TypeError: forward() takes 2 positional arguments but 3 were given,flownetc_flow2 = self.flownetc(x)[0].why!!!!

[zzk1212]

I spent a lot of time and still couldn't solve this problem. Has anyone encountered this problem? Can you provide a solution? Thank you so much!

Tasks

Beta Give feedback

No tasks being tracked yet.

import argparse

parser = argparse.ArgumentParser(description='Calculate volume of a cylinder')
parser.add_argument("--rgb_max", default=1)
parser.add_argument("--fp16", default=True)
args = parser.parse_args() # 获取所有参数

flownet2 = FlowNet2(args=args)

获取 conv1 的第一个子模块

first_module = flownet2.flownetc.conv1[0] # 假设 conv1 是一个包含一个子模块的 Sequential 对象

获取第一个子模块的输入通道数

expected_channels = first_module.in_channels
print("模型的期望输入通道数:", expected_channels)

创建符合期望通道数的随机输入张量

batch_size = 1
channels = 3
height = 64
width = 64
inputs = torch.randn(batch_size, channels, height, width) # 随机生成一个输入张量

将输入张量传递给模型获取输出

output = flownet2(inputs)