mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
aa27b03bc0
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
* success run ngram * Revert "[Code Simplification] remove cum_offsets (#3410)" This reverts commit32b39620bc. * success run ngram5 tp4 42bs * success run ngram5 tp4 42bs * mtp draft commit * add decorator for target model * enable draft model in cudagraph v0.5 * revert revrt cum_offset * enable target model in cudagraph v0.9 And clean debug code * Revert "success run ngram" This reverts commit8351e83993. * add reverted code * enable target model in cudagraph v0.9 * solve comment * fix bid < 0 * Enable Target Model Padding And Draft Model in cudagraph * solve problem * delete rebuild padding debug note * fast compile * Add capture list for mtp * success run 256 tp1 mtp * Enable Lite TP2 Bsz256 * realy enable tp2 bsz 256 * fix problem * Solve problem for Draft model in cudagraph * Solve comment * replace emptytensor as zeros * Solve comments * Revert "fast compile" This reverts commit834639a7ff. * fix bug * fix merge bug * fix typo * fix bug --------- Co-authored-by: lizexu <2694294196@qq.com> Co-authored-by: littledgg <1658565283@qq.com> Co-authored-by: zeroRains <linjunlu@zerorains.top> Co-authored-by: gongshaotian <gstain5555@outlook.com>
69 lines
2.3 KiB
Python
69 lines
2.3 KiB
Python
"""
|
|
# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License"
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
"""
|
|
|
|
import paddle
|
|
|
|
from fastdeploy.config import FDConfig
|
|
from fastdeploy.model_executor.ops.gpu import ngram_match
|
|
|
|
from .base import Proposer
|
|
|
|
|
|
class NgramProposer(Proposer):
|
|
"""
|
|
Proposer for Ngram match method.
|
|
|
|
Matching corresponding tokens in input and output as draft tokens.
|
|
"""
|
|
|
|
def __init__(self, fd_config: FDConfig):
|
|
super().__init__(fd_config)
|
|
self.max_ngram_size = self.speculative_config.max_ngram_size
|
|
self.input_ids_len = paddle.zeros(shape=[self.max_num_seqs, 1], dtype="int64").cpu()
|
|
|
|
def update(self, bid: int, seq_len: int):
|
|
"""
|
|
update
|
|
"""
|
|
self.input_ids_len[bid] = seq_len
|
|
|
|
def _run_impl(self, share_inputs):
|
|
"""
|
|
run
|
|
"""
|
|
draft_tokens = share_inputs["draft_tokens"].cpu()
|
|
seq_lens_this_time = share_inputs["seq_lens_this_time"].cpu()
|
|
seq_lens_encoder = share_inputs["seq_lens_encoder"].cpu()
|
|
seq_lens_decoder = share_inputs["seq_lens_decoder"].cpu()
|
|
|
|
ngram_match(
|
|
share_inputs["input_ids_cpu"],
|
|
self.input_ids_len.cpu(),
|
|
share_inputs["pre_ids"].cpu(),
|
|
share_inputs["step_idx"].cpu(),
|
|
share_inputs["actual_draft_token_num"].cpu(),
|
|
draft_tokens,
|
|
seq_lens_this_time,
|
|
seq_lens_encoder,
|
|
seq_lens_decoder,
|
|
share_inputs["max_dec_len"].cpu(),
|
|
self.max_ngram_size,
|
|
self.max_draft_token_num,
|
|
)
|
|
share_inputs["draft_tokens"][:] = draft_tokens.cuda()
|
|
share_inputs["seq_lens_encoder"][:] = seq_lens_encoder.cuda()
|
|
share_inputs["seq_lens_this_time"][:] = seq_lens_this_time.cuda()
|