调用层conv2d出现异常

[leejq666]

作者您好，我将基于pytorch训练得到的onnx模型转化为tflite的过程中出现报错：
ValueError: Exception encountered when calling layer "conv2d" (type Conv2D).

Negative dimension size caused by subtracting 3 from 2 for '{{node conv2d/Conv2D}} = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], explicit_paddings=[], padding="VALID", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true](Placeholder, conv2d/Conv2D/ReadVariableOp)' with input shapes: [1,2,2,3], [3,3,3,32].

Call arguments received by layer "conv2d" (type Conv2D):
• inputs=tf.Tensor(shape=(1, 2, 2, 3), dtype=float32)

[leejq666]

看了下评论区没有相似的错误，请问您有思路吗