ValueError: Layer conv2d_246 expects 1 inputs, but it received 2 input tensors. Inputs received: [<tf.Tensor 'max_pooling2d_84/MaxPool:0' shape=(None, 48, 64, 32) dtype=float32>, <tf.Tensor 'conv2d_245/Relu:0' shape=(None, 96, 128, 32) dtype=float32>]

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

getting the following errors :ValueError: Layer conv2d_246 expects 1 inputs, but it received 2 input tensors. Inputs received: [<tf.Tensor ‘max_pooling2d_84/MaxPool:0’ shape=(None, 48, 64, 32) dtype=float32>, <tf.Tensor ‘conv2d_245/Relu:0’ shape=(None, 96, 128, 32) dtype=float32>]

2

Hi shivraj,

Have a look of the output of the conv_block function. What does this mean for cblock1 through cblock4?

3

conv_block (cblocks) gives 2 outputs a=& you’ve to use 1st one in all 5 lines in the first half (while downsampling) & 2nd one in all 5 lines in the second half (while upsampling). While you’re directly using the output of block (Except for ublock6 = ...., you’ve not indexed the block outputs anywhere).

Example:
Else everything is correct 100%. Hope this explains.

4

Hi abdur75648,

Good that you give useful advice! But could you please remove the code suggestions? This will give shivraj the possibility to think about the coding… Thanks!

5

Sure, But how can I edit?

6

Hi abdur75648,

Thanks a lot for your response and indicating that you cannot edit your post. I’ll see if I can find someone at the backend who can remove the code. Thanks!