时间:2020-12-25 12:02:23 | 栏目:Python代码 | 点击:次
1 Pytorch以ONNX方式保存模型
def saveONNX(model, filepath): ''' 保存ONNX模型 :param model: 神经网络模型 :param filepath: 文件保存路径 ''' # 神经网络输入数据类型 dummy_input = torch.randn(self.config.BATCH_SIZE, 1, 28, 28, device='cuda') torch.onnx.export(model, dummy_input, filepath, verbose=True)
2 利用TensorRT5中ONNX解析器构建Engine
def ONNX_build_engine(onnx_file_path): ''' 通过加载onnx文件,构建engine :param onnx_file_path: onnx文件路径 :return: engine ''' # 打印日志 G_LOGGER = trt.Logger(trt.Logger.WARNING) with trt.Builder(G_LOGGER) as builder, builder.create_network() as network, trt.OnnxParser(network, G_LOGGER) as parser: builder.max_batch_size = 100 builder.max_workspace_size = 1 << 20 print('Loading ONNX file from path {}...'.format(onnx_file_path)) with open(onnx_file_path, 'rb') as model: print('Beginning ONNX file parsing') parser.parse(model.read()) print('Completed parsing of ONNX file') print('Building an engine from file {}; this may take a while...'.format(onnx_file_path)) engine = builder.build_cuda_engine(network) print("Completed creating Engine") # 保存计划文件 # with open(engine_file_path, "wb") as f: # f.write(engine.serialize()) return engine
3 构建TensorRT运行引擎进行预测
def loadONNX2TensorRT(filepath): ''' 通过onnx文件,构建TensorRT运行引擎 :param filepath: onnx文件路径 ''' # 计算开始时间 Start = time() engine = self.ONNX_build_engine(filepath) # 读取测试集 datas = DataLoaders() test_loader = datas.testDataLoader() img, target = next(iter(test_loader)) img = img.numpy() target = target.numpy() img = img.ravel() context = engine.create_execution_context() output = np.empty((100, 10), dtype=np.float32) # 分配内存 d_input = cuda.mem_alloc(1 * img.size * img.dtype.itemsize) d_output = cuda.mem_alloc(1 * output.size * output.dtype.itemsize) bindings = [int(d_input), int(d_output)] # pycuda操作缓冲区 stream = cuda.Stream() # 将输入数据放入device cuda.memcpy_htod_async(d_input, img, stream) # 执行模型 context.execute_async(100, bindings, stream.handle, None) # 将预测结果从从缓冲区取出 cuda.memcpy_dtoh_async(output, d_output, stream) # 线程同步 stream.synchronize() print("Test Case: " + str(target)) print("Prediction: " + str(np.argmax(output, axis=1))) print("tensorrt time:", time() - Start) del context del engine
补充知识:Pytorch/Caffe可以先转换为ONNX,再转换为TensorRT
近来工作,试图把Pytorch用TensorRT运行。折腾了半天,没有完成。github中的转换代码,只能处理pytorch 0.2.0的功能(也明确表示不维护了)。和同事一起处理了很多例外,还是没有通过。吾以为,实际上即使勉强过了,能不能跑也是问题。
后来有高手建议,先转换为ONNX,再转换为TensorRT。这个思路基本可行。
是不是这样就万事大吉?当然不是,还是有严重问题要解决的。这只是个思路。