# How to determine input shape in keras?

I am having difficulty finding where my error is while building deep learning models, but I typically have issues when setting the input layer input shape.

This is my model:

model = Sequential([
Dense(32, activation='relu', input_shape=(1461, 75)),
Dense(32, activation='relu'),
Dense(ytrain.size),])


It is returning the following error:

 ValueError: Error when checking input: expected dense_1_input to have 3

dimensions, but got array with shape (1461, 75)


The array is the training set from the kaggle housing price competition and my dataset has 75 columns and 1461 rows. My array is 2 dimensional, so why are 3 dimensions expected? I have tried adding a redundant 3rd dimension of 1 or flattening the array before the first dense layer but the error simply becomes:

ValueError: Input 0 is incompatible with layer flatten_1: expected

min_ndim=3, found ndim=2


How do you determine what the input size should be and why do the dimensions it expects seem so arbitrary?

For reference, I attached the rest of my code:

xtrain = pd.read_csv("pricetrain.csv")
xtrain.fillna(xtrain.mean(), inplace=True)
xtrain.drop(["Alley"], axis=1, inplace=True)
xtrain.drop(["PoolQC"], axis=1, inplace=True)
xtrain.drop(["Fence"], axis=1, inplace=True)
xtrain.drop(["MiscFeature"], axis=1, inplace=True)
xtrain.drop(["PoolArea"], axis=1, inplace=True)
columns = list(xtrain)
for i in columns:
if xtrain[i].dtypes == 'object':
xtrain[i] = pd.Categorical(pd.factorize(xtrain[i])[0])
from sklearn import preprocessing

le = preprocessing.LabelEncoder()
for i in columns:
if xtrain[i].dtypes == 'object':
xtrain[i] = le.fit_transform(xtrain[i])
ytrain = xtrain["SalePrice"]
xtrain.drop(["SalePrice"], axis=1, inplace=True)
ytrain = ytrain.values
xtrain = xtrain.values
ytrain.astype("float32")

size = xtrain.size
print(ytrain)
model = Sequential(
[Flatten(),
Dense(32, activation='relu', input_shape=(109575,)),
Dense(32, activation='relu'),
Dense(ytrain.size),
])
model.fit(xtrain, ytrain, epochs=10, verbose=1)


Thank you.

• I checked this piece of code and I am wondering on the layer definition in the Sequential model. Running similar structure on my end, ValueError is thrown as in the first command, "Flatten()", the input shape is missing / not defined... Dec 28, 2021 at 18:11

The number of rows in your training data is not part of the input shape of the network because the training process feeds the network one sample per batch (or, more precisely, batch_size samples per batch). And in input_shape, the batch dimension is not included for the first layer. You can read more on this here.

So, the input shape for your problem will be:

input_shape=(75, )

• I changed my input shape to your suggestion and the error simply becomes Error when checking target: expected dense_3 to have shape (1461,) but got array with shape (1,) I tried adding a flatten layer but then the error became Input 0 is incompatible with layer flatten_1: expected min_ndim=3, found ndim=2  How can I fix this issue? Jun 12, 2019 at 4:00
• In which line are you getting this error? This is not the same as previous one as that one was on checking input and this is while checking target. Jun 12, 2019 at 4:09
• what is the difference between those two notes? and the error comes from the model.compile() but it corresponds to dense_3 does that make it the third dense layer? Jun 12, 2019 at 4:29
• Sorry for the late reply, and yes the error is now in 3rd layer. I think the size of last layer is the problem. If you are trying to predict a single value of housing price then the last layer must have 1 unit rather than ytrain.size Jun 13, 2019 at 4:14
• that is perfect the model is now compiling. Thank you! one last question, do you have any idea why the loss function might not be decreasing? (i.e not learning)? Jun 13, 2019 at 20:15

Try this bunch of code:

x_train=x_train.reshape(-1,75,1)
`

but before you train(fit) model

Negative one (-1) in reshape(-1,75,1) simply mean, that you don't know how much should be in first dimension, but you know that second one should be equals 75 and last one 1.