mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
523 lines
57 KiB
HTML
523 lines
57 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.redrafter.model — 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=e59714d7" />
|
|
<link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
|
|
|
|
|
|
<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=9bcbadda"></script>
|
|
<script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
|
|
<script src="../../../../_static/clipboard.min.js?v=a7894cd8"></script>
|
|
<script src="../../../../_static/copybutton.js?v=65e89d2a"></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">Getting Started</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../overview.html">Overview</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../quick-start-guide.html">Quick Start Guide</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../key-features.html">Key Features</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../release-notes.html">Release Notes</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Installation</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../installation/linux.html">Installing on Linux</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../installation/build-from-source-linux.html">Building from Source Code on Linux</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../installation/windows.html">Installing on Windows</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../installation/build-from-source-windows.html">Building from Source Code on Windows</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../installation/grace-hopper.html">Installing on Grace Hopper</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">LLM API</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../llm-api/index.html">API Introduction</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../llm-api/reference.html">API Reference</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">LLM API Examples</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../llm-api-examples/index.html">LLM Examples Introduction</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../llm-api-examples/customization.html">Common Customizations</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../llm-api-examples/llm_api_examples.html">Examples</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Model Definition 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/executor.html">Executor</a></li>
|
|
<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">Command-Line Reference</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../commands/trtllm-build.html">trtllm-build</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../commands/trtllm-serve.html">trtllm-serve</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Architecture</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/overview.html">TensorRT-LLM Architecture</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/core-concepts.html">Model Definition</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/core-concepts.html#compilation">Compilation</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/core-concepts.html#runtime">Runtime</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/core-concepts.html#multi-gpu-and-multi-node-support">Multi-GPU and Multi-Node Support</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/checkpoint.html">TensorRT-LLM Checkpoint</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/workflow.html">TensorRT-LLM Build Workflow</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../architecture/add-model.html">Adding a Model</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Advanced</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/gpt-attention.html">Multi-Head, Multi-Query, and Group-Query Attention</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/gpt-runtime.html">C++ GPT Runtime</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/executor.html">Executor API</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/graph-rewriting.html">Graph Rewriting Module</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/inference-request.html">Inference Request</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/inference-request.html#responses">Responses</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/lora.html">Run gpt-2b + LoRA using GptManager / cpp runtime</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/expert-parallelism.html">Expert Parallelism in TensorRT-LLM</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/kv-cache-reuse.html">KV cache reuse</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/speculative-decoding.html">Speculative Sampling</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../advanced/disaggregated-service.html">Disaggregated-Service (experimental)</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Performance</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../performance/perf-overview.html">Overview</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../performance/perf-benchmarking.html">Benchmarking</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../performance/perf-best-practices.html">Best Practices</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../performance/perf-analysis.html">Performance Analysis</a></li>
|
|
</ul>
|
|
<p class="caption" role="heading"><span class="caption-text">Reference</span></p>
|
|
<ul>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../reference/troubleshooting.html">Troubleshooting</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../reference/support-matrix.html">Support Matrix</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../reference/precision.html">Numerical Precision</a></li>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../reference/memory.html">Memory Usage of TensorRT-LLM</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>
|
|
<li class="toctree-l1"><a class="reference internal" href="../../../../blogs/XQA-kernel.html">New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget</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.redrafter.model</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.redrafter.model</h1><div class="highlight"><pre>
|
|
<span></span><span class="c1"># SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.</span>
|
|
<span class="c1"># SPDX-License-Identifier: Apache-2.0</span>
|
|
<span class="c1">#</span>
|
|
<span class="c1"># Licensed under the Apache License, Version 2.0 (the "License");</span>
|
|
<span class="c1"># you may not use this file except in compliance with the License.</span>
|
|
<span class="c1"># You may obtain a copy of the License at</span>
|
|
<span class="c1">#</span>
|
|
<span class="c1"># http://www.apache.org/licenses/LICENSE-2.0</span>
|
|
<span class="c1">#</span>
|
|
<span class="c1"># Unless required by applicable law or agreed to in writing, software</span>
|
|
<span class="c1"># distributed under the License is distributed on an "AS IS" BASIS,</span>
|
|
<span class="c1"># WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.</span>
|
|
<span class="c1"># See the License for the specific language governing permissions and</span>
|
|
<span class="c1"># limitations under the License.</span>
|
|
|
|
<span class="kn">from</span> <span class="nn">collections</span> <span class="kn">import</span> <span class="n">OrderedDict</span>
|
|
|
|
<span class="kn">import</span> <span class="nn">tensorrt</span> <span class="k">as</span> <span class="nn">trt</span>
|
|
|
|
<span class="kn">from</span> <span class="nn">tensorrt_llm._common</span> <span class="kn">import</span> <span class="n">default_net</span>
|
|
<span class="kn">from</span> <span class="nn">tensorrt_llm.bindings</span> <span class="kn">import</span> <span class="n">KVCacheType</span>
|
|
<span class="kn">from</span> <span class="nn">tensorrt_llm.functional</span> <span class="kn">import</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">cast</span><span class="p">,</span> <span class="n">categorical_sample</span>
|
|
<span class="kn">from</span> <span class="nn">tensorrt_llm.models</span> <span class="kn">import</span> <span class="n">LLaMAForCausalLM</span>
|
|
<span class="kn">from</span> <span class="nn">tensorrt_llm.models.generation_mixin</span> <span class="kn">import</span> <span class="n">GenerationMixin</span>
|
|
|
|
<span class="kn">from</span> <span class="nn">..._utils</span> <span class="kn">import</span> <span class="n">pad_vocab_size</span><span class="p">,</span> <span class="n">str_dtype_to_trt</span>
|
|
<span class="kn">from</span> <span class="nn">.drafter</span> <span class="kn">import</span> <span class="n">Drafter</span>
|
|
<span class="kn">from</span> <span class="nn">.redrafter_helper</span> <span class="kn">import</span> <span class="p">(</span><span class="n">_beam_search_candidates</span><span class="p">,</span> <span class="n">_beams2tree</span><span class="p">,</span>
|
|
<span class="n">_process_logits_and_hidden_states</span><span class="p">)</span>
|
|
|
|
|
|
<div class="viewcode-block" id="ReDrafterForCausalLM">
|
|
<a class="viewcode-back" href="../../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.ReDrafterForCausalLM">[docs]</a>
|
|
<span class="k">class</span> <span class="nc">ReDrafterForCausalLM</span><span class="p">(</span><span class="n">LLaMAForCausalLM</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="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">dtype</span> <span class="o">=</span> <span class="n">str_dtype_to_trt</span><span class="p">(</span><span class="n">config</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">vocab_size</span>
|
|
<span class="n">vocab_size_padded</span> <span class="o">=</span> <span class="n">pad_vocab_size</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">,</span>
|
|
<span class="n">config</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="bp">self</span><span class="o">.</span><span class="n">drafter</span> <span class="o">=</span> <span class="n">Drafter</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">vocab_size_padded</span><span class="p">)</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">redrafter_num_beams</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">beam_candidate_length</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">redrafter_draft_len_per_beam</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_candidate_length</span> <span class="o">+</span> <span class="mi">1</span> <span class="c1"># including true token</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">greedy_search</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">redrafter_greedy_search</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">is_rnn</span> <span class="o">=</span> <span class="n">config</span><span class="o">.</span><span class="n">redrafter_is_rnn</span>
|
|
<span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span> <span class="o">==</span> <span class="bp">self</span><span class="o">.</span><span class="n">drafter</span><span class="o">.</span><span class="n">dtype</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="n">dtype</span><span class="si">}</span><span class="s2"> != </span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">drafter</span><span class="o">.</span><span class="n">dtype</span><span class="si">}</span><span class="s2">"</span>
|
|
|
|
<span class="k">def</span> <span class="nf">_fwd_helper</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">lm_logits</span><span class="p">,</span> <span class="n">embedding</span><span class="p">,</span> <span class="n">drafter</span><span class="p">,</span>
|
|
<span class="n">kwargs</span><span class="p">:</span> <span class="nb">dict</span><span class="p">):</span>
|
|
<span class="w"> </span><span class="sd">'''</span>
|
|
<span class="sd"> Must enable remove_input_padding:</span>
|
|
<span class="sd"> hidden_states [total_tokens, H]</span>
|
|
<span class="sd"> lm_logits [total_tokens, V]</span>
|
|
<span class="sd"> 1. process_logits: context vs gen</span>
|
|
<span class="sd"> a. Context: just return the last hidden states, and logits/probs</span>
|
|
<span class="sd"> b. Gen:</span>
|
|
<span class="sd"> i. verify: use lm_logits, draft_probs, draft_indices, draft_tokens</span>
|
|
<span class="sd"> ii. select hidden state and update probs</span>
|
|
<span class="sd"> 3. Sample token based on probs</span>
|
|
<span class="sd"> 4. Generate candidates using hidden_states, sampled token</span>
|
|
<span class="sd"> 5. Using beams, generate validation buffers, mark them as output</span>
|
|
<span class="sd"> 6. Mark all the outputs</span>
|
|
<span class="sd"> '''</span>
|
|
|
|
<span class="n">num_beams</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span>
|
|
<span class="n">beam_length</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span>
|
|
|
|
<span class="c1"># Get the inputs needed</span>
|
|
<span class="n">rand_data_sample</span> <span class="o">=</span> <span class="n">kwargs</span><span class="p">[</span><span class="s1">'rand_data_sample'</span><span class="p">]</span>
|
|
<span class="n">position_ids_base</span> <span class="o">=</span> <span class="n">kwargs</span><span class="p">[</span><span class="s1">'position_ids_base'</span><span class="p">]</span>
|
|
|
|
<span class="c1"># Step 1: Process logits and hidden states</span>
|
|
<span class="c1"># process the base model output (verify for gen-phase)</span>
|
|
<span class="n">probs</span><span class="p">,</span> <span class="n">draft_input</span><span class="p">,</span> <span class="n">num_accepted_tokens</span><span class="p">,</span> \
|
|
<span class="n">accepted_beam_index</span> <span class="o">=</span> <span class="n">_process_logits_and_hidden_states</span><span class="p">(</span>
|
|
<span class="bp">self</span><span class="p">,</span> <span class="n">lm_logits</span><span class="p">,</span> <span class="n">hidden_states</span><span class="p">,</span> <span class="n">kwargs</span><span class="p">)</span>
|
|
<span class="c1"># NOTE: num_accepted_tokens doesn't include true token so add 1 here</span>
|
|
<span class="n">num_accepted_tokens</span> <span class="o">=</span> <span class="n">num_accepted_tokens</span> <span class="o">+</span> <span class="mi">1</span>
|
|
|
|
<span class="c1"># At this point:</span>
|
|
<span class="c1"># probs : [bs, V]</span>
|
|
<span class="c1"># hidden_states : [bs, H]</span>
|
|
|
|
<span class="c1"># Step 2: Sample token</span>
|
|
<span class="n">next_token</span> <span class="o">=</span> <span class="n">categorical_sample</span><span class="p">(</span><span class="n">probs</span><span class="p">,</span> <span class="n">rand_data_sample</span><span class="p">)</span>
|
|
|
|
<span class="c1"># Step 3: beam search</span>
|
|
<span class="n">new_draft_tokens</span><span class="p">,</span> <span class="n">new_draft_logits</span> <span class="o">=</span> <span class="n">_beam_search_candidates</span><span class="p">(</span>
|
|
<span class="n">draft_input</span><span class="p">,</span> <span class="n">next_token</span><span class="p">,</span> <span class="n">embedding</span><span class="p">,</span> <span class="n">drafter</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">is_rnn</span><span class="p">)</span>
|
|
|
|
<span class="c1"># Step 4: tree processing</span>
|
|
<span class="n">active_tokens_flattened</span><span class="p">,</span> <span class="n">new_draft_token_indices</span><span class="p">,</span> <span class="n">new_mask</span><span class="p">,</span> \
|
|
<span class="n">new_position_offsets</span><span class="p">,</span> <span class="n">packed_position_ids</span><span class="p">,</span> <span class="n">next_num_gen_tokens</span><span class="p">,</span> <span class="n">max_gen_token</span><span class="p">,</span> \
|
|
<span class="n">total_gen_token</span> <span class="o">=</span> <span class="n">_beams2tree</span><span class="p">(</span><span class="n">new_draft_tokens</span><span class="p">,</span> <span class="n">num_beams</span><span class="p">,</span> <span class="n">beam_length</span><span class="p">,</span>
|
|
<span class="n">position_ids_base</span> <span class="o">+</span> <span class="n">num_accepted_tokens</span><span class="p">)</span>
|
|
|
|
<span class="c1"># Step 5: mark all the tensors we need</span>
|
|
<span class="n">num_accepted_tokens</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'num_accepted_tokens'</span><span class="p">)</span>
|
|
<span class="n">accepted_beam_index</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'accepted_beam_index'</span><span class="p">)</span>
|
|
<span class="n">max_gen_token</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'max_gen_token'</span><span class="p">)</span>
|
|
<span class="n">total_gen_token</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'total_gen_token'</span><span class="p">)</span>
|
|
<span class="n">next_num_gen_tokens</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_spec_decoding_generation_lengths'</span><span class="p">)</span>
|
|
<span class="n">active_tokens_flattened</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_flat_tokens'</span><span class="p">)</span>
|
|
<span class="n">new_draft_tokens</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_draft_tokens'</span><span class="p">)</span>
|
|
<span class="n">new_draft_logits</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_draft_probs'</span><span class="p">)</span>
|
|
<span class="n">new_draft_token_indices</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_draft_indices'</span><span class="p">)</span>
|
|
<span class="n">new_mask</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'spec_decoding_mask'</span><span class="p">)</span>
|
|
<span class="n">new_position_offsets</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'next_spec_decoding_position_offsets'</span><span class="p">)</span>
|
|
<span class="n">packed_position_ids</span><span class="o">.</span><span class="n">mark_output</span><span class="p">(</span><span class="s1">'packed_position_ids'</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="n">next_token</span><span class="p">,</span> <span class="n">probs</span><span class="p">,</span> <span class="n">draft_input</span>
|
|
|
|
<div class="viewcode-block" id="ReDrafterForCausalLM.forward">
|
|
<a class="viewcode-back" href="../../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.ReDrafterForCausalLM.forward">[docs]</a>
|
|
<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="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="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> 0. run base model, get logits, hidden_states</span>
|
|
<span class="sd"> """</span>
|
|
|
|
<span class="n">extra_args</span> <span class="o">=</span> <span class="p">[</span>
|
|
<span class="s1">'draft_tokens'</span><span class="p">,</span>
|
|
<span class="s1">'draft_indices'</span><span class="p">,</span>
|
|
<span class="s1">'draft_probs'</span><span class="p">,</span>
|
|
<span class="s1">'device_request_types'</span><span class="p">,</span>
|
|
<span class="s1">'redrafter_inverted_temperature'</span><span class="p">,</span>
|
|
<span class="s1">'rand_data_validation'</span><span class="p">,</span>
|
|
<span class="s1">'rand_data_sample'</span><span class="p">,</span>
|
|
<span class="s1">'position_ids_base'</span><span class="p">,</span>
|
|
<span class="p">]</span>
|
|
<span class="n">use_cache</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="n">base_kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="n">k</span><span class="p">:</span> <span class="n">v</span> <span class="k">for</span> <span class="n">k</span><span class="p">,</span> <span class="n">v</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">if</span> <span class="n">k</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">extra_args</span><span class="p">}</span>
|
|
<span class="k">if</span> <span class="n">use_cache</span> <span class="ow">and</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="ow">is</span> <span class="kc">False</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="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">forward</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">base_kwargs</span><span class="p">)</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">lm_logits</span><span class="p">,</span> <span class="n">hidden_states</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">forward</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">base_kwargs</span><span class="p">)</span>
|
|
|
|
<span class="c1"># lm_logits could be in fp32</span>
|
|
<span class="n">lm_logits_cast</span> <span class="o">=</span> <span class="n">cast</span><span class="p">(</span><span class="n">lm_logits</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">)</span> <span class="c1"># no-op if same type</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">register_network_output</span><span class="p">(</span><span class="s2">"hidden_states"</span><span class="p">,</span>
|
|
<span class="n">hidden_states</span><span class="p">)</span> <span class="c1"># debugging</span>
|
|
|
|
<span class="n">new_draft_tokens</span><span class="p">,</span> <span class="n">new_draft_logits</span><span class="p">,</span> <span class="n">probs</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_fwd_helper</span><span class="p">(</span>
|
|
<span class="n">hidden_states</span><span class="p">,</span>
|
|
<span class="n">lm_logits_cast</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">vocab_embedding</span><span class="p">,</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">drafter</span><span class="p">,</span>
|
|
<span class="n">kwargs</span><span class="o">=</span><span class="n">kwargs</span><span class="p">)</span>
|
|
|
|
<span class="k">return</span> <span class="n">new_draft_tokens</span><span class="p">,</span> <span class="n">new_draft_logits</span><span class="p">,</span> <span class="n">probs</span></div>
|
|
|
|
|
|
<div class="viewcode-block" id="ReDrafterForCausalLM.prepare_inputs">
|
|
<a class="viewcode-back" href="../../../../python-api/tensorrt_llm.models.html#tensorrt_llm.models.ReDrafterForCausalLM.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="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="w"> </span><span class="sd">"""</span>
|
|
<span class="sd"> Inputs needed:</span>
|
|
<span class="sd"> Assuming, max_gen_tokens = 1 + nb*(bl - 1), counting true token</span>
|
|
<span class="sd"> device_request_types: [bs]</span>
|
|
<span class="sd"> draft_tokens: [bs, nb, bl]</span>
|
|
<span class="sd"> draft_indices: [bs, nb, bl]</span>
|
|
<span class="sd"> draft_probs: [bs, nb, bl-1, V]</span>
|
|
<span class="sd"> spec_decoding_generation_lengths: [bs]</span>
|
|
<span class="sd"> spec_decoding_position_offsets: [bs, max_gen_tokens]</span>
|
|
<span class="sd"> spec_decoding_packed_mask: [bs, max_gen_tokens, packed_length] **</span>
|
|
<span class="sd"> redrafter_inverted_temperature: [bs]</span>
|
|
<span class="sd"> rand_data_sample: [bs]</span>
|
|
<span class="sd"> rand_data_validation: [bs, nb, bl-1]</span>
|
|
|
|
<span class="sd"> ** The mask is tricky since the boolean mask will need to be</span>
|
|
<span class="sd"> packed in runtime. So, the last dim will be:</span>
|
|
<span class="sd"> packed_length = ceil(max_gen_tokens/32)</span>
|
|
<span class="sd"> """</span>
|
|
<span class="n">default_range</span> <span class="o">=</span> <span class="n">GenerationMixin</span><span class="o">.</span><span class="n">default_range</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">max_batch_size</span> <span class="o">=</span> <span class="n">kwargs</span><span class="p">[</span><span class="s1">'max_batch_size'</span><span class="p">]</span>
|
|
<span class="k">assert</span> <span class="n">max_batch_size</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
|
|
<span class="n">bb_range</span> <span class="o">=</span> <span class="n">default_range</span><span class="p">(</span><span class="n">max_batch_size</span><span class="p">)</span>
|
|
<span class="n">bb0_range</span> <span class="o">=</span> <span class="n">default_range</span><span class="p">(</span><span class="n">max_batch_size</span><span class="p">,</span> <span class="n">min_range</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">opt_offset</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
|
|
<span class="n">num_beam_tokens</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span>
|
|
<span class="n">max_draft_tokens</span> <span class="o">=</span> <span class="n">num_beam_tokens</span> <span class="o">-</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span> <span class="c1"># ignore the true token</span>
|
|
<span class="n">max_gen_token_len</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">+</span> <span class="n">max_draft_tokens</span> <span class="c1"># for the true token</span>
|
|
<span class="n">max_gen_token_len_range</span> <span class="o">=</span> <span class="n">default_range</span><span class="p">(</span><span class="n">max_gen_token_len</span><span class="p">)</span>
|
|
<span class="n">bb_max_gen_token_len_range</span> <span class="o">=</span> <span class="n">default_range</span><span class="p">(</span><span class="n">max_gen_token_len</span> <span class="o">*</span>
|
|
<span class="n">max_batch_size</span><span class="p">,</span>
|
|
<span class="n">min_range</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
|
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'speculative_decoding_draft_tokens_external'</span><span class="p">]</span> <span class="o">=</span> <span class="kc">False</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'max_draft_len'</span><span class="p">]</span> <span class="o">=</span> <span class="n">max_draft_tokens</span>
|
|
<span class="n">kwargs</span><span class="p">[</span><span class="s1">'spec_decoding_is_generation_length_variable'</span><span class="p">]</span> <span class="o">=</span> <span class="kc">True</span>
|
|
<span class="n">inputs</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">prepare_inputs</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="k">assert</span> <span class="n">inputs</span><span class="p">[</span><span class="s1">'spec_decoding_params'</span><span class="p">]</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
|
|
|
|
<span class="n">enable_two_optimization_profiles</span> <span class="o">=</span> <span class="n">GenerationMixin</span><span class="o">.</span><span class="n">has_ctx_gen_opt_profiles</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">remove_input_padding</span><span class="o">=</span><span class="n">remove_input_padding</span><span class="p">,</span>
|
|
<span class="n">kv_cache_type</span><span class="o">=</span><span class="n">KVCacheType</span><span class="o">.</span><span class="n">PAGED</span>
|
|
<span class="k">if</span> <span class="n">paged_kv_cache</span> <span class="k">else</span> <span class="n">KVCacheType</span><span class="o">.</span><span class="n">CONTINUOUS</span><span class="p">)</span>
|
|
<span class="k">if</span> <span class="n">enable_two_optimization_profiles</span><span class="p">:</span>
|
|
<span class="n">bb_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">bb_range</span><span class="p">,</span> <span class="n">bb_range</span><span class="p">]</span>
|
|
<span class="n">bb0_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">bb0_range</span><span class="p">,</span> <span class="n">bb0_range</span><span class="p">]</span>
|
|
<span class="n">max_gen_token_len_range</span> <span class="o">=</span> <span class="p">[</span>
|
|
<span class="n">max_gen_token_len_range</span><span class="p">,</span> <span class="n">max_gen_token_len_range</span>
|
|
<span class="p">]</span>
|
|
<span class="n">bb_max_gen_token_len_range</span> <span class="o">=</span> <span class="p">[</span>
|
|
<span class="n">bb_max_gen_token_len_range</span><span class="p">,</span> <span class="n">bb_max_gen_token_len_range</span>
|
|
<span class="p">]</span>
|
|
<span class="n">num_beams_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">]</span>
|
|
<span class="n">beam_length_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">]</span>
|
|
<span class="n">candidate_length_range</span> <span class="o">=</span> <span class="p">[</span>
|
|
<span class="bp">self</span><span class="o">.</span><span class="n">beam_candidate_length</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_candidate_length</span>
|
|
<span class="p">]</span>
|
|
<span class="n">vocab_size_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">]</span>
|
|
<span class="k">else</span><span class="p">:</span>
|
|
<span class="n">bb_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">bb_range</span><span class="p">]</span>
|
|
<span class="n">bb0_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">bb0_range</span><span class="p">]</span>
|
|
<span class="n">max_gen_token_len_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">max_gen_token_len_range</span><span class="p">]</span>
|
|
<span class="n">bb_max_gen_token_len_range</span> <span class="o">=</span> <span class="p">[</span><span class="n">bb_max_gen_token_len_range</span><span class="p">]</span>
|
|
<span class="n">num_beams_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">]</span>
|
|
<span class="n">beam_length_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">]</span>
|
|
<span class="n">candidate_length_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">beam_candidate_length</span><span class="p">]</span>
|
|
<span class="n">vocab_size_range</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">]</span>
|
|
|
|
<span class="n">device_request_types</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'device_request_types'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">trt</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size'</span><span class="p">,</span> <span class="n">bb_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">draft_tokens</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'draft_tokens'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">trt</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size_wt0'</span><span class="p">,</span> <span class="n">bb0_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'num_beams'</span><span class="p">,</span> <span class="n">num_beams_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'beam_length'</span><span class="p">,</span> <span class="n">beam_length_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">draft_indices</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'draft_indices'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">trt</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size_wt0'</span><span class="p">,</span> <span class="n">bb0_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'num_beams'</span><span class="p">,</span> <span class="n">num_beams_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'beam_length'</span><span class="p">,</span> <span class="n">beam_length_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">draft_probs</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span>
|
|
<span class="n">name</span><span class="o">=</span><span class="s1">'draft_probs'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span> <span class="o">-</span> <span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">vocab_size</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size_wt0'</span><span class="p">,</span> <span class="n">bb0_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'num_beams'</span><span class="p">,</span> <span class="n">num_beams_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'candidate_length'</span><span class="p">,</span> <span class="n">candidate_length_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'vocab_size'</span><span class="p">,</span> <span class="n">vocab_size_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">redrafter_inverted_temperature</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span>
|
|
<span class="n">name</span><span class="o">=</span><span class="s1">'redrafter_inverted_temperature'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s2">"batch_size"</span><span class="p">,</span> <span class="n">bb_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">rand_data_validation</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span>
|
|
<span class="n">name</span><span class="o">=</span><span class="s1">'rand_data_validation'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_beams</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">beam_length</span> <span class="o">-</span> <span class="mi">1</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size_wt0'</span><span class="p">,</span> <span class="n">bb0_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'num_beams'</span><span class="p">,</span> <span class="n">num_beams_range</span><span class="p">),</span>
|
|
<span class="p">(</span><span class="s1">'candidate_length'</span><span class="p">,</span> <span class="n">candidate_length_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">rand_data_sample</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">'rand_data_sample'</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s1">'batch_size'</span><span class="p">,</span> <span class="n">bb_range</span><span class="p">),</span>
|
|
<span class="p">]))</span>
|
|
<span class="n">position_ids_base</span> <span class="o">=</span> <span class="n">Tensor</span><span class="p">(</span>
|
|
<span class="n">name</span><span class="o">=</span><span class="s2">"position_ids_base"</span><span class="p">,</span>
|
|
<span class="n">dtype</span><span class="o">=</span><span class="n">trt</span><span class="o">.</span><span class="n">int32</span><span class="p">,</span>
|
|
<span class="n">shape</span><span class="o">=</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">],</span>
|
|
<span class="n">dim_range</span><span class="o">=</span><span class="n">OrderedDict</span><span class="p">([</span>
|
|
<span class="p">(</span><span class="s2">"batch_size"</span><span class="p">,</span> <span class="n">bb_range</span><span class="p">),</span>
|
|
<span class="p">]),</span>
|
|
<span class="p">)</span>
|
|
|
|
<span class="n">inputs</span><span class="p">[</span>
|
|
<span class="s1">'device_request_types'</span><span class="p">]</span> <span class="o">=</span> <span class="n">device_request_types</span> <span class="c1"># needed by process_logits</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'draft_tokens'</span><span class="p">]</span> <span class="o">=</span> <span class="n">draft_tokens</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'draft_indices'</span><span class="p">]</span> <span class="o">=</span> <span class="n">draft_indices</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'draft_probs'</span><span class="p">]</span> <span class="o">=</span> <span class="n">draft_probs</span>
|
|
<span class="n">inputs</span><span class="p">[</span>
|
|
<span class="s1">'redrafter_inverted_temperature'</span><span class="p">]</span> <span class="o">=</span> <span class="n">redrafter_inverted_temperature</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'rand_data_validation'</span><span class="p">]</span> <span class="o">=</span> <span class="n">rand_data_validation</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'rand_data_sample'</span><span class="p">]</span> <span class="o">=</span> <span class="n">rand_data_sample</span>
|
|
<span class="n">inputs</span><span class="p">[</span><span class="s1">'position_ids_base'</span><span class="p">]</span> <span class="o">=</span> <span class="n">position_ids_base</span>
|
|
<span class="k">return</span> <span class="n">inputs</span></div>
|
|
</div>
|
|
|
|
</pre></div>
|
|
|
|
</div>
|
|
</div>
|
|
<footer>
|
|
|
|
<hr/>
|
|
|
|
<div role="contentinfo">
|
|
<jinja2.runtime.BlockReference object at 0x7f1ac950ddc0>
|
|
|
|
<div class="footer">
|
|
<p>
|
|
Copyright © 2024 NVIDIA Corporation
|
|
</p>
|
|
<p>
|
|
<a class="Link" href="https://www.nvidia.com/en-us/about-nvidia/privacy-policy/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Privacy Policy</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Manage My Privacy</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/preferences/start/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Do Not Sell or Share My Data</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/about-nvidia/terms-of-service/" target="_blank"
|
|
rel="noopener" data-cms-ai="0">Terms of Service</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/about-nvidia/accessibility/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Accessibility</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/about-nvidia/company-policies/" target="_blank"
|
|
rel="noopener" data-cms-ai="0">Corporate Policies</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/product-security/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Product Security</a> |
|
|
<a class="Link" href="https://www.nvidia.com/en-us/contact/" target="_blank" rel="noopener"
|
|
data-cms-ai="0">Contact</a>
|
|
</p>
|
|
</div>
|
|
|
|
|
|
</div>
|
|
|
|
|
|
|
|
</footer>
|
|
</div>
|
|
</div>
|
|
</section>
|
|
</div>
|
|
<script>
|
|
jQuery(function () {
|
|
SphinxRtdTheme.Navigation.enable(true);
|
|
});
|
|
</script>
|
|
|
|
</body>
|
|
</html> |