I want to deploy Inference of TVM on Windows. To this end, I did the following work:
1. Clone the tvm source code from github
2. Download LLVM source code，Visual studio build it
3. Build tvm with llvm, modify config.cmake (USE LLVM OFF to ON)
4. into tvm root path
cd python; python setup.py install --user; cd .. cd topi/python; python setup.py install --user; cd ../.. cd nnvm/python; python setup.py install --user; cd ../..`
After finishing the above work:
we can find this file in Release
Generate the TVM model
import numpy as np
import nnvm.compiler
import nnvm.testing
import tvm
from tvm.contrib import graph_runtime
import mxnet as mx
from mxnet import ndarray as nd
prefix,epoch = "model",0
sym, arg_params, aux_params = mx.model.load_checkpoint(prefix, epoch)
image_size = (112, 112)
opt_level = 3
shape_dict = {'data': (1, 3, *image_size)}
target = tvm.target.create("llvm -mcpu=haswell")
# "target" means your target platform you want to compile.
#target = tvm.target.create("llvm -mcpu=broadwell")
nnvm_sym, nnvm_params = nnvm.frontend.from_mxnet(sym, arg_params, aux_params)
with nnvm.compiler.build_config(opt_level=opt_level):
graph, lib, params = nnvm.compiler.build(nnvm_sym, target, shape_dict, params=nnvm_params)
lib.export_library("./deploy_lib.dll")
print('lib export succeefully')
with open("./deploy_graph.json", "w") as fo:
fo.write(graph.json())
with open("./deploy_param.params", "wb") as fo:
fo.write(nnvm.compiler.save_param_dict(params))
Then we can find tvm model in current folder
