1. 新功能: caffe load API添加指定输入name的接口;添加了caffe lrn(WithinChannel)的支持 2. 添加了Caffe新OP支持:crop/flatten/normalize/proposal/reduction 3. 添加了onnx/pytorch/tensorflow/darknet/tflite新OP支持 4. 移除了aciq以及Kl散度量化功能 5. 修复一些已知的bug: 1) 最后一层是reshape转换bug;...
..., bin[i-1]] outliers_count = sum( bin[i], bin[i+1], ... , bin[2047]) reference_distribution_P[i-1] += outliers_count // take first 'i' bins from H P /= sum(P) // normalize distribution
(np.float32) + return data + + +class NormalizeTransformer(Transformer): + def __init__(self, std): + self.std = std + super(NormalizeTransformer, self).__init__() + + def run_transform(self, data): + data = data / self.std + data = data.astype(np.float32...
inputs, outputs, bindings, stream = common.allocate_buffers(engine) with engine.create_execution_context() as context: case_num = load_normalized_test_case(data_paths, pagelocked_buffer=inputs[0].host) [output] = common.do_inference(context, bindings=bindings, inputs=inputs, outputs=outputs,...