Skip to content

Commit 7ce8e9c

Browse files
ViktoriaNovViktoriaNov
committed
changes in the ‘quick start’ section
1 parent 702f18b commit 7ce8e9c

File tree

5 files changed

+57
-21
lines changed

5 files changed

+57
-21
lines changed

‎docs/_build/html/_static/styles/front_page.css‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -197,7 +197,7 @@ code {
197197
h1 { font-size: 1.5rem; }
198198
.badges img { height: 18px; }
199199
}
200-
/* Fix Sphinx weird highlighting for custom code blocks */
200+
201201
pre, code {
202202
background: #2d2d2d !important;
203203
color: #eee !important;

‎docs/_build/html/genindex.html‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
<title>Index &#8212; LLMSQL 0.2.0 documentation</title>
88
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8e8a900e" />
99
<link rel="stylesheet" type="text/css" href="_static/basic.css?v=29da98fa" />
10-
<link rel="stylesheet" type="text/css" href="_static/styles/front_page.css?v=ccf8d1d0" />
10+
<link rel="stylesheet" type="text/css" href="_static/styles/front_page.css?v=e05e6e91" />
1111
<script src="_static/documentation_options.js?v=938c9ccc"></script>
1212
<script src="_static/doctools.js?v=9bcbadda"></script>
1313
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>

‎docs/_build/html/index.html‎

Lines changed: 27 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -86,15 +86,33 @@ <h2 id="quick-start">⚡ Quick Start</h2>
8686
</p>
8787
</div>
8888

89-
<h3>Installation</h3>
90-
<pre><code>pip3 install llmsql</code></pre>
91-
92-
<h3>Recommended Workflow (vLLM)</h3>
93-
<pre><code>pip install llmsql[vllm]
94-
llmsql evaluate --model gpt-4 --dataset llmsql_dev</code></pre>
95-
96-
<h3>Evaluation API (Python)</h3>
97-
<pre><code>from llmsql import LLMSQLEvaluator
89+
<h3>1️⃣ Installation</h3>
90+
<p>Install the base package:</p>
91+
<pre><code>pip install llmsql</code></pre>
92+
93+
<p>To enable the vLLM backend:</p>
94+
<pre><code>pip install llmsql[vllm]</code></pre>
95+
96+
<h3>2️⃣ Inference from CLI</h3>
97+
98+
<p><strong>vLLM Backend (Recommended)</strong></p>
99+
<pre><code>llmsql inference --method vllm \
100+
--model-name Qwen/Qwen2.5-1.5B-Instruct \
101+
--output-file outputs/preds.jsonl \
102+
--batch-size 8 \
103+
--num_fewshots 5 \
104+
--temperature 0.0</code></pre>
105+
106+
<p><strong>Transformers Backend</strong></p>
107+
<pre><code>llmsql inference --method transformers \
108+
--model-or-model-name-or-path Qwen/Qwen2.5-1.5B-Instruct \
109+
--output-file outputs/preds.jsonl \
110+
--batch-size 8 \
111+
--temperature 0.9 \
112+
--generate-kwargs '{"do_sample": false, "top_p": 0.95}'</code></pre>
113+
114+
<h3>3️⃣ Evaluation API (Python)</h3>
115+
<pre><code>from llmsql import LLMSQLEvaluator
98116

99117
evaluator = LLMSQLEvaluator(workdir_path="llmsql_workdir")
100118
report = evaluator.evaluate(outputs_path="path_to_your_outputs.jsonl")

‎docs/_build/html/search.html‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
<title>Search &#8212; LLMSQL 0.2.0 documentation</title>
88
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=8e8a900e" />
99
<link rel="stylesheet" type="text/css" href="_static/basic.css?v=29da98fa" />
10-
<link rel="stylesheet" type="text/css" href="_static/styles/front_page.css?v=ccf8d1d0" />
10+
<link rel="stylesheet" type="text/css" href="_static/styles/front_page.css?v=e05e6e91" />
1111

1212
<script src="_static/documentation_options.js?v=938c9ccc"></script>
1313
<script src="_static/doctools.js?v=9bcbadda"></script>

‎docs/_templates/index.html‎

Lines changed: 27 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -86,15 +86,33 @@ <h2 id="quick-start">⚡ Quick Start</h2>
8686
</p>
8787
</div>
8888

89-
<h3>Installation</h3>
90-
<pre><code>pip3 install llmsql</code></pre>
91-
92-
<h3>Recommended Workflow (vLLM)</h3>
93-
<pre><code>pip install llmsql[vllm]
94-
llmsql evaluate --model gpt-4 --dataset llmsql_dev</code></pre>
95-
96-
<h3>Evaluation API (Python)</h3>
97-
<pre><code>from llmsql import LLMSQLEvaluator
89+
<h3>1️⃣ Installation</h3>
90+
<p>Install the base package:</p>
91+
<pre><code>pip install llmsql</code></pre>
92+
93+
<p>To enable the vLLM backend:</p>
94+
<pre><code>pip install llmsql[vllm]</code></pre>
95+
96+
<h3>2️⃣ Inference from CLI</h3>
97+
98+
<p><strong>vLLM Backend (Recommended)</strong></p>
99+
<pre><code>llmsql inference --method vllm \
100+
--model-name Qwen/Qwen2.5-1.5B-Instruct \
101+
--output-file outputs/preds.jsonl \
102+
--batch-size 8 \
103+
--num_fewshots 5 \
104+
--temperature 0.0</code></pre>
105+
106+
<p><strong>Transformers Backend</strong></p>
107+
<pre><code>llmsql inference --method transformers \
108+
--model-or-model-name-or-path Qwen/Qwen2.5-1.5B-Instruct \
109+
--output-file outputs/preds.jsonl \
110+
--batch-size 8 \
111+
--temperature 0.9 \
112+
--generate-kwargs '{"do_sample": false, "top_p": 0.95}'</code></pre>
113+
114+
<h3>3️⃣ Evaluation API (Python)</h3>
115+
<pre><code>from llmsql import LLMSQLEvaluator
98116

99117
evaluator = LLMSQLEvaluator(workdir_path="llmsql_workdir")
100118
report = evaluator.evaluate(outputs_path="path_to_your_outputs.jsonl")

0 commit comments

Comments
 (0)