pytorch/caffe2/python/onnx/onnxifi.py
Yinghai Lu 70ee257ad4 Fix batch insert (#17158)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17158

Because of Reshape op, batch size can be changed. This diff addresses first order issue raised from multiple batch size system. We need to export different real_batch_size for different max_batch_size input and attach it to the right output.

It also fixes a false exception.

Reviewed By: ipiszy

Differential Revision: D14099541

fbshipit-source-id: 0fa9e86826f417a11d2b5dd2ee60dff64a7ce8c4
2019-02-15 12:28:23 -08:00

43 lines
1.2 KiB
Python

## @package onnx
#Module caffe2.python.onnx.onnxifi
"""
ONNXIFI a Caffe2 net
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals
from caffe2.proto import caffe2_pb2
from caffe2.python import core, workspace
import caffe2.python._import_c_extension as C
import numpy as np
def onnxifi_caffe2_net(
pred_net,
input_shapes,
max_batch_size=1,
max_seq_size=1,
debug=False,
use_onnx=True,
black_list=None):
"""
Transform the caffe2_net by collapsing ONNXIFI-runnable nodes into Onnxifi c2 ops
"""
shape_hints = {}
for k, v in input_shapes.items():
shape_hints[k] = v
pred_net_str = C.onnxifi(pred_net.SerializeToString(),
shape_hints,
black_list if black_list else [],
max_batch_size,
max_seq_size,
debug,
use_onnx)
pred_net_cut = caffe2_pb2.NetDef()
pred_net_cut.ParseFromString(pred_net_str)
return pred_net_cut