Buckets:

hf-doc-build/doc-dev / lighteval /pr_722 /en /using-the-python-api.html
rtrm's picture
download
raw
8.73 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Using the Python API&quot;,&quot;local&quot;:&quot;using-the-python-api&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}">
<link href="/docs/lighteval/pr_722/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/entry/start.5b03f48b.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/scheduler.7da89386.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/singletons.ab0c7ea1.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/paths.3c8b78a7.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/entry/app.90ee4dda.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/index.20910acc.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/nodes/0.d90b7330.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/each.e59479a4.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/nodes/23.364a5f09.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/CodeBlock.143bd81e.js">
<link rel="modulepreload" href="/docs/lighteval/pr_722/en/_app/immutable/chunks/getInferenceSnippets.fc2ce523.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Using the Python API&quot;,&quot;local&quot;:&quot;using-the-python-api&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="using-the-python-api" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#using-the-python-api"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>Using the Python API</span></h1> <p data-svelte-h="svelte-1fsv9y8">Lighteval can be used from a custom python script. To evaluate a model you will need to set up an
<a href="/docs/lighteval/pr_722/en/package_reference/evaluation_tracker#lighteval.logging.evaluation_tracker.EvaluationTracker">EvaluationTracker</a>, <a href="/docs/lighteval/pr_722/en/package_reference/pipeline#lighteval.pipeline.PipelineParameters">PipelineParameters</a>,
a <a href="package_reference/models"><code>model</code></a> or a <a href="package_reference/model_config"><code>model_config</code></a>,
and a <a href="/docs/lighteval/pr_722/en/package_reference/pipeline#lighteval.pipeline.Pipeline">Pipeline</a>.</p> <p data-svelte-h="svelte-cbze7g">After that, simply run the pipeline and save the results.</p> <div class="code-block relative "><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-keyword">import</span> lighteval
<span class="hljs-keyword">from</span> lighteval.logging.evaluation_tracker <span class="hljs-keyword">import</span> EvaluationTracker
<span class="hljs-keyword">from</span> lighteval.models.vllm.vllm_model <span class="hljs-keyword">import</span> VLLMModelConfig
<span class="hljs-keyword">from</span> lighteval.pipeline <span class="hljs-keyword">import</span> ParallelismManager, Pipeline, PipelineParameters
<span class="hljs-keyword">from</span> lighteval.utils.utils <span class="hljs-keyword">import</span> EnvConfig
<span class="hljs-keyword">from</span> lighteval.utils.imports <span class="hljs-keyword">import</span> is_accelerate_available
<span class="hljs-keyword">if</span> is_accelerate_available():
<span class="hljs-keyword">from</span> datetime <span class="hljs-keyword">import</span> timedelta
<span class="hljs-keyword">from</span> accelerate <span class="hljs-keyword">import</span> Accelerator, InitProcessGroupKwargs
accelerator = Accelerator(kwargs_handlers=[InitProcessGroupKwargs(timeout=timedelta(seconds=<span class="hljs-number">3000</span>))])
<span class="hljs-keyword">else</span>:
accelerator = <span class="hljs-literal">None</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">main</span>():
evaluation_tracker = EvaluationTracker(
output_dir=<span class="hljs-string">&quot;./results&quot;</span>,
save_details=<span class="hljs-literal">True</span>,
push_to_hub=<span class="hljs-literal">True</span>,
hub_results_org=<span class="hljs-string">&quot;your user name&quot;</span>,
)
pipeline_params = PipelineParameters(
launcher_type=ParallelismManager.ACCELERATE,
env_config=EnvConfig(cache_dir=<span class="hljs-string">&quot;tmp/&quot;</span>),
custom_task_directory=<span class="hljs-literal">None</span>, <span class="hljs-comment"># if using a custom task</span>
<span class="hljs-comment"># Remove the 2 parameters below once your configuration is tested</span>
override_batch_size=<span class="hljs-number">1</span>,
max_samples=<span class="hljs-number">10</span>
)
model_config = VLLMModelConfig(
model_name=<span class="hljs-string">&quot;HuggingFaceH4/zephyr-7b-beta&quot;</span>,
dtype=<span class="hljs-string">&quot;float16&quot;</span>,
use_chat_template=<span class="hljs-literal">True</span>,
)
task = <span class="hljs-string">&quot;helm|mmlu|5|1&quot;</span>
pipeline = Pipeline(
tasks=task,
pipeline_parameters=pipeline_params,
evaluation_tracker=evaluation_tracker,
model_config=model_config,
)
pipeline.evaluate()
pipeline.save_and_push_results()
pipeline.show_results()
<span class="hljs-keyword">if</span> __name__ == <span class="hljs-string">&quot;__main__&quot;</span>:
main()<!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/lighteval/blob/main/docs/source/using-the-python-api.mdx" target="_blank"><span data-svelte-h="svelte-1kd6by1">&lt;</span> <span data-svelte-h="svelte-x0xyl0">&gt;</span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a> <p></p>
<script>
{
__sveltekit_1sktna2 = {
assets: "/docs/lighteval/pr_722/en",
base: "/docs/lighteval/pr_722/en",
env: {}
};
const element = document.currentScript.parentElement;
const data = [null,null];
Promise.all([
import("/docs/lighteval/pr_722/en/_app/immutable/entry/start.5b03f48b.js"),
import("/docs/lighteval/pr_722/en/_app/immutable/entry/app.90ee4dda.js")
]).then(([kit, app]) => {
kit.start(app, element, {
node_ids: [0, 23],
data,
form: null,
error: null
});
});
}
</script>

Xet Storage Details

Size:
8.73 kB
·
Xet hash:
7e99435f3c53c0a82216e479d8fdd0e1ce03af4fd4b9b3ce1028509da29881ae

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.