[Speculative Decoding] Return accepted tokens per head in response

[Deleter-D]

Motivation

💡 If this PR is a Cherry Pick, the PR title needs to follow the format by adding the [Cherry-Pick] label at the very beginning and appending the original PR ID at the end. For example, [Cherry-Pick][CI] Add check trigger and logic(#5191)

💡 如若此PR是Cherry Pick，PR标题需遵循格式，在最开始加上[Cherry-Pick]标签，以及最后面加上原PR ID，例如[Cherry-Pick][CI] Add check trigger and logic(#5191)

Modifications

Based on PR #5518, the number of accepted tokens per head is newly added to the response, with an example return shown below.

{
    "id": "xxx",
    "model": "xxx",
    "choices": [
        {
            ...
            "speculate_metrics": {
                "accepted_tokens": 128,
                "rejected_tokens": 112,
                "accept_ratio": 0.53125,
                "average_accept_length": 2.1333333333333333,
                "accepted_tokens_per_head": [
                   60,
                   36,
                   23,
                   9
                ]
                "accept_ratio_per_head": [
                    0.6,
                    0.6388888888888888,
                    0.391304347826087
                ]
            }
        }
    ],
    ...
}

The acceptance rate for each head is calculated as follows.

for i in range (1, len(accepted_tokens_per_head)):
    accept_ratio_per_head[i - 1] = accepted_tokens_per_head[i] / accepted_tokens_per_head[i - 1]

accepted_tokens_per_head[0] is equivalent to the number of decoding steps.

Usage or Command

Accuracy Tests

Checklist

• Add at least a tag in the PR title.
  • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
  • You can add new tags based on the PR content, but the semantics must be clear.
• Format your code, run pre-commit before commit.
• Add unit tests. Please write the reason in this PR if no unit tests.
• Provide accuracy results.
• If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

[paddle-bot]

Thanks for your contribution!

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (develop@d8c6ba6). Learn more about missing BASE report.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5947   +/-   ##
==========================================
  Coverage           ?   67.05%           
==========================================
  Files              ?      348           
  Lines              ?    44654           
  Branches           ?     6867           
==========================================
  Hits               ?    29943           
  Misses             ?    12502           
  Partials           ?     2209           

Flag	Coverage Δ
GPU	67.05% <100.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.