TVM Compiler中文教程：TVM部署SSD模型

一时失言乱红尘 2022-01-23 07:41 353阅读 0赞

# 部署SSD模型 #

我们使用GluonCV预训练SSD模型和转换成Relay IR中间表示

import tvm
    
    from matplotlib import pyplot as plt
    from tvm.relay.testing.config import ctx_list
    from tvm import relay
    from tvm.contrib import graph_runtime
    from tvm.contrib.download import download_testdata
    from gluoncv import model_zoo, data, utils
    
    supported_model = [
        'ssd_512_resnet50_v1_voc',
        'ssd_512_resnet50_v1_coco',
        'ssd_512_resnet101_v2_voc',
        'ssd_512_mobilenet1.0_voc',
        'ssd_512_mobilenet1.0_coco',
    ]
    
    model_name = supported_model[0]
    dshape = (1, 3, 512, 512)
    target_list = ctx_list()
    
    # 下载和预处理demo图像
    im_fname = download_testdata('https://github.com/dmlc/web-data/blob/master/' +
                                 'gluoncv/detection/street_small.jpg?raw=true',
                                 'street_small.jpg', module='data')
    x, img = data.transforms.presets.ssd.load_test(im_fname, short=512)
    
    #转换和编译模型for CPU
    block = model_zoo.get_model(model_name, pretrained=True)
    def build(target):
        net, params = relay.frontend.from_mxnet(block,{ "data": dshape})
        with relay.build_config(opt_level=3):
            graph, lib, params = relay.build(net,target,params=params)
            return graph, lib, params
    
    #创建TVM运行时和推理
    def run(graph, lib, params, ctx):
        #创建TVM运行时
        m = graph_runtime.create(graph, lib, ctx)
        tvm_input = tvm.nd.array(x.asnumpy(), ctx=ctx)
        m.set_input('data', tvm_input)
        m.set_input(**params)
        #执行
        m.run()
        #获取输出
        class_IDs, scores, bounding_boxs = m.get_output(0), m.get_output(1), m.get_output(2)
        return class_IDs, scores, bounding_boxs
    
    
    for target, ctx in target_list:
        graph, lib, params = build(target)
        class_IDs, scores, bounding_boxs = run(graph, lib, params, ctx)
    
    #显示结果
    ax = utils.viz.plot_bbox(img, bounding_boxs,asnumpy()[0], scores.asnumpy()[0],class_IDs.asnumpy()[0],class_names=block.classes)
    plt.show