mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
779 lines
104 KiB
HTML
779 lines
104 KiB
HTML
<!DOCTYPE html>
|
|
<html class="writer-html5" lang="en" data-content_root="../../../">
|
|
<head>
|
|
<meta charset="utf-8" />
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
|
<title>tensorrt_llm.models.modeling_utils — tensorrt_llm documentation</title>
|
|
<link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=80d5e7a1" />
|
|
<link rel="stylesheet" type="text/css" href="../../../_static/css/theme.css?v=19f00094" />
|
|
|
|
|
|
<!--[if lt IE 9]>
|
|
<script src="../../../_static/js/html5shiv.min.js"></script>
|
|
<![endif]-->
|
|
|
|
<script src="../../../_static/jquery.js?v=5d32c60e"></script>
|
|
<script src="../../../_static/_sphinx_javascript_frameworks_compat.js?v=2cd50e6c"></script>
|
|
<script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
|
|
<script src="../../../_static/doctools.js?v=888ff710"></script>
|
|
<script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
|
|
<script src="../../../_static/js/theme.js"></script>
|
|
<link rel="index" title="Index" href="../../../genindex.html" />
|
|
<link rel="search" title="Search" href="../../../search.html" />
|
|
</head>
|
|
|
|
<body class="wy-body-for-nav">
|
|
<div class="wy-grid-for-nav">
|
|
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
|
|
<div class="wy-side-scroll">
|
|
<div class="wy-side-nav-search" >
|
|
|
|
|
|
|
|
<a href="../../../index.html" class="icon icon-home">
|
|
tensorrt_llm
|
|
</a>
|
|
<div role="search">
|
|
<form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
|
|
<input type="text" name="q" placeholder="Search docs" aria-label="Search docs" />
|
|
<input type="hidden" name="check_keywords" value="yes" />
|
|
<input type="hidden" name="area" value="default" />
|
|
</form>
|
|
</div>
|
|
</div><div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="Navigation menu">
|
|
<p class="caption" role="heading"><span class="caption-text">Contents:</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../architecture.html">TensorRT-LLM Architecture</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../gpt_runtime.html">C++ GPT Runtime</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../batch_manager.html">The Batch Manager in TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../inference_request.html">Inference Request</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../gpt_attention.html">Multi-head, Multi-query and Group-query Attention</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../precision.html">Numerical Precision</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../build_from_source.html">Build from Source</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../performance.html">Performance of TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../2023-05-19-how-to-debug.html">How to debug</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../2023-05-17-how-to-add-a-new-model.html">How to add a new model</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../graph-rewriting.html">Graph Rewriting Module</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../memory.html">Memory Usage of TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../new_workflow.html">New Workflow</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../lora.html">Run gpt-2b + LoRA using GptManager / cpp runtime</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../perf_best_practices.html">Best Practices for Tuning the Performance of TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../performance_analysis.html">Performance Analysis of TensorRT-LLM</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Python API</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.layers.html">Layers</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.functional.html">Functionals</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.models.html">Models</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.plugin.html">Plugin</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.quantization.html">Quantization</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../python-api/tensorrt_llm.runtime.html">Runtime</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">C++ API</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../_cpp_gen/runtime.html">Runtime</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Blogs</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../blogs/H100vsA100.html">H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../blogs/H200launch.html">H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../blogs/Falcon180B-H200.html">Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../blogs/quantization-in-TRT-LLM.html">Speed up inference with SOTA quantization techniques in TRT-LLM</a></li>
|
|
</ul>
|
|
|
|
</div>
|
|
</div>
|
|
</nav>
|
|
|
|
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"><nav class="wy-nav-top" aria-label="Mobile navigation menu" >
|
|
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
|
|
<a href="../../../index.html">tensorrt_llm</a>
|
|
</nav>
|
|
|
|
<div class="wy-nav-content">
|
|
<div class="rst-content">
|
|
<div role="navigation" aria-label="Page navigation">
|
|
<ul class="wy-breadcrumbs">
|
|
<li><a href="../../../index.html" class="icon icon-home" aria-label="Home"></a></li>
|
|
<li class="breadcrumb-item"><a href="../../index.html">Module code</a></li>
|
|
<li class="breadcrumb-item active">tensorrt_llm.models.modeling_utils</li>
|
|
<li class="wy-breadcrumbs-aside">
|
|
</li>
|
|
</ul>
|
|
<hr/>
|
|
</div>
|
|
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
|
|
<div itemprop="articleBody">
|
|
|
|
<h1>Source code for tensorrt_llm.models.modeling_utils</h1><div class="highlight"><pre>
|
|
<span></span><span class="kn">import</span> <span class="nn">copy</span>
|
|
<span class="kn">import</span> <span class="nn">json</span>
|
|
<span class="kn">import</span> <span class="nn">os</span>
|
|
<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span>
|
|
|
|
<span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
|
|
<span class="kn">import</span> <span class="nn">safetensors</span>
|
|
<span class="kn">import</span> <span class="nn">torch</span>
|
|
|
|
<span class="kn">from</span> <span class="nn">.._common</span> <span class="kn">import</span> <span class="n">default_net</span>
|
|
<span class="kn">from</span> <span class="nn">.._utils</span> <span class="kn">import</span> <span class="p">(</span><span class="n">numpy_to_torch</span><span class="p">,</span> <span class="n">str_dtype_to_torch</span><span class="p">,</span> <span class="n">str_dtype_to_trt</span><span class="p">,</span>
|
|
<span class="n">trt_dtype_to_torch</span><span class="p">)</span>
|
|
<span class="kn">from</span> <span class="nn">..functional</span> <span class="kn">import</span> <span class="n">PositionEmbeddingType</span><span class="p">,</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">gather_last_token_logits</span>
|
|
<span class="kn">from</span> <span class="nn">..layers</span> <span class="kn">import</span> <span class="p">(</span><span class="n">AttentionParams</span><span class="p">,</span> <span class="n">FusedGatedMLP</span><span class="p">,</span> <span class="n">GatedMLP</span><span class="p">,</span>
|
|
<span class="n">KeyValueCacheParams</span><span class="p">,</span> <span class="n">LoraParams</span><span class="p">)</span>
|
|
<span class="kn">from</span> <span class="nn">..mapping</span> <span class="kn">import</span> <span class="n">Mapping</span>
|
|
<span class="kn">from</span> <span class="nn">..module</span> <span class="kn">import</span> <span class="n">Module</span><span class="p">,</span> <span class="n">ModuleList</span>
|
|
<span class="kn">from</span> <span class="nn">..quantization</span> <span class="kn">import</span> <span class="n">QuantMode</span>
|
|
<span class="kn">from</span> <span class="nn">..quantization.quantize</span> <span class="kn">import</span> <span class="n">quantize</span>
|
|
<span class="kn">from</span> <span class="nn">.generation_mixin</span> <span class="kn">import</span> <span class="n">GenerationMixin</span>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">PretrainedConfig</span><span class="p">:</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
|
<span class="n">architecture</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">logits_dtype</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">vocab_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">max_position_embeddings</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">hidden_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">num_hidden_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">num_attention_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">num_key_value_heads</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">hidden_act</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">intermediate_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">norm_epsilon</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
|
|
<span class="n">position_embedding_type</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">world_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">tp_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">pp_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
|
<span class="n">quant_mode</span><span class="p">:</span> <span class="n">QuantMode</span><span class="p">,</span>
|
|
<span class="n">quant_kwargs</span><span class="p">:</span> <span class="nb">dict</span><span class="p">,</span>
|
|
<span class="n">use_prompt_tuning</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">use_parallel_embedding</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">embedding_sharding_dim</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
|
|
<span class="n">share_embedding_table</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">max_lora_rank</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">64</span><span class="p">,</span>
|
|
<span class="n">head_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">architecture</span> <span class="o">=</span> <span class="n">architecture</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">dtype</span> <span class="o">=</span> <span class="n">dtype</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">logits_dtype</span> <span class="o">=</span> <span class="n">logits_dtype</span>
|
|
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span> <span class="o">=</span> <span class="n">vocab_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">max_position_embeddings</span> <span class="o">=</span> <span class="n">max_position_embeddings</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">hidden_size</span> <span class="o">=</span> <span class="n">hidden_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_hidden_layers</span> <span class="o">=</span> <span class="n">num_hidden_layers</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_attention_heads</span> <span class="o">=</span> <span class="n">num_attention_heads</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_key_value_heads</span> <span class="o">=</span> <span class="n">num_key_value_heads</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">head_size</span> <span class="o">=</span> <span class="n">hidden_size</span> <span class="o">//</span> <span class="n">num_attention_heads</span> <span class="k">if</span> <span class="n">head_size</span> <span class="ow">is</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">head_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">hidden_act</span> <span class="o">=</span> <span class="n">hidden_act</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">intermediate_size</span> <span class="o">=</span> <span class="n">intermediate_size</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">norm_epsilon</span> <span class="o">=</span> <span class="n">norm_epsilon</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">position_embedding_type</span> <span class="o">=</span> <span class="n">PositionEmbeddingType</span><span class="o">.</span><span class="n">from_string</span><span class="p">(</span>
|
|
<span class="n">position_embedding_type</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">use_prompt_tuning</span> <span class="o">=</span> <span class="n">use_prompt_tuning</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">use_parallel_embedding</span> <span class="o">=</span> <span class="n">use_parallel_embedding</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">embedding_sharding_dim</span> <span class="o">=</span> <span class="n">embedding_sharding_dim</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">share_embedding_table</span> <span class="o">=</span> <span class="n">share_embedding_table</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">mapping</span> <span class="o">=</span> <span class="n">Mapping</span><span class="p">(</span><span class="n">world_size</span><span class="o">=</span><span class="n">world_size</span><span class="p">,</span>
|
|
<span class="n">tp_size</span><span class="o">=</span><span class="n">tp_size</span><span class="p">,</span>
|
|
<span class="n">pp_size</span><span class="o">=</span><span class="n">pp_size</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_mode</span> <span class="o">=</span> <span class="n">quant_mode</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span> <span class="o">=</span> <span class="n">quant_kwargs</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">kv_dtype</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">max_lora_rank</span> <span class="o">=</span> <span class="n">max_lora_rank</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant_mode</span><span class="o">.</span><span class="n">has_int8_kv_cache</span><span class="p">():</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">kv_dtype</span> <span class="o">=</span> <span class="s1">'int8'</span>
|
|
<span class="k">elif</span> <span class="bp">self</span><span class="o">.</span><span class="n">quant_mode</span><span class="o">.</span><span class="n">has_fp8_kv_cache</span><span class="p">():</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">kv_dtype</span> <span class="o">=</span> <span class="s1">'fp8'</span>
|
|
|
|
<span class="k">for</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span> <span class="ow">in</span> <span class="n">kwargs</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
|
|
<span class="k">try</span><span class="p">:</span>
|
|
<span class="nb">setattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span>
|
|
<span class="k">except</span> <span class="ne">AttributeError</span> <span class="k">as</span> <span class="n">err</span><span class="p">:</span>
|
|
<span class="k">raise</span> <span class="n">err</span>
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig.set_if_not_exist">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig.set_if_not_exist">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">set_if_not_exist</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">):</span>
|
|
<span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">):</span>
|
|
<span class="nb">setattr</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">key</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig.from_dict">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig.from_dict">[docs]</a>
|
|
<span class="nd">@classmethod</span>
|
|
<span class="k">def</span> <span class="nf">from_dict</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">config</span><span class="p">):</span>
|
|
<span class="n">architecture</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'architecture'</span><span class="p">)</span>
|
|
<span class="n">dtype</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'dtype'</span><span class="p">)</span>
|
|
<span class="n">vocab_size</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'vocab_size'</span><span class="p">)</span>
|
|
<span class="n">hidden_size</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'hidden_size'</span><span class="p">)</span>
|
|
<span class="n">num_hidden_layers</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'num_hidden_layers'</span><span class="p">)</span>
|
|
<span class="n">num_attention_heads</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'num_attention_heads'</span><span class="p">)</span>
|
|
<span class="n">hidden_act</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'hidden_act'</span><span class="p">)</span>
|
|
<span class="n">norm_epsilon</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'norm_epsilon'</span><span class="p">,</span> <span class="mf">1e-5</span><span class="p">)</span>
|
|
<span class="n">position_embedding_type</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'position_embedding_type'</span><span class="p">,</span>
|
|
<span class="s1">'learned_absolute'</span><span class="p">)</span>
|
|
<span class="n">logits_dtype</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'logits_dtype'</span><span class="p">,</span> <span class="s1">'float32'</span><span class="p">)</span>
|
|
<span class="n">num_key_value_heads</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'num_key_value_heads'</span><span class="p">,</span>
|
|
<span class="n">num_attention_heads</span><span class="p">)</span>
|
|
<span class="n">intermediate_size</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'intermediate_size'</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
|
|
<span class="n">max_position_embeddings</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'max_position_embeddings'</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
|
|
<span class="n">use_prompt_tuning</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'use_prompt_tuning'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
<span class="n">use_parallel_embedding</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'use_parallel_embedding'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
<span class="n">embedding_sharding_dim</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'embedding_sharding_dim'</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
|
|
<span class="n">share_embedding_table</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'share_embedding_table'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
|
|
<span class="n">mapping</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'mapping'</span><span class="p">,</span> <span class="p">{</span>
|
|
<span class="s1">'world_size'</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
|
|
<span class="s1">'tp_size'</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
|
|
<span class="s1">'pp_size'</span><span class="p">:</span> <span class="mi">1</span>
|
|
<span class="p">})</span>
|
|
<span class="n">world_size</span> <span class="o">=</span> <span class="n">mapping</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'world_size'</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
|
|
<span class="n">tp_size</span> <span class="o">=</span> <span class="n">mapping</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'tp_size'</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
|
|
<span class="n">pp_size</span> <span class="o">=</span> <span class="n">mapping</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'pp_size'</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">share_embedding_table</span> <span class="ow">and</span> <span class="n">mapping</span><span class="o">.</span><span class="n">tp_size</span> <span class="o">></span> <span class="mi">1</span><span class="p">:</span>
|
|
<span class="k">if</span> <span class="p">(</span><span class="ow">not</span> <span class="n">use_parallel_embedding</span><span class="p">)</span> <span class="ow">or</span> <span class="p">(</span><span class="n">use_parallel_embedding</span> <span class="ow">and</span>
|
|
<span class="n">embedding_sharding_dim</span> <span class="o">==</span> <span class="mi">1</span><span class="p">):</span>
|
|
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
|
|
<span class="s2">"For multiple-processes cases, sharing the embedding table must set"</span> \
|
|
<span class="s2">"use_parallel_embedding=True and embedding_sharding_dim=0"</span>
|
|
<span class="p">)</span>
|
|
|
|
<span class="n">quantization</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span>
|
|
<span class="s1">'quantization'</span><span class="p">,</span> <span class="p">{</span>
|
|
<span class="s1">'quant_algo'</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="s1">'kv_cache_quant_algo'</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="s1">'group_size'</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span>
|
|
<span class="s1">'has_zero_point'</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="s1">'pre_quant_scale'</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="s1">'exclude_modules'</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="s1">'sq_use_plugin'</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="p">})</span>
|
|
<span class="n">quant_algo</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'quant_algo'</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
|
|
<span class="n">kv_cache_quant_algo</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'kv_cache_quant_algo'</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
|
|
<span class="n">group_size</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'group_size'</span><span class="p">,</span> <span class="mi">128</span><span class="p">)</span>
|
|
<span class="n">has_zero_point</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'has_zero_point'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
<span class="n">pre_quant_scale</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'pre_quant_scale'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
<span class="n">exclude_modules</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'exclude_modules'</span><span class="p">,</span> <span class="kc">None</span><span class="p">)</span>
|
|
<span class="n">sq_use_plugin</span> <span class="o">=</span> <span class="n">quantization</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'sq_use_plugin'</span><span class="p">,</span> <span class="kc">False</span><span class="p">)</span>
|
|
|
|
<span class="n">quant_mode</span> <span class="o">=</span> <span class="n">QuantMode</span><span class="o">.</span><span class="n">from_quant_algo</span><span class="p">(</span><span class="n">quant_algo</span><span class="p">,</span> <span class="n">kv_cache_quant_algo</span><span class="p">)</span>
|
|
<span class="n">quant_kwargs</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="s1">'quant_algo'</span><span class="p">:</span> <span class="n">quant_algo</span><span class="p">,</span>
|
|
<span class="s1">'kv_cache_quant_algo'</span><span class="p">:</span> <span class="n">kv_cache_quant_algo</span><span class="p">,</span>
|
|
<span class="s1">'group_size'</span><span class="p">:</span> <span class="n">group_size</span><span class="p">,</span>
|
|
<span class="s1">'zero'</span><span class="p">:</span> <span class="n">has_zero_point</span><span class="p">,</span>
|
|
<span class="s1">'pre_quant_scale'</span><span class="p">:</span> <span class="n">pre_quant_scale</span><span class="p">,</span>
|
|
<span class="s1">'exclude_modules'</span><span class="p">:</span> <span class="n">exclude_modules</span><span class="p">,</span>
|
|
<span class="s1">'sq_use_plugin'</span><span class="p">:</span> <span class="n">sq_use_plugin</span><span class="p">,</span>
|
|
<span class="p">}</span>
|
|
|
|
<span class="n">max_lora_rank</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'max_lora_rank'</span><span class="p">,</span> <span class="mi">64</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="bp">cls</span><span class="p">(</span><span class="n">architecture</span><span class="p">,</span> <span class="n">dtype</span><span class="p">,</span> <span class="n">logits_dtype</span><span class="p">,</span> <span class="n">vocab_size</span><span class="p">,</span>
|
|
<span class="n">max_position_embeddings</span><span class="p">,</span> <span class="n">hidden_size</span><span class="p">,</span> <span class="n">num_hidden_layers</span><span class="p">,</span>
|
|
<span class="n">num_attention_heads</span><span class="p">,</span> <span class="n">num_key_value_heads</span><span class="p">,</span> <span class="n">hidden_act</span><span class="p">,</span>
|
|
<span class="n">intermediate_size</span><span class="p">,</span> <span class="n">norm_epsilon</span><span class="p">,</span> <span class="n">position_embedding_type</span><span class="p">,</span>
|
|
<span class="n">world_size</span><span class="p">,</span> <span class="n">tp_size</span><span class="p">,</span> <span class="n">pp_size</span><span class="p">,</span> <span class="n">quant_mode</span><span class="p">,</span> <span class="n">quant_kwargs</span><span class="p">,</span>
|
|
<span class="n">use_prompt_tuning</span><span class="p">,</span> <span class="n">use_parallel_embedding</span><span class="p">,</span>
|
|
<span class="n">embedding_sharding_dim</span><span class="p">,</span> <span class="n">share_embedding_table</span><span class="p">,</span> <span class="n">max_lora_rank</span><span class="p">,</span>
|
|
<span class="o">**</span><span class="n">config</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig.from_json_file">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig.from_json_file">[docs]</a>
|
|
<span class="nd">@classmethod</span>
|
|
<span class="k">def</span> <span class="nf">from_json_file</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">config_file</span><span class="p">:</span> <span class="nb">str</span><span class="p">):</span>
|
|
<span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">config_file</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
|
|
<span class="n">config</span> <span class="o">=</span> <span class="n">json</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">PretrainedConfig</span><span class="o">.</span><span class="n">from_dict</span><span class="p">(</span><span class="n">config</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig.to_dict">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig.to_dict">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">to_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="n">output</span> <span class="o">=</span> <span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="vm">__dict__</span><span class="p">)</span>
|
|
|
|
<span class="n">output</span><span class="p">[</span><span class="s1">'position_embedding_type'</span><span class="p">]</span> <span class="o">=</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">position_embedding_type</span><span class="p">)</span>
|
|
<span class="n">output</span><span class="p">[</span><span class="s1">'mapping'</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="s1">'world_size'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
|
|
<span class="s1">'tp_size'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">tp_size</span><span class="p">,</span>
|
|
<span class="s1">'pp_size'</span><span class="p">:</span> <span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">pp_size</span><span class="p">,</span>
|
|
<span class="p">}</span>
|
|
<span class="n">output</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'quant_mode'</span><span class="p">)</span>
|
|
<span class="n">output</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="s1">'quant_kwargs'</span><span class="p">)</span>
|
|
<span class="n">output</span><span class="p">[</span><span class="s1">'quantization'</span><span class="p">]</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="s1">'quant_algo'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'quant_algo'</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span>
|
|
<span class="s1">'kv_cache_quant_algo'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'kv_cache_quant_algo'</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span>
|
|
<span class="s1">'group_size'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'group_size'</span><span class="p">,</span> <span class="mi">128</span><span class="p">),</span>
|
|
<span class="s1">'has_zero_point'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'zero'</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
|
|
<span class="s1">'pre_quant_scale'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'pre_quant_scale'</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
|
|
<span class="s1">'exclude_modules'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'exclude_modules'</span><span class="p">,</span> <span class="kc">None</span><span class="p">),</span>
|
|
<span class="s1">'sq_use_plugin'</span><span class="p">:</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">'sq_use_plugin'</span><span class="p">,</span> <span class="kc">False</span><span class="p">),</span>
|
|
<span class="p">}</span>
|
|
|
|
<span class="k">return</span> <span class="n">output</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedConfig.set_rank">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedConfig.set_rank">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">set_rank</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">rank</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">mapping</span> <span class="o">=</span> <span class="n">Mapping</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">world_size</span><span class="p">,</span>
|
|
<span class="n">rank</span><span class="o">=</span><span class="n">rank</span><span class="p">,</span>
|
|
<span class="n">tp_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">tp_size</span><span class="p">,</span>
|
|
<span class="n">pp_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">pp_size</span><span class="p">)</span></div>
|
|
</div>
|
|
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">DecoderLayerList</span><span class="p">(</span><span class="n">ModuleList</span><span class="p">):</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="bp">cls</span><span class="p">,</span> <span class="n">config</span><span class="p">):</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">layer_list</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">pp_layers</span><span class="p">(</span><span class="n">config</span><span class="o">.</span><span class="n">num_hidden_layers</span><span class="p">)</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">([</span><span class="bp">cls</span><span class="p">(</span><span class="n">config</span><span class="p">,</span> <span class="n">idx</span><span class="p">)</span> <span class="k">for</span> <span class="n">idx</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer_list</span><span class="p">])</span>
|
|
|
|
<span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
|
<span class="n">hidden_states</span><span class="p">,</span>
|
|
<span class="n">use_cache</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">attention_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">attention_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">lora_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">medusa_position_offsets</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">medusa_packed_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
|
|
<span class="n">kv_cache_params</span><span class="o">.</span><span class="n">fill_none_tensor_list</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layer_list</span><span class="p">))</span>
|
|
|
|
<span class="k">if</span> <span class="n">use_cache</span><span class="p">:</span>
|
|
<span class="n">presents</span> <span class="o">=</span> <span class="p">[]</span>
|
|
|
|
<span class="k">for</span> <span class="n">layer_idx</span><span class="p">,</span> <span class="p">(</span>
|
|
<span class="n">layer</span><span class="p">,</span> <span class="n">past</span><span class="p">,</span> <span class="n">pointer</span><span class="p">,</span> <span class="n">host_pointer</span><span class="p">,</span>
|
|
<span class="n">max_attention_window_size</span><span class="p">)</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span>
|
|
<span class="nb">zip</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">kv_cache_params</span><span class="o">.</span><span class="n">past_key_value</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">.</span><span class="n">kv_cache_block_pointers</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">.</span><span class="n">host_kv_cache_block_pointers</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">.</span><span class="n">host_max_attention_window_sizes</span><span class="p">)):</span>
|
|
|
|
<span class="n">lora_layer_params</span> <span class="o">=</span> <span class="kc">None</span>
|
|
<span class="k">if</span> <span class="n">lora_params</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">lora_params</span><span class="o">.</span><span class="n">lora_ranks</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">lora_layer_params</span> <span class="o">=</span> <span class="n">lora_params</span><span class="o">.</span><span class="n">get_layer_params</span><span class="p">(</span><span class="n">layer_idx</span><span class="p">)</span>
|
|
|
|
<span class="n">kwargs</span> <span class="o">=</span> <span class="p">{}</span>
|
|
|
|
<span class="k">if</span> <span class="n">lora_layer_params</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'lora_layer_params'</span><span class="p">]</span> <span class="o">=</span> <span class="n">lora_layer_params</span>
|
|
<span class="k">if</span> <span class="n">medusa_position_offsets</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'medusa_position_offsets'</span><span class="p">]</span> <span class="o">=</span> <span class="n">medusa_position_offsets</span>
|
|
<span class="k">if</span> <span class="n">medusa_packed_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'medusa_packed_mask'</span><span class="p">]</span> <span class="o">=</span> <span class="n">medusa_packed_mask</span>
|
|
<span class="n">hidden_states</span> <span class="o">=</span> <span class="n">layer</span><span class="p">(</span>
|
|
<span class="n">hidden_states</span><span class="p">,</span>
|
|
<span class="n">use_cache</span><span class="o">=</span><span class="n">use_cache</span><span class="p">,</span>
|
|
<span class="n">attention_mask</span><span class="o">=</span><span class="n">attention_mask</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">=</span><span class="n">KeyValueCacheParams</span><span class="p">(</span>
|
|
<span class="n">past_key_value</span><span class="o">=</span><span class="p">[</span><span class="n">past</span><span class="p">],</span>
|
|
<span class="n">host_past_key_value_lengths</span><span class="o">=</span><span class="n">kv_cache_params</span><span class="o">.</span>
|
|
<span class="n">host_past_key_value_lengths</span><span class="p">,</span>
|
|
<span class="n">host_max_attention_window_sizes</span><span class="o">=</span><span class="n">max_attention_window_size</span><span class="p">,</span>
|
|
<span class="n">host_sink_token_length</span><span class="o">=</span><span class="n">kv_cache_params</span><span class="o">.</span>
|
|
<span class="n">host_sink_token_length</span><span class="p">,</span>
|
|
<span class="n">kv_cache_block_pointers</span><span class="o">=</span><span class="p">[</span><span class="n">pointer</span><span class="p">],</span>
|
|
<span class="n">host_kv_cache_block_pointers</span><span class="o">=</span><span class="p">[</span><span class="n">host_pointer</span><span class="p">],</span>
|
|
<span class="n">cache_indirection</span><span class="o">=</span><span class="n">kv_cache_params</span><span class="o">.</span><span class="n">cache_indirection</span><span class="p">),</span>
|
|
<span class="n">attention_params</span><span class="o">=</span><span class="n">attention_params</span><span class="p">,</span>
|
|
<span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">use_cache</span><span class="p">:</span>
|
|
<span class="n">presents</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
|
|
<span class="n">hidden_states</span> <span class="o">=</span> <span class="n">hidden_states</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
|
|
|
|
<span class="k">if</span> <span class="n">use_cache</span><span class="p">:</span>
|
|
<span class="k">return</span> <span class="n">hidden_states</span><span class="p">,</span> <span class="n">presents</span>
|
|
<span class="k">return</span> <span class="n">hidden_states</span>
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">PostInitCaller</span><span class="p">(</span><span class="nb">type</span><span class="p">):</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
|
|
<span class="n">obj</span> <span class="o">=</span> <span class="nb">type</span><span class="o">.</span><span class="fm">__call__</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="o">*</span><span class="n">args</span><span class="p">,</span> <span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
|
|
<span class="n">obj</span><span class="o">.</span><span class="n">__post_init__</span><span class="p">()</span>
|
|
<span class="k">return</span> <span class="n">obj</span>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedModel">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">PretrainedModel</span><span class="p">(</span><span class="n">Module</span><span class="p">,</span> <span class="n">GenerationMixin</span><span class="p">,</span> <span class="n">metaclass</span><span class="o">=</span><span class="n">PostInitCaller</span><span class="p">):</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">config</span><span class="p">:</span> <span class="n">PretrainedConfig</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">config</span> <span class="o">=</span> <span class="n">config</span>
|
|
|
|
<span class="k">def</span> <span class="nf">__post_init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
|
<span class="n">quantize</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">quant_mode</span><span class="p">,</span> <span class="o">**</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="p">)</span>
|
|
|
|
<div class="viewcode-block" id="PretrainedModel.check_config">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel.check_config">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">check_config</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">config</span><span class="p">):</span>
|
|
<span class="k">raise</span> <span class="ne">NotImplementedError</span><span class="p">(</span>
|
|
<span class="sa">f</span><span class="s2">"</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="vm">__class__</span><span class="si">}</span><span class="s2"> is an abstract class. Only classes inheriting this class can be called."</span>
|
|
<span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedModel.from_config">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel.from_config">[docs]</a>
|
|
<span class="nd">@classmethod</span>
|
|
<span class="k">def</span> <span class="nf">from_config</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">config</span><span class="p">:</span> <span class="n">PretrainedConfig</span><span class="p">):</span>
|
|
<span class="k">return</span> <span class="bp">cls</span><span class="p">(</span><span class="n">config</span><span class="p">)</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedModel.from_checkpoint">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel.from_checkpoint">[docs]</a>
|
|
<span class="nd">@classmethod</span>
|
|
<span class="k">def</span> <span class="nf">from_checkpoint</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span>
|
|
<span class="n">ckpt_dir</span><span class="p">:</span> <span class="nb">str</span><span class="p">,</span>
|
|
<span class="n">rank</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
|
|
<span class="n">config</span><span class="p">:</span> <span class="n">PretrainedConfig</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
|
|
<span class="k">if</span> <span class="n">config</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">config</span> <span class="o">=</span> <span class="n">PretrainedConfig</span><span class="o">.</span><span class="n">from_json_file</span><span class="p">(</span>
|
|
<span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">ckpt_dir</span><span class="p">,</span> <span class="s1">'config.json'</span><span class="p">))</span>
|
|
<span class="n">config</span><span class="o">.</span><span class="n">set_rank</span><span class="p">(</span><span class="n">rank</span><span class="p">)</span>
|
|
<span class="n">model</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">from_config</span><span class="p">(</span><span class="n">config</span><span class="p">)</span>
|
|
|
|
<span class="n">weights</span> <span class="o">=</span> <span class="p">{}</span>
|
|
<span class="k">with</span> <span class="n">safetensors</span><span class="o">.</span><span class="n">safe_open</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">ckpt_dir</span><span class="p">,</span>
|
|
<span class="sa">f</span><span class="s1">'rank</span><span class="si">{</span><span class="n">rank</span><span class="si">}</span><span class="s1">.safetensors'</span><span class="p">),</span>
|
|
<span class="n">framework</span><span class="o">=</span><span class="s1">'pt'</span><span class="p">,</span>
|
|
<span class="n">device</span><span class="o">=</span><span class="s1">'cpu'</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
|
|
<span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">f</span><span class="o">.</span><span class="n">keys</span><span class="p">():</span>
|
|
<span class="n">weights</span><span class="p">[</span><span class="n">key</span><span class="p">]</span> <span class="o">=</span> <span class="n">f</span><span class="o">.</span><span class="n">get_tensor</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>
|
|
<span class="n">model</span><span class="o">.</span><span class="n">load</span><span class="p">(</span><span class="n">weights</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="n">model</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedModel.load">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel.load">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">load</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">weights</span><span class="p">):</span>
|
|
<span class="n">expected_names</span> <span class="o">=</span> <span class="nb">set</span><span class="p">([</span><span class="n">name</span> <span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">param</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">()])</span>
|
|
<span class="n">provided_names</span> <span class="o">=</span> <span class="nb">set</span><span class="p">(</span><span class="n">weights</span><span class="o">.</span><span class="n">keys</span><span class="p">())</span>
|
|
<span class="k">if</span> <span class="n">provided_names</span> <span class="o">!=</span> <span class="n">expected_names</span><span class="p">:</span>
|
|
<span class="n">err_msg</span> <span class="o">=</span> <span class="s2">"Provided tensor names are different from those expected by the engine."</span>
|
|
<span class="k">if</span> <span class="n">expected_names</span><span class="o">.</span><span class="n">difference</span><span class="p">(</span><span class="n">provided_names</span><span class="p">):</span>
|
|
<span class="n">err_msg</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">"</span><span class="se">\n</span><span class="s2">Expected but not provided tensors: </span><span class="si">{</span><span class="n">expected_names</span><span class="o">.</span><span class="n">difference</span><span class="p">(</span><span class="n">provided_names</span><span class="p">)</span><span class="si">}</span><span class="s2">"</span>
|
|
<span class="k">if</span> <span class="n">provided_names</span><span class="o">.</span><span class="n">difference</span><span class="p">(</span><span class="n">expected_names</span><span class="p">):</span>
|
|
<span class="n">err_msg</span> <span class="o">+=</span> <span class="sa">f</span><span class="s2">"</span><span class="se">\n</span><span class="s2">Provided but not expected tensors: </span><span class="si">{</span><span class="n">provided_names</span><span class="o">.</span><span class="n">difference</span><span class="p">(</span><span class="n">expected_names</span><span class="p">)</span><span class="si">}</span><span class="s2">"</span>
|
|
<span class="k">raise</span> <span class="ne">RuntimeError</span><span class="p">(</span><span class="n">err_msg</span><span class="p">)</span>
|
|
|
|
<span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">param</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">named_parameters</span><span class="p">():</span>
|
|
<span class="n">param</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">weights</span><span class="p">[</span><span class="n">name</span><span class="p">]</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="PretrainedModel.prepare_inputs">
|
|
<a class="viewcode-back" href="../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.PretrainedModel.prepare_inputs">[docs]</a>
|
|
<span class="k">def</span> <span class="nf">prepare_inputs</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
|
<span class="n">max_batch_size</span><span class="p">,</span>
|
|
<span class="n">max_input_len</span><span class="p">,</span>
|
|
<span class="n">max_seq_len</span><span class="p">,</span>
|
|
<span class="n">use_cache</span><span class="p">,</span>
|
|
<span class="n">max_beam_width</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span><span class="p">,</span>
|
|
<span class="n">max_num_tokens</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">prompt_embedding_table_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
|
|
<span class="n">position_encoding_2d</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">max_draft_len</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span><span class="p">,</span>
|
|
<span class="n">gather_context_logits</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">gather_generation_logits</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">lora_target_modules</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">'''@brief: Prepare inputs Tensors for the model, the given sizes are used to determine the</span>
|
|
<span class="sd"> ranges of the dimensions of when using TRT dynamic shapes.</span>
|
|
|
|
<span class="sd"> @return: a list contains values which can be fed into the self.forward()</span>
|
|
<span class="sd"> '''</span>
|
|
|
|
<span class="c1"># Prepare inputs</span>
|
|
<span class="n">remove_input_padding</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">remove_input_padding</span>
|
|
<span class="n">use_gpt_attention_plugin</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">(</span>
|
|
<span class="p">)</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">gpt_attention_plugin</span>
|
|
<span class="n">use_gemm_plugin</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">gemm_plugin</span>
|
|
<span class="n">paged_kv_cache</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">paged_kv_cache</span>
|
|
<span class="n">tokens_per_block</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">tokens_per_block</span>
|
|
<span class="n">use_custom_all_reduce</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">(</span>
|
|
<span class="p">)</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">use_custom_all_reduce</span>
|
|
<span class="n">use_lora_plugin</span> <span class="o">=</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">lora_plugin</span>
|
|
|
|
<span class="n">model_inputs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_basic_inputs</span><span class="p">(</span>
|
|
<span class="n">max_batch_size</span><span class="o">=</span><span class="n">max_batch_size</span><span class="p">,</span>
|
|
<span class="n">max_beam_width</span><span class="o">=</span><span class="n">max_beam_width</span><span class="p">,</span>
|
|
<span class="n">max_input_len</span><span class="o">=</span><span class="n">max_input_len</span><span class="p">,</span>
|
|
<span class="n">max_seq_len</span><span class="o">=</span><span class="n">max_seq_len</span><span class="p">,</span>
|
|
<span class="n">num_kv_heads</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">num_key_value_heads</span><span class="p">,</span>
|
|
<span class="n">head_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">head_size</span><span class="p">,</span>
|
|
<span class="n">num_layers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">num_hidden_layers</span><span class="p">,</span>
|
|
<span class="n">kv_dtype</span><span class="o">=</span><span class="n">str_dtype_to_trt</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">kv_dtype</span><span class="p">),</span>
|
|
<span class="n">remove_input_padding</span><span class="o">=</span><span class="n">remove_input_padding</span><span class="p">,</span>
|
|
<span class="n">use_gpt_attention_plugin</span><span class="o">=</span><span class="n">use_gpt_attention_plugin</span><span class="p">,</span>
|
|
<span class="n">use_gemm_plugin</span><span class="o">=</span><span class="n">use_gemm_plugin</span><span class="p">,</span>
|
|
<span class="n">paged_kv_cache</span><span class="o">=</span><span class="n">paged_kv_cache</span><span class="p">,</span>
|
|
<span class="n">tokens_per_block</span><span class="o">=</span><span class="n">tokens_per_block</span><span class="p">,</span>
|
|
<span class="n">num_heads</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">num_attention_heads</span><span class="p">,</span>
|
|
<span class="n">max_num_tokens</span><span class="o">=</span><span class="n">max_num_tokens</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">str_dtype_to_trt</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">dtype</span><span class="p">),</span>
|
|
<span class="n">prompt_embedding_table_size</span><span class="o">=</span><span class="n">prompt_embedding_table_size</span><span class="p">,</span>
|
|
<span class="n">position_encoding_2d</span><span class="o">=</span><span class="n">position_encoding_2d</span><span class="p">,</span>
|
|
<span class="n">mapping</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="p">,</span>
|
|
<span class="n">gather_context_logits</span><span class="o">=</span><span class="n">gather_context_logits</span><span class="p">,</span>
|
|
<span class="n">gather_generation_logits</span><span class="o">=</span><span class="n">gather_generation_logits</span><span class="p">,</span>
|
|
<span class="n">use_custom_all_reduce</span><span class="o">=</span><span class="n">use_custom_all_reduce</span><span class="p">,</span>
|
|
<span class="n">use_lora_plugin</span><span class="o">=</span><span class="n">use_lora_plugin</span><span class="p">,</span>
|
|
<span class="n">max_draft_len</span><span class="o">=</span><span class="n">max_draft_len</span><span class="p">,</span>
|
|
<span class="n">lora_target_modules</span><span class="o">=</span><span class="n">lora_target_modules</span><span class="p">)</span>
|
|
|
|
<span class="n">result</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="s1">'input_ids'</span><span class="p">:</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'input_ids'</span><span class="p">],</span>
|
|
<span class="s1">'position_ids'</span><span class="p">:</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'position_ids'</span><span class="p">],</span>
|
|
<span class="s1">'use_cache'</span><span class="p">:</span>
|
|
<span class="kc">True</span><span class="p">,</span>
|
|
<span class="s1">'last_token_ids'</span><span class="p">:</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'last_token_ids'</span><span class="p">],</span>
|
|
<span class="s1">'attention_mask'</span><span class="p">:</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'attention_mask'</span><span class="p">],</span>
|
|
<span class="s1">'kv_cache_params'</span><span class="p">:</span>
|
|
<span class="n">KeyValueCacheParams</span><span class="p">(</span>
|
|
<span class="n">past_key_value</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'past_key_value'</span><span class="p">],</span>
|
|
<span class="n">host_past_key_value_lengths</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span>
|
|
<span class="s1">'host_past_key_value_lengths'</span><span class="p">],</span>
|
|
<span class="n">host_max_attention_window_sizes</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span>
|
|
<span class="s1">'host_max_attention_window_sizes'</span><span class="p">],</span>
|
|
<span class="n">host_sink_token_length</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'host_sink_token_length'</span><span class="p">],</span>
|
|
<span class="n">kv_cache_block_pointers</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span>
|
|
<span class="s1">'kv_cache_block_pointers_list'</span><span class="p">],</span>
|
|
<span class="n">host_kv_cache_block_pointers</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span>
|
|
<span class="s1">'host_kv_cache_block_pointers_list'</span><span class="p">],</span>
|
|
<span class="n">cache_indirection</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'cache_indirection'</span><span class="p">],</span>
|
|
<span class="p">),</span>
|
|
<span class="s1">'attention_params'</span><span class="p">:</span>
|
|
<span class="n">AttentionParams</span><span class="p">(</span>
|
|
<span class="n">sequence_length</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'sequence_length'</span><span class="p">],</span>
|
|
<span class="n">context_lengths</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'context_lengths'</span><span class="p">],</span>
|
|
<span class="n">host_context_lengths</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'host_context_lengths'</span><span class="p">],</span>
|
|
<span class="n">max_context_length</span><span class="o">=</span><span class="n">max_input_len</span><span class="p">,</span>
|
|
<span class="n">host_request_types</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'host_request_types'</span><span class="p">])</span>
|
|
<span class="p">}</span>
|
|
|
|
<span class="k">if</span> <span class="n">prompt_embedding_table_size</span> <span class="o">></span> <span class="mi">0</span><span class="p">:</span>
|
|
<span class="n">result</span><span class="p">[</span><span class="s1">'prompt_embedding_table'</span><span class="p">]</span> <span class="o">=</span> <span class="n">model_inputs</span><span class="p">[</span>
|
|
<span class="s1">'prompt_embedding_table'</span><span class="p">]</span>
|
|
<span class="n">result</span><span class="p">[</span><span class="s1">'prompt_tasks'</span><span class="p">]</span> <span class="o">=</span> <span class="n">model_inputs</span><span class="p">[</span><span class="s1">'tasks'</span><span class="p">]</span>
|
|
<span class="n">result</span><span class="p">[</span><span class="s1">'prompt_vocab_size'</span><span class="p">]</span> <span class="o">=</span> <span class="n">model_inputs</span><span class="p">[</span><span class="s1">'prompt_vocab_size'</span><span class="p">]</span>
|
|
<span class="k">if</span> <span class="n">model_inputs</span><span class="p">[</span><span class="s1">'hidden_states_input'</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">result</span><span class="p">[</span><span class="s1">'hidden_states'</span><span class="p">]</span> <span class="o">=</span> <span class="n">model_inputs</span><span class="p">[</span><span class="s1">'hidden_states_input'</span><span class="p">]</span>
|
|
<span class="k">if</span> <span class="n">use_lora_plugin</span><span class="p">:</span>
|
|
<span class="n">result</span><span class="p">[</span><span class="s1">'lora_params'</span><span class="p">]</span> <span class="o">=</span> <span class="n">LoraParams</span><span class="p">(</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'lora_ranks'</span><span class="p">],</span>
|
|
<span class="n">model_inputs</span><span class="p">[</span><span class="s1">'lora_weights_pointers'</span><span class="p">],</span>
|
|
<span class="n">host_context_lengths</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'host_context_lengths'</span><span class="p">],</span>
|
|
<span class="n">max_context_length</span><span class="o">=</span><span class="n">max_input_len</span><span class="p">,</span>
|
|
<span class="n">host_request_types</span><span class="o">=</span><span class="n">model_inputs</span><span class="p">[</span><span class="s1">'host_request_types'</span><span class="p">])</span>
|
|
|
|
<span class="k">return</span> <span class="n">result</span></div>
|
|
</div>
|
|
|
|
|
|
|
|
<span class="k">class</span> <span class="nc">DecoderModelForCausalLM</span><span class="p">(</span><span class="n">PretrainedModel</span><span class="p">):</span>
|
|
|
|
<span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">config</span><span class="p">,</span> <span class="n">transformer</span><span class="p">,</span> <span class="n">lm_head</span><span class="p">):</span>
|
|
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">config</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">transformer</span> <span class="o">=</span> <span class="n">transformer</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">lm_head</span> <span class="o">=</span> <span class="n">lm_head</span>
|
|
|
|
<span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
|
<span class="n">input_ids</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span>
|
|
<span class="n">position_ids</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">use_cache</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
|
|
<span class="n">last_token_ids</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">attention_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">kv_cache_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">attention_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">hidden_states</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">prompt_embedding_table</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">prompt_tasks</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">prompt_vocab_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Tensor</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">lora_params</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">medusa_position_offsets</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
|
<span class="n">medusa_packed_mask</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
|
|
<span class="n">kwargs</span> <span class="o">=</span> <span class="p">{</span>
|
|
<span class="s1">'input_ids'</span><span class="p">:</span> <span class="n">input_ids</span><span class="p">,</span>
|
|
<span class="s1">'position_ids'</span><span class="p">:</span> <span class="n">position_ids</span><span class="p">,</span>
|
|
<span class="s1">'use_cache'</span><span class="p">:</span> <span class="n">use_cache</span><span class="p">,</span>
|
|
<span class="s1">'attention_mask'</span><span class="p">:</span> <span class="n">attention_mask</span><span class="p">,</span>
|
|
<span class="s1">'kv_cache_params'</span><span class="p">:</span> <span class="n">kv_cache_params</span><span class="p">,</span>
|
|
<span class="s1">'attention_params'</span><span class="p">:</span> <span class="n">attention_params</span><span class="p">,</span>
|
|
<span class="p">}</span>
|
|
<span class="k">if</span> <span class="n">lora_params</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'lora_params'</span><span class="p">]</span> <span class="o">=</span> <span class="n">lora_params</span>
|
|
<span class="k">if</span> <span class="n">hidden_states</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'hidden_states'</span><span class="p">]</span> <span class="o">=</span> <span class="n">hidden_states</span>
|
|
<span class="k">if</span> <span class="n">prompt_embedding_table</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'prompt_embedding_table'</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt_embedding_table</span>
|
|
<span class="k">if</span> <span class="n">prompt_tasks</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'prompt_tasks'</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt_tasks</span>
|
|
<span class="k">if</span> <span class="n">prompt_vocab_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'prompt_vocab_size'</span><span class="p">]</span> <span class="o">=</span> <span class="n">prompt_vocab_size</span>
|
|
|
|
<span class="k">if</span> <span class="n">medusa_position_offsets</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'medusa_position_offsets'</span><span class="p">]</span> <span class="o">=</span> <span class="n">medusa_position_offsets</span>
|
|
<span class="k">if</span> <span class="n">medusa_packed_mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'medusa_packed_mask'</span><span class="p">]</span> <span class="o">=</span> <span class="n">medusa_packed_mask</span>
|
|
|
|
<span class="n">hidden_states</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">transformer</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">use_cache</span><span class="p">:</span>
|
|
<span class="n">hidden_states</span><span class="p">,</span> <span class="n">presents</span> <span class="o">=</span> <span class="n">hidden_states</span>
|
|
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">is_last_pp_rank</span><span class="p">():</span>
|
|
<span class="n">hidden_states</span> <span class="o">=</span> <span class="n">gather_last_token_logits</span><span class="p">(</span>
|
|
<span class="n">hidden_states</span><span class="p">,</span> <span class="n">last_token_ids</span><span class="p">,</span>
|
|
<span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">remove_input_padding</span><span class="p">)</span>
|
|
|
|
<span class="c1"># [batch_size, hidden_size] -> [batch_size, vocab_size]</span>
|
|
<span class="n">lm_logits</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">lm_head</span><span class="p">(</span><span class="n">hidden_states</span><span class="p">)</span>
|
|
<span class="n">lm_logits</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'logits'</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">logits_dtype</span><span class="p">)</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">hidden_states</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'hidden_states_output'</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">use_cache</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">default_net</span><span class="p">()</span><span class="o">.</span><span class="n">plugin_config</span><span class="o">.</span><span class="n">paged_kv_cache</span><span class="p">:</span>
|
|
<span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">present</span> <span class="ow">in</span> <span class="nb">zip</span><span class="p">(</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">pp_layers</span><span class="p">(</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">num_hidden_layers</span><span class="p">),</span> <span class="n">presents</span><span class="p">):</span>
|
|
<span class="n">present</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="sa">f</span><span class="s1">'present_key_value_</span><span class="si">{</span><span class="n">i</span><span class="si">}</span><span class="s1">'</span><span class="p">,</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">kv_dtype</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">is_last_pp_rank</span><span class="p">():</span>
|
|
<span class="k">return</span> <span class="p">(</span><span class="n">lm_logits</span><span class="p">,</span> <span class="n">presents</span><span class="p">,</span> <span class="n">hidden_states</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="p">(</span><span class="n">hidden_states</span><span class="p">,</span> <span class="n">presents</span><span class="p">)</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">mapping</span><span class="o">.</span><span class="n">is_last_pp_rank</span><span class="p">():</span>
|
|
<span class="k">return</span> <span class="n">lm_logits</span><span class="p">,</span> <span class="n">hidden_states</span>
|
|
<span class="k">return</span> <span class="n">hidden_states</span>
|
|
|
|
|
|
<span class="k">def</span> <span class="nf">fuse_gate_mlp</span><span class="p">(</span><span class="n">model</span><span class="p">):</span>
|
|
<span class="k">for</span> <span class="n">layer</span> <span class="ow">in</span> <span class="n">model</span><span class="o">.</span><span class="n">transformer</span><span class="o">.</span><span class="n">layers</span><span class="p">:</span>
|
|
<span class="k">if</span> <span class="ow">not</span> <span class="nb">hasattr</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="s1">'mlp'</span><span class="p">):</span>
|
|
<span class="k">continue</span>
|
|
|
|
<span class="n">quant_algo</span> <span class="o">=</span> <span class="n">model</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">quant_kwargs</span><span class="p">[</span><span class="s1">'quant_algo'</span><span class="p">]</span>
|
|
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="p">,</span> <span class="n">GatedMLP</span><span class="p">):</span>
|
|
<span class="n">fused_layer</span> <span class="o">=</span> <span class="n">FusedGatedMLP</span><span class="p">(</span>
|
|
<span class="n">hidden_size</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">hidden_size</span><span class="p">,</span>
|
|
<span class="n">ffn_hidden_size</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">ffn_hidden_size</span><span class="p">,</span>
|
|
<span class="n">hidden_act</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">hidden_act</span><span class="p">,</span>
|
|
<span class="n">bias</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">bias</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
|
|
<span class="n">tp_group</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">tp_group</span><span class="p">,</span>
|
|
<span class="n">tp_size</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">tp_size</span><span class="p">,</span>
|
|
<span class="n">quant_mode</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">quant_mode</span><span class="p">,</span>
|
|
<span class="n">max_lora_rank</span><span class="o">=</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">max_lora_rank</span><span class="p">)</span>
|
|
|
|
<span class="k">if</span> <span class="n">quant_algo</span> <span class="o">==</span> <span class="s1">'FP8'</span><span class="p">:</span>
|
|
<span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
|
|
<span class="n">dtype</span> <span class="o">=</span> <span class="n">str_dtype_to_torch</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">dtype</span> <span class="o">=</span> <span class="n">trt_dtype_to_torch</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
|
|
|
|
<span class="c1"># dequantize</span>
|
|
<span class="n">gate_weight</span> <span class="o">=</span> <span class="n">numpy_to_torch</span><span class="p">(</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">raw_value</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span> <span class="o">*</span> <span class="n">numpy_to_torch</span><span class="p">(</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">weights_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span><span class="p">)</span>
|
|
<span class="n">fc_weight</span> <span class="o">=</span> <span class="n">numpy_to_torch</span><span class="p">(</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">raw_value</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">dtype</span><span class="p">)</span> <span class="o">*</span> <span class="n">numpy_to_torch</span><span class="p">(</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">weights_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span><span class="p">)</span>
|
|
|
|
<span class="c1"># concat</span>
|
|
<span class="n">fused_weight</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">([</span><span class="n">gate_weight</span><span class="p">,</span> <span class="n">fc_weight</span><span class="p">],</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
|
|
|
<span class="c1"># quantize</span>
|
|
<span class="n">fused_weight_scaling_factor</span> <span class="o">=</span> <span class="n">numpy_to_torch</span><span class="p">(</span>
|
|
<span class="nb">max</span><span class="p">(</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">weights_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span><span class="p">,</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">weights_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span><span class="p">,</span>
|
|
<span class="p">))</span>
|
|
<span class="n">fused_weight</span> <span class="o">=</span> <span class="p">(</span><span class="n">fused_weight</span> <span class="o">/</span> <span class="n">fused_weight_scaling_factor</span><span class="p">)</span><span class="o">.</span><span class="n">to</span><span class="p">(</span>
|
|
<span class="n">torch</span><span class="o">.</span><span class="n">float8_e4m3fn</span><span class="p">)</span>
|
|
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">fused_fc</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">fused_weight</span>
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">fused_fc</span><span class="o">.</span><span class="n">weights_scaling_factor</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">fused_weight_scaling_factor</span>
|
|
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">fused_fc</span><span class="o">.</span><span class="n">activation_scaling_factor</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> \
|
|
<span class="nb">max</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">activation_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span><span class="p">,</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">activation_scaling_factor</span><span class="o">.</span><span class="n">raw_value</span>
|
|
<span class="p">)</span>
|
|
<span class="k">elif</span> <span class="n">quant_algo</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">fused_fc</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">raw_value</span><span class="p">,</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">raw_value</span>
|
|
<span class="p">],</span>
|
|
<span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">bias</span><span class="p">:</span>
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">fused_fc</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">value</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">([</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">gate</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">raw_value</span><span class="p">,</span>
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">fc</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">raw_value</span>
|
|
<span class="p">],</span>
|
|
<span class="n">axis</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s1">'Unsupported quant algo: </span><span class="si">{</span><span class="n">quant_algo</span><span class="si">}</span><span class="s1">'</span><span class="p">)</span>
|
|
|
|
<span class="n">fused_layer</span><span class="o">.</span><span class="n">proj</span> <span class="o">=</span> <span class="n">layer</span><span class="o">.</span><span class="n">mlp</span><span class="o">.</span><span class="n">proj</span>
|
|
|
|
<span class="n">layer</span><span class="o">.</span><span class="n">mlp</span> <span class="o">=</span> <span class="n">fused_layer</span>
|
|
|
|
<span class="k">return</span> <span class="n">model</span>
|
|
|
|
|
|
<span class="k">def</span> <span class="nf">optimize_model</span><span class="p">(</span><span class="n">model</span><span class="p">,</span> <span class="n">use_fused_mlp</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
|
|
<span class="k">if</span> <span class="n">use_fused_mlp</span><span class="p">:</span>
|
|
<span class="n">model</span> <span class="o">=</span> <span class="n">fuse_gate_mlp</span><span class="p">(</span><span class="n">model</span><span class="p">)</span>
|
|
<span class="k">return</span> <span class="n">model</span>
|
|
</pre></div>
|
|
|
|
</div>
|
|
</div>
|
|
<footer>
|
|
|
|
<hr/>
|
|
|
|
<div role="contentinfo">
|
|
<p>© Copyright 2023, NVidia.</p>
|
|
</div>
|
|
|
|
Built with <a href="https://www.sphinx-doc.org/">Sphinx</a> using a
|
|
<a href="https://github.com/readthedocs/sphinx_rtd_theme">theme</a>
|
|
provided by <a href="https://readthedocs.org">Read the Docs</a>.
|
|
|
|
|
|
</footer>
|
|
</div>
|
|
</div>
|
|
</section>
|
|
</div>
|
|
<script>
|
|
jQuery(function () {
|
|
SphinxRtdTheme.Navigation.enable(true);
|
|
});
|
|
</script>
|
|
|
|
</body>
|
|
</html> |