".gitbook/assets/blog_image_hnsw.png"
 ---

 # Speeding up vector recall 5x with HNSW
diff --git a/pgml-cms/blog/using-postgresml-with-django-and-embedding-search.md b/pgml-cms/blog/using-postgresml-with-django-and-embedding-search.md
 ---
 tags:
  - engineering
 description: >-
  An example application using PostgresML and Django to build embedding based search.
 tags: [engineering]
 ---

 # Using PostgresML with Django and embedding search

 <div align="left">

 <figure><img src=".gitbook/assets/lev.jpg" alt="Author" width="100"><figcaption></figcaption></figure>

 </div>

 Lev Kokotov

 Feb 15, 2024

 Building web apps on top of PostgresML allows anyone to integrate advanced machine learning and AI features into their products without much work or needing to understand how it really works. In this blog post, we'll talk about building a classic to-do Django app, with the spicy addition of semantic search powered by embedding models running inside your PostgreSQL database.

 ### Getting the code

 Djago Rest Framework provides the bulk of the implementation. We just added a `ModelViewSet` for the `TodoItem` model, with just one addition: a search endpoint. The search endpoint required us to write a bit of SQL to embed the search query and accept a few filters, but the core of it can be summarized in a single annotation on the query set:

 <pre class="language-python"><code class="lang-python"><strong>results = TodoItem.objects.annotate(
 </strong>    similarity=RawSQL(
 ```python
 results = TodoItem.objects.annotate(
    similarity=RawSQL(
        "pgml.embed('intfloat/e5-small', %s)::vector(384) &#x3C;=> embedding",
        [query],
    )
 ).order_by("similarity")
 </code></pre>
 ```

 This single line of SQL does quite a bit:


 Before going forward, make sure you have the app running either locally or in a cloud provider of your choice. If hosting it somewhere, replace `localhost:8000` with the URL and port of your service.

 The simplest way to interact with it is to use cURL oranHTTP client of your preference. If running in debug mode locally, the Rest Framework provides a nice web UI which you can access on [http://localhost:8000/api/todos/](http://localhost:8000/api/todo/) using a browser.
 The simplest way to interact with it is to use cURL oryour preferredHTTP client. If running in debug mode locally, the Rest Framework provides a nice web UI which you can access on [http://localhost:8000/api/todo/](http://localhost:8000/api/todo/) using a browser.

 To create a to-do item with cURL, you can just run this:


 The embedding contains 384 floating point numbers; we removed most of them in this blog post to make sure it fits on the page.

 You can try creating multiple to-do items for fun and profit. If the description is changed, so will the embedding, demonstrating how the `intfloat/e5-small` modelis understanding the semantic meaning of your text.
 You can try creating multiple to-do items for fun and profit. If the description is changed, so will the embedding, demonstrating how the `intfloat/e5-small` modelunderstands the semantic meaning of your text.

 ### Searching

 Once you have a few embeddings and to-dos stored in your database, the fun part of searching can begin. In a typical search example with PostgreSQL, you'd now be using `tsvector` to keyword match your to-dos to the search query with term frequency. That's a good technique, but semantic search is better.

 We've created a simplesearch endpointthataccepts a query, a completed to-do filter, and a limit. To use it, you can justdo this:
 Oursearch endpoint accepts a query, a completed to-do filter, and a limit. To use it, you can justrun this:

 ```bash
 curl \
 If you've created a bunch of different to-do items, you should get only one search result back, and exactly the one you were expecting:

 ```json
 "Make a New Year resolution list"
 "Make a New Year resolution"
 ```

 You can increase the `limit` to something larger and you should get more documents, in decreasing order of relevance.
Original file line number	Diff line number	Diff line change
Expand Up		@@ -3,6 +3,9 @@ description: >-
		GPTQ & GGML allow PostgresML to fit larger models in less RAM. These
		algorithms perform inference significantly faster on NVIDIA, Apple and Intel
		hardware.
		featured:false
		tags:[engineering]
		image:".gitbook/assets/image (14).png"
		---

		#Announcing GPTQ & GGML Quantized LLM support for Huggingface Transformers
Expand Down
Original file line number	Diff line number	Diff line change
		@@ -1,3 +1,10 @@
		---
		description: >-
		We added aws us east 1 to our list of support aws regions.
		featured: false
		tags: [product]
		---

		# Announcing Support for AWS us-east-1 Region

		<div align="left">
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -3,6 +3,8 @@ description: >-
		A common problem with data science and machine learning tutorials is the
		published and studied datasets are often nothing like what you’ll find in
		industry.
		featured: false
		tags: [engineering]
		---

		# Data is Living and Relational
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -2,6 +2,8 @@
		description:>-
		How to use the pgml.embed(...) function to generate embeddings with free and
		open source models in your own database.
		image:".gitbook/assets/blog_image_generating_llm_embeddings.png"
		features:true
		---

		#Generating LLM embeddings with open source models in PostgresML
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -3,6 +3,9 @@ description: >-
		PostgresML makes it easy to use machine learning on your data and scale
		workloads horizontally in our cloud. One of the most common use cases is to
		improve search results.
		featured: true
		image: ".gitbook/assets/image (2) (2).png"
		tags: ["Engineering"]
		---

		# How-to Improve Search Results with Machine Learning
Expand Down
Original file line number	Diff line number	Diff line change
		@@ -1,8 +1,11 @@
		---
		featured: true
		tags: [engineering, product]
		image: https://postgresml.org/dashboard/static/images/open_source_ai_social_share.png
		description: >-
		Quickly and easily transition from the confines of the OpenAI APIs to higher
		quality embeddings and unrestricted text generation models.
		image: ".gitbook/assets/blog_image_switch_kit.png"
		---

		# Introducing the OpenAI Switch Kit: Move from closed to open-source AI in minutes
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -2,6 +2,7 @@
		description:>-
		If you want to improve your search results, don't rely on expensive O(n*m)
		word frequency statistics. Get new sources of data instead.
		image:".gitbook/assets/image (53).png"
		---

		#Postgres Full Text Search is Awesome!
Expand Down
Original file line number	Diff line number	Diff line change
		@@ -1,5 +1,16 @@
		#PostgresML is going multicloud

		<divalign="left">

		<figure><imgsrc=".gitbook/assets/lev.jpg"alt="Author"width="100"><figcaption></figcaption></figure>

		</div>

		Lev Kokotov

		Jan 18, 2024


		We started PostgresML two years ago with the goal of making machine learning and AI accessible and easy for everyone. To make this a reality, we needed to deploy PostgresML as closely as possible to our end users. With that goal mind, today we're proud to announce support for a new cloud provider: Azure.

		###How we got here
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -3,6 +3,9 @@ description: >-
		HNSW indexing is the latest upgrade in vector recall performance. In this post
		we announce our updated SDK that utilizes HNSW indexing to give world class
		performance in vector search.
		tags: [engineering]
		featured: true
		image: ".gitbook/assets/blog_image_hnsw.png"
		---

		# Speeding up vector recall 5x with HNSW
Expand Down
Original file line number	Diff line number	Diff line change
		@@ -1,10 +1,21 @@
		---
		tags:
		- engineering
		description: >-
		An example application using PostgresML and Django to build embedding based search.
		tags: [engineering]
		---

		# Using PostgresML with Django and embedding search

		<div align="left">

		<figure><img src=".gitbook/assets/lev.jpg" alt="Author" width="100"><figcaption></figcaption></figure>

		</div>

		Lev Kokotov

		Feb 15, 2024

		Building web apps on top of PostgresML allows anyone to integrate advanced machine learning and AI features into their products without much work or needing to understand how it really works. In this blog post, we'll talk about building a classic to-do Django app, with the spicy addition of semantic search powered by embedding models running inside your PostgreSQL database.

		### Getting the code
Expand DownExpand Up		@@ -56,13 +67,14 @@ And that's it! In just a few lines of code, we're generating and storing high qu

		Djago Rest Framework provides the bulk of the implementation. We just added a `ModelViewSet` for the `TodoItem` model, with just one addition: a search endpoint. The search endpoint required us to write a bit of SQL to embed the search query and accept a few filters, but the core of it can be summarized in a single annotation on the query set:

		<pre class="language-python"><code class="lang-python"><strong>results = TodoItem.objects.annotate(
		</strong> similarity=RawSQL(
		```python
		results = TodoItem.objects.annotate(
		similarity=RawSQL(
		"pgml.embed('intfloat/e5-small', %s)::vector(384) <=> embedding",
		[query],
		)
		).order_by("similarity")
		</code></pre>
		```

		This single line of SQL does quite a bit:

Expand All		@@ -76,7 +88,7 @@ All of this happens inside PostgresML. Our Django app doesn't need to implement

		Before going forward, make sure you have the app running either locally or in a cloud provider of your choice. If hosting it somewhere, replace `localhost:8000` with the URL and port of your service.

		The simplest way to interact with it is to use cURL oranHTTP client of your preference. If running in debug mode locally, the Rest Framework provides a nice web UI which you can access on [http://localhost:8000/api/todos/](http://localhost:8000/api/todo/) using a browser.
		The simplest way to interact with it is to use cURL oryour preferredHTTP client. If running in debug mode locally, the Rest Framework provides a nice web UI which you can access on [http://localhost:8000/api/todo/](http://localhost:8000/api/todo/) using a browser.

		To create a to-do item with cURL, you can just run this:

Expand All		@@ -103,13 +115,13 @@ In return, you'll get your to-do item alongside the embedding of the `descriptio

		The embedding contains 384 floating point numbers; we removed most of them in this blog post to make sure it fits on the page.

		You can try creating multiple to-do items for fun and profit. If the description is changed, so will the embedding, demonstrating how the `intfloat/e5-small` modelis understanding the semantic meaning of your text.
		You can try creating multiple to-do items for fun and profit. If the description is changed, so will the embedding, demonstrating how the `intfloat/e5-small` modelunderstands the semantic meaning of your text.

		### Searching

		Once you have a few embeddings and to-dos stored in your database, the fun part of searching can begin. In a typical search example with PostgreSQL, you'd now be using `tsvector` to keyword match your to-dos to the search query with term frequency. That's a good technique, but semantic search is better.

		We've created a simplesearch endpointthataccepts a query, a completed to-do filter, and a limit. To use it, you can justdo this:
		Oursearch endpoint accepts a query, a completed to-do filter, and a limit. To use it, you can justrun this:

		```bash
		curl \
Expand All		@@ -122,7 +134,7 @@ curl \
		If you've created a bunch of different to-do items, you should get only one search result back, and exactly the one you were expecting:

		```json
		"Make a New Year resolution list"
		"Make a New Year resolution"
		```

		You can increase the `limit` to something larger and you should get more documents, in decreasing order of relevance.
Expand Down