onnx export

游雁

2023-02-27 f98c4bf6d2bb5202488cd4243efdbca65288c313

onnx export

 .gitignore

@@ -7,3 +7,4 @@
init_model/
*.tar.gz
test_local/
RapidASR

 funasr/export/export_model.py

@@ -7,10 +7,12 @@
import logging
import torch

from funasr.bin.asr_inference_paraformer import Speech2Text
from funasr.export.models import get_model
import numpy as np
import random

torch_version = float(".".join(torch.__version__.split(".")[:2]))
assert torch_version > 1.9

class ASRModelExportParaformer:
    def __init__(self, cache_dir: Union[Path, str] = None, onnx: bool = True):
@@ -30,7 +32,7 @@

    def _export(
        self,
        model: Speech2Text,
        model,
        tag_name: str = None,
        verbose: bool = False,
    ):
@@ -112,110 +114,6 @@
            os.path.join(path, f'{model.model_name}.onnx'),
            verbose=verbose,
            opset_version=14,
            input_names=model.get_input_names(),
            output_names=model.get_output_names(),
            dynamic_axes=model.get_dynamic_axes()
        )


class ASRModelExport:
    def __init__(self, cache_dir: Union[Path, str] = None, onnx: bool = True):
        assert check_argument_types()
        self.set_all_random_seed(0)
        if cache_dir is None:
            cache_dir = Path.home() / ".cache" / "export"
        
        self.cache_dir = Path(cache_dir)
        self.export_config = dict(
            feats_dim=560,
            onnx=False,
        )
        print("output dir: {}".format(self.cache_dir))
        self.onnx = onnx
    
    def _export(
        self,
        model: Speech2Text,
        tag_name: str = None,
        verbose: bool = False,
    ):
        
        export_dir = self.cache_dir / tag_name.replace(' ', '-')
        os.makedirs(export_dir, exist_ok=True)
        
        # export encoder1
        self.export_config["model_name"] = "model"
        model = get_model(
            model,
            self.export_config,
        )
        model.eval()
        # self._export_onnx(model, verbose, export_dir)
        if self.onnx:
            self._export_onnx(model, verbose, export_dir)
        else:
            self._export_torchscripts(model, verbose, export_dir)
        
        print("output dir: {}".format(export_dir))
    
    def _export_torchscripts(self, model, verbose, path, enc_size=None):
        if enc_size:
            dummy_input = model.get_dummy_inputs(enc_size)
        else:
            dummy_input = model.get_dummy_inputs_txt()
        
        # model_script = torch.jit.script(model)
        model_script = torch.jit.trace(model, dummy_input)
        model_script.save(os.path.join(path, f'{model.model_name}.torchscripts'))
    
    def set_all_random_seed(self, seed: int):
        random.seed(seed)
        np.random.seed(seed)
        torch.random.manual_seed(seed)
    
    def export(self,
               tag_name: str = 'damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch',
               mode: str = 'paraformer',
               ):
        
        model_dir = tag_name
        if model_dir.startswith('damo/'):
            from modelscope.hub.snapshot_download import snapshot_download
            model_dir = snapshot_download(model_dir, cache_dir=self.cache_dir)
        asr_train_config = os.path.join(model_dir, 'config.yaml')
        asr_model_file = os.path.join(model_dir, 'model.pb')
        cmvn_file = os.path.join(model_dir, 'am.mvn')
        json_file = os.path.join(model_dir, 'configuration.json')
        if mode is None:
            import json
            with open(json_file, 'r') as f:
                config_data = json.load(f)
                mode = config_data['model']['model_config']['mode']
        if mode.startswith('paraformer'):
            from funasr.tasks.asr import ASRTaskParaformer as ASRTask
        elif mode.startswith('uniasr'):
            from funasr.tasks.asr import ASRTaskUniASR as ASRTask
        
        model, asr_train_args = ASRTask.build_model_from_file(
            asr_train_config, asr_model_file, cmvn_file, 'cpu'
        )
        self._export(model, tag_name)
    
    def _export_onnx(self, model, verbose, path, enc_size=None):
        if enc_size:
            dummy_input = model.get_dummy_inputs(enc_size)
        else:
            dummy_input = model.get_dummy_inputs()
        
        # model_script = torch.jit.script(model)
        model_script = model  # torch.jit.trace(model)
        
        torch.onnx.export(
            model_script,
            dummy_input,
            os.path.join(path, f'{model.model_name}.onnx'),
            verbose=verbose,
            opset_version=12,
            input_names=model.get_input_names(),
            output_names=model.get_output_names(),
            dynamic_axes=model.get_dynamic_axes()

 funasr/export/models/predictor/cif.py

@@ -77,6 +77,53 @@
        return hidden, alphas, token_num_floor





# @torch.jit.script

# def cif(hidden, alphas, threshold: float):

#     batch_size, len_time, hidden_size = hidden.size()

#     threshold = torch.tensor([threshold], dtype=alphas.dtype).to(alphas.device)

#

#     # loop varss

#     integrate = torch.zeros([batch_size], device=hidden.device)

#     frame = torch.zeros([batch_size, hidden_size], device=hidden.device)

#     # intermediate vars along time

#     list_fires = []

#     list_frames = []

#

#     for t in range(len_time):

#         alpha = alphas[:, t]

#         distribution_completion = torch.ones([batch_size], device=hidden.device) - integrate

#

#         integrate += alpha

#         list_fires.append(integrate)

#

#         fire_place = integrate >= threshold

#         integrate = torch.where(fire_place,

#                                 integrate - torch.ones([batch_size], device=hidden.device),

#                                 integrate)

#         cur = torch.where(fire_place,

#                           distribution_completion,

#                           alpha)

#         remainds = alpha - cur

#

#         frame += cur[:, None] * hidden[:, t, :]

#         list_frames.append(frame)

#         frame = torch.where(fire_place[:, None].repeat(1, hidden_size),

#                             remainds[:, None] * hidden[:, t, :],

#                             frame)

#

#     fires = torch.stack(list_fires, 1)

#     frames = torch.stack(list_frames, 1)

#     list_ls = []

#     len_labels = torch.floor(alphas.sum(-1)).int()

#     max_label_len = len_labels.max()

#     for b in range(batch_size):

#         fire = fires[b, :]

#         l = torch.index_select(frames[b, :, :], 0, torch.nonzero(fire >= threshold).squeeze())

#         pad_l = torch.zeros([int(max_label_len - l.size(0)), int(hidden_size)], device=hidden.device)

#         list_ls.append(torch.cat([l, pad_l], 0))

#     return torch.stack(list_ls, 0), fires





@torch.jit.script

def cif(hidden, alphas, threshold: float):

    batch_size, len_time, hidden_size = hidden.size()

@@ -113,15 +160,11 @@
    

    fires = torch.stack(list_fires, 1)

    frames = torch.stack(list_frames, 1)

    # list_ls = []

    len_labels = torch.round(alphas.sum(-1)).type(torch.int32)

    # max_label_len = int(torch.max(len_labels).item())

    # print("type: {}".format(type(max_label_len)))



    fire_idxs = fires >= threshold

    frame_fires = torch.zeros_like(hidden)

    max_label_len = frames[0, fire_idxs[0]].size(0)

    for b in range(batch_size):

        # fire = fires[b, :]

        frame_fire = frames[b, fire_idxs[b]]

        frame_len = frame_fire.size(0)

        frame_fires[b, :frame_len, :] = frame_fire


 funasr/runtime/python/onnxruntime/rapid_paraformer/utils/utils.py

@@ -148,6 +148,7 @@

class OrtInferSession():
    def __init__(self, model_file, device_id=-1):
        device_id = str(device_id)
        sess_opt = SessionOptions()
        sess_opt.log_severity_level = 4
        sess_opt.enable_cpu_mem_arena = False
@@ -166,7 +167,7 @@
        }

        EP_list = []
        if device_id != -1 and get_device() == 'GPU' \
        if device_id != "-1" and get_device() == 'GPU' \
                and cuda_ep in get_available_providers():
            EP_list = [(cuda_ep, cuda_provider_options)]
        EP_list.append((cpu_ep, cpu_provider_options))
@@ -176,7 +177,7 @@
                                        sess_options=sess_opt,
                                        providers=EP_list)

        if device_id != -1 and cuda_ep not in self.session.get_providers():
        if device_id != "-1" and cuda_ep not in self.session.get_providers():
            warnings.warn(f'{cuda_ep} is not avaiable for current env, the inference part is automatically shifted to be executed under {cpu_ep}.\n'
                          'Please ensure the installed onnxruntime-gpu version matches your cuda and cudnn version, '
                          'you can check their relations from the offical web site: '

 scan.py