issues in help docs of Con2D and Conv1D

[benwu232]

https://keras.io/layers/convolutional/#conv2d

data_format: A string, one of channels_last (default) or channels_first. The ordering of the dimensions in the inputs. channels_last corresponds to inputs with shape (batch, width, height, channels) while channels_first corresponds to inputs with shape (batch, channels, width, height)

I think channels_last should be (batch, height, width, channels), which is NHWC. And similar to channels_first. There are many places in the help doc like this.

https://keras.io/layers/convolutional/#conv1d

Input shape

3D tensor with shape: (batch_size, steps, input_dim)

The means of steps and input_dim are not clear. I think steps is the length of the time series, input_dim is features in each time step, similar to the channels in Conv2D. Am I right?

[abhaikollara]

The issue has been resolved in #5854

[benwu232]

Seems the doc in keras.io is not modified.

BTW, is my understanding of Conv1D right?

Thanks!

[ghost]

I agree that the documentation is way too brief and it is not clear for conv1d, parameter 'data_format', what the batch_size and what steps means in the (batch_size, steps, channels) ordering. It would be sometimes best if there could be a brief example, for example telling 'if you have 7 features and 10000 samples in your dataset, of which each one has exactly one numeric value, then you have 1 channel, 7 steps and batch_size=10000' (or steps = 10000 and batch_size=7, at the moment I don't know what is correct).