yujunjun/graphrag
1
0
Fork 0
mirror of https://github.com/microsoft/graphrag.git synced 2025-09-25 16:16:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Deploying to gh-pages from @ microsoft/graphrag@6ee349a4d9 🚀

Browse Source
This commit is contained in:
AlonsoGuevara 2024-06-28 23:03:53 +00:00
parent 677729407f
commit 0a3adefc5e
25 changed files with 71 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
data/operation_dulce/dataset.zip
View File

Binary file not shown.

6
index.html
View File

@ -301,7 +301,9 @@ Figure 1: An LLM-generated knowledge graph built using GPT-4 Turbo.
<p>GraphRAG is a structured, hierarchical approach to Retrieval Augmented Generation (RAG), as opposed to naive semantic-search
approaches using plain text snippets. The GraphRAG process involves extracting a knowledge graph out of raw text, building a community hierarchy, generating summaries for these communities, and then leveraging these structures when perform RAG-based tasks.</p>
<p>To learn more about GraphRAG and how it can be used to enhance your LLMs ability to reason about your private data, please visit the <a href="https://www.microsoft.com/en-us/research/blog/graphrag-unlocking-llm-discovery-on-narrative-private-data/">Microsoft Research Blog Post</a>.</p>
<h2>Get Started 🚀</h2>
<h2>Solution Accelerator 🚀</h2>
<p>To quickstart the GraphRAG system we recommend trying the <a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a> package. This provides a user-friendly end-to-end experience with Azure resources.</p>
<h2>Get Started with GraphRAG 🚀</h2>
<p>To start using GraphRAG, check out the <a href="posts/get_started"><em>Get Started</em></a> guide.
For a deeper dive into the main sub-systems, please visit the docpages for the <a href="posts/index/overview">Indexer</a> and <a href="posts/query/overview">Query</a> packages.</p>
<h2>GraphRAG vs Baseline RAG 🔍</h2>
@ -346,6 +348,8 @@ We strongly recommend to fine-tune your prompts following the <a href="posts/ind
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/custom/index.html
View File

@ -485,6 +485,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/env_vars/index.html
View File

@ -1247,6 +1247,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/init/index.html
View File

@ -339,6 +339,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/json_yaml/index.html
View File

@ -492,6 +492,8 @@ API_KEY=some_api_key
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/overview/index.html
View File

@ -316,6 +316,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/config/template/index.html
View File

@ -475,6 +475,8 @@ the <code>--root</code> parameter on your Indexing Pipeline execution.</p>
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/developing/index.html
View File

@ -404,6 +404,8 @@ to reduce concurrency. Please refer to the <a href="../config/overview">Configur
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

38
posts/get_started/index.html
View File

@ -295,7 +295,9 @@ a {
<p>👉 <a href="https://github.com/Azure-Samples/graphrag-accelerator">Use the GraphRAG Accelerator solution</a> <br>
👉 <a href="https://pypi.org/project/graphrag/">Install from pypi</a>. <br>
👉 <a href="/posts/developing">Use it from source</a><br></p>
<h1>Top-Level Packages</h1>
<h2>Quickstart</h2>
<p>To get started with the GraphRAG system we recommend trying the <a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a> package. This provides a user-friendly end-to-end experience with Azure resources.</p>
<h1>Top-Level Modules</h1>
<p><a href="/posts/index/overview">Indexing Pipeline Overview</a><br>
<a href="/posts/query/overview">Query Engine Overview</a></p>
<h1>Overview</h1>
@ -304,9 +306,9 @@ It shows how to use the system to index some text, and then use the indexed data
<h1>Install GraphRAG</h1>
<div style="position: relative">
<pre class="language-bash"><code id="code-27" class="language-bash">pip <span class="token function">install</span> graphrag</code></pre>
<pre class="language-bash"><code id="code-33" class="language-bash">pip <span class="token function">install</span> graphrag</code></pre>
<button class="code-copy " data-clipboard-target="#code-27" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-33" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -315,18 +317,18 @@ It shows how to use the system to index some text, and then use the indexed data
<p>First let's get a sample dataset ready:</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-37" class="language-sh"><span class="token function">mkdir</span> <span class="token parameter variable">-p</span> ./ragtest/input</code></pre>
<pre class="language-sh"><code id="code-43" class="language-sh"><span class="token function">mkdir</span> <span class="token parameter variable">-p</span> ./ragtest/input</code></pre>
<button class="code-copy " data-clipboard-target="#code-37" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-43" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
<p>Now let's get a copy of A Christmas Carol by Charles Dickens from a trusted source</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-41" class="language-sh"><span class="token function">curl</span> https://www.gutenberg.org/cache/epub/24022/pg24022.txt <span class="token operator">></span> ./ragtest/input/book.txt</code></pre>
<pre class="language-sh"><code id="code-47" class="language-sh"><span class="token function">curl</span> https://www.gutenberg.org/cache/epub/24022/pg24022.txt <span class="token operator">></span> ./ragtest/input/book.txt</code></pre>
<button class="code-copy " data-clipboard-target="#code-41" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-47" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -337,9 +339,9 @@ It shows how to use the system to index some text, and then use the indexed data
Since we have already configured a directory named .ragtest` in the previous step, we can run the following command:</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-54" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.index <span class="token parameter variable">--init</span> <span class="token parameter variable">--root</span> ./ragtest</code></pre>
<pre class="language-sh"><code id="code-60" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.index <span class="token parameter variable">--init</span> <span class="token parameter variable">--root</span> ./ragtest</code></pre>
<button class="code-copy " data-clipboard-target="#code-54" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-60" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -356,12 +358,12 @@ Since we have already configured a directory named .ragtest` in the previous ste
<p>In addition, Azure OpenAI users should set the following variables in the settings.yaml file. To find the appropriate sections, just search for the <code>llm:</code> configuration, you should see two sections, one for the chat endpoint and one for the embeddings endpoint. Here is an example of how to configure the chat endpoint:</p>
<div style="position: relative">
<pre class="language-yaml"><code id="code-82" class="language-yaml"><span class="token key atrule">type</span><span class="token punctuation">:</span> azure_openai_chat <span class="token comment"># Or azure_openai_embedding for embeddings</span>
<pre class="language-yaml"><code id="code-88" class="language-yaml"><span class="token key atrule">type</span><span class="token punctuation">:</span> azure_openai_chat <span class="token comment"># Or azure_openai_embedding for embeddings</span>
<span class="token key atrule">api_base</span><span class="token punctuation">:</span> https<span class="token punctuation">:</span>//&lt;instance<span class="token punctuation">></span>.openai.azure.com
<span class="token key atrule">api_version</span><span class="token punctuation">:</span> 2024<span class="token punctuation">-</span>02<span class="token punctuation">-</span>15<span class="token punctuation">-</span>preview <span class="token comment"># You can customize this for other versions</span>
<span class="token key atrule">deployment_name</span><span class="token punctuation">:</span> &lt;azure_model_deployment_name<span class="token punctuation">></span></code></pre>
<button class="code-copy " data-clipboard-target="#code-82" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-88" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -374,9 +376,9 @@ Since we have already configured a directory named .ragtest` in the previous ste
<p>Finally we'll run the pipeline!</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-106" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.index <span class="token parameter variable">--root</span> ./ragtest</code></pre>
<pre class="language-sh"><code id="code-112" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.index <span class="token parameter variable">--root</span> ./ragtest</code></pre>
<button class="code-copy " data-clipboard-target="#code-106" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-112" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -389,24 +391,24 @@ Once the pipeline is complete, you should see a new folder called <code>./ragtes
<p>Here is an example using Global search to ask a high-level question:</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-125" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.query <span class="token punctuation">\</span>
<pre class="language-sh"><code id="code-131" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.query <span class="token punctuation">\</span>
<span class="token parameter variable">--root</span> ./ragtest <span class="token punctuation">\</span>
<span class="token parameter variable">--method</span> global <span class="token punctuation">\</span>
<span class="token string">"What are the top themes in this story?"</span></code></pre>
<button class="code-copy " data-clipboard-target="#code-125" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-131" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
<p>Here is an example using Local search to ask a more specific question about a particular character:</p>
<div style="position: relative">
<pre class="language-sh"><code id="code-129" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.query <span class="token punctuation">\</span>
<pre class="language-sh"><code id="code-135" class="language-sh">python <span class="token parameter variable">-m</span> graphrag.query <span class="token punctuation">\</span>
<span class="token parameter variable">--root</span> ./ragtest <span class="token punctuation">\</span>
<span class="token parameter variable">--method</span> <span class="token builtin class-name">local</span> <span class="token punctuation">\</span>
<span class="token string">"Who is Scrooge, and what are his main relationships?"</span></code></pre>
<button class="code-copy " data-clipboard-target="#code-129" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<button class="code-copy " data-clipboard-target="#code-135" style="position: absolute; top: 7.5px; right: 6px; padding-top: 3px; cursor: pointer; outline: none; opacity: 0.8;" title="Copy">
<span style="display:inline-block;background:url(https://api.iconify.design/mdi/content-copy.svg) no-repeat center center / contain;width: 16px; height: 16px;" class=""></span>
</button>
</div>
@ -428,6 +430,8 @@ Once the pipeline is complete, you should see a new folder called <code>./ragtes
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/index/0-architecture/index.html
View File

@ -336,6 +336,8 @@ This allows our indexer to be more resilient to network issues, to act idempoten
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/index/1-default_dataflow/index.html
View File

@ -377,6 +377,8 @@ Entities and Relationships are extracted at once in our <em>entity_extract</em>
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/index/2-cli/index.html
View File

@ -328,6 +328,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/index/overview/index.html
View File

@ -381,6 +381,8 @@ pipeline_result <span class="token operator">=</span> outputs<span class="token
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/prompt_tuning/auto_prompt_tuning/index.html
View File

@ -386,6 +386,8 @@ After that, it uses one of the following selection methods to pick a sample to w
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/prompt_tuning/manual_prompt_tuning/index.html
View File

@ -346,6 +346,8 @@ The default value is</p>
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

4
posts/prompt_tuning/overview/index.html
View File

@ -293,7 +293,7 @@ a {
<h2>Default Prompts</h2>
<p>The default prompts are the simplest way to get started with the GraphRAG system. It is designed to work out-of-the-box with minimal configuration. You can find more detail about these prompts in the following links:</p>
<ul>
<li>[Entity/Relationship Extraction] (http://github.com/microsoft/graphrag/blob/main/graphrag/index/graph/extractors/graph/prompts.py)</li>
<li><a href="http://github.com/microsoft/graphrag/blob/main/graphrag/index/graph/extractors/graph/prompts.py">Entity/Relationship Extraction</a></li>
<li><a href="http://github.com/microsoft/graphrag/blob/main/graphrag/index/graph/extractors/summarize/prompts.py">Entity/Relationship Description Summarization</a></li>
<li><a href="http://github.com/microsoft/graphrag/blob/main/graphrag/index/graph/extractors/claims/prompts.py">Claim Extraction</a></li>
<li><a href="http://github.com/microsoft/graphrag/blob/main/graphrag/index/graph/extractors/community_reports/prompts.py">Community Reports</a></li>
@ -319,6 +319,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/0-global_search/index.html
View File

@ -332,6 +332,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/1-local_search/index.html
View File

@ -324,6 +324,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/2-question_generation/index.html
View File

@ -322,6 +322,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/3-cli/index.html
View File

@ -350,6 +350,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/notebooks/global_search_nb/index.html
View File

@ -487,6 +487,8 @@ result<span class="token punctuation">.</span>context_data<span class="token pun
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/notebooks/local_search_nb/index.html
View File

@ -647,6 +647,8 @@ candidate_questions <span class="token operator">=</span> <span class="token key
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/notebooks/overview/index.html
View File

@ -312,6 +312,8 @@ a {
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>

2
posts/query/overview/index.html
View File

@ -322,6 +322,8 @@ It is responsible for the following tasks:</p>
<a href="https://www.microsoft.com" id="copyright"></a>
|
<a href="https://github.com/microsoft/graphrag">GitHub</a>
|
<a href="https://github.com/Azure-Samples/graphrag-accelerator">Solution Accelerator</a>
</footer>
</body>
</html>