input and filter must have the same depth: 4 vs 3

[TD-101]

Hi,

I get this error / exception, and while it is being, handled, the same error/exception occurs.
I am fairly new to this so it could be a multitude of problems, from the way I have set up python, tf etc, to improper hardware, but I thought I would put it up here in case someone had an easy fix!

Traceback (most recent call last):
File "/usr/local/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 965, in _do_call
return fn(*args)
File "/usr/local/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 947, in _run_fn
status, run_metadata)
File "/usr/local/Cellar/python3/3.5.2_1/Frameworks/Python.framework/Versions/3.5/lib/python3.5/contextlib.py", line 66, in exit
next(self.gen)
File "/usr/local/lib/python3.5/site-packages/tensorflow/python/framework/errors.py", line 450, in raise_exception_on_not_ok_status
pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors.InvalidArgumentError: input and filter must have the same depth: 4 vs 3
[[Node: import/conv2d0_pre_relu/conv = Conv2D[T=DT_FLOAT, data_format="NHWC", padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](ExpandDims, import/conv2d0_w)]]

[ghost]

@tomdawson91 have you solved it? I had the same problem.

[gonzalolc]

Same issue here! Have anyone solved it?

[ghost]

img0 = np.float32(img0)[:,:,:3]

[gonzalolc]

It works! thanks @dattranx

[TD-101]

@dattranx Thanks - as I understand it, this just cuts out the alpha channel

[R-Miner]

I get an error like: Status(StatusCode=InvalidArgument, Detail="input and filter must have the same depth: 1 vs 3
[[Node: conv2d_1/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="VALID", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_input_1_0_0, conv2d_1/kernel/read)]]")'

Any thoughts on this? for me it is 1 Vs 3

[CyLouisKoo]

I have also met the problem and How you resolved it then? Thank you for your reply. @TD-101

[vishalvanpariya]

@gonzalolc please explain in deep i cannot understood

[khanfarhan10]

I am coding Grad-CAM with keras vis
I tried
seed_input =tf.convert_to_tensor(img[:,:,:3])
seed_input=np.float32(img)[:,:,:3]

But none of them worked for me
I get the following error

InvalidArgumentError Traceback (most recent call last)
in ()
20 penultimate_layer_idx = penultimate_layer_idx,#None,
21 backprop_modifier = None,
---> 22 grad_modifier = None)

8 frames
/usr/local/lib/python3.6/dist-packages/vis/visualization/saliency.py in visualize_cam(model, layer_idx, filter_indices, seed_input, penultimate_layer_idx, backprop_modifier, grad_modifier)
237 (ActivationMaximization(model.layers[layer_idx], filter_indices), -1)
238 ]
--> 239 return visualize_cam_with_losses(model.input, losses, seed_input, penultimate_layer, grad_modifier)

/usr/local/lib/python3.6/dist-packages/vis/visualization/saliency.py in visualize_cam_with_losses(input_tensor, losses, seed_input, penultimate_layer, grad_modifier)
158 penultimate_output = penultimate_layer.output
159 opt = Optimizer(input_tensor, losses, wrt_tensor=penultimate_output, norm_grads=False)
--> 160 _, grads, penultimate_output_value = opt.minimize(seed_input, max_iter=1, grad_modifier=grad_modifier, verbose=False)
161
162 # For numerical stability. Very small grad values along with small penultimate_output_value can cause

/usr/local/lib/python3.6/dist-packages/vis/optimizer.py in minimize(self, seed_input, max_iter, input_modifiers, grad_modifier, callbacks, verbose)
141
142 # 0 learning phase for 'test'
--> 143 computed_values = self.compute_fn([seed_input, 0])
144 losses = computed_values[:len(self.loss_names)]
145 named_losses = zip(self.loss_names, losses)

/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/backend.py in call(self, inputs)
3790 value = math_ops.cast(value, tensor.dtype)
3791 converted_inputs.append(value)
-> 3792 outputs = self._graph_fn(*converted_inputs)
3793
3794 # EagerTensor.numpy() will often make a copy to ensure memory safety.

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in call(self, *args, **kwargs)
1603 TypeError: For invalid positional/keyword argument combinations.
1604 """
-> 1605 return self._call_impl(args, kwargs)
1606
1607 def _call_impl(self, args, kwargs, cancellation_manager=None):

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in _call_impl(self, args, kwargs, cancellation_manager)
1643 raise TypeError("Keyword arguments {} unknown. Expected {}.".format(
1644 list(kwargs.keys()), list(self._arg_keywords)))
-> 1645 return self._call_flat(args, self.captured_inputs, cancellation_manager)
1646
1647 def _filtered_call(self, args, kwargs):

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in _call_flat(self, args, captured_inputs, cancellation_manager)
1744 # No tape is watching; skip to running the function.
1745 return self._build_call_outputs(self._inference_function.call(
-> 1746 ctx, args, cancellation_manager=cancellation_manager))
1747 forward_backward = self._select_forward_and_backward_functions(
1748 args,

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py in call(self, ctx, args, cancellation_manager)
596 inputs=args,
597 attrs=attrs,
--> 598 ctx=ctx)
599 else:
600 outputs = execute.execute_with_cancellation(

/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
58 ctx.ensure_initialized()
59 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
---> 60 inputs, attrs, num_outputs)
61 except core._NotOkStatusException as e:
62 if name is not None:

InvalidArgumentError: input depth must be evenly divisible by filter depth: 443 vs 3
[[node conv2d_1_1/convolution (defined at /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3009) ]] [Op:__inference_keras_scratch_graph_10171]

Function call stack:
keras_scratch_graph

[asr-aditya]

The error is because of mismatch in dimensions of the input provided. The model requires a depth of '3' for the input but is given '4'.

[AnjanaChankya]

whats depth of 3 means

[asr-aditya]

whats depth of 3 means

It means if you are giving an image, it has 3 channels i.e. size of image is (256,256,3)

[LucasColas]

Yes you have to change the depth.

[Umraz-Hussain-MyWorld]

[Umraz-Hussain-MyWorld]

some one please help

[Zapbbx]

I'm not an expert, but it seems that the 2 source images I'm using don't have the same number of channels. IE, I'm using PNG files and one set of them has a transparent background (Alpha 0) - Wheras the other set of images has colored backgrounds..
Saving both sets of files as JPG images, works around the error. I think because this gets rid of the transparent (alpha) channel. IE, the "shape" of your data needs to be the same.. If you feed a grayscale image into CNN which expects a color image. Find shape of input, e.g. print(model.input.shape) in Keras, you get (None, 224, 224, 3) and your input blob must have a corresponding shape, so having a grayscale image you have to convert it into a color image all three channels need to be the same.