tensorflow - Graph disconnected : cannot obtain value for tensor Tensor

Question

I am trying to preprocess the data for the Titanic problem on Kaggle, following this link

train_data = pd.read_csv("/kaggle/input/titanic/train.csv")

train_data = train_data.drop(columns=['Cabin'])

titanic_features = train_data.copy()
titanic_labels = titanic_features.pop('Survived')



inputs = {}

for name, column in titanic_features.items():
    dtype = column.dtype
    if dtype == object:
        dtype = tf.string
    else:
        dtype = tf.float32
    
    inputs[name] = tf.keras.Input(shape=(1,), name=name, dtype=dtype)



numeric_inputs = {name:input for name,input in inputs.items()
                  if input.dtype==tf.float32}



x = layers.Concatenate()(list(numeric_inputs.values()))
norm = preprocessing.Normalization()
norm.adapt(np.array(train_data[numeric_inputs.keys()]))
all_numeric_inputs = norm(x)


preprocessed_inputs = [all_numeric_inputs]



for name, inpute in inputs.items():
    if inpute.dtype == tf.float32:
        continue

    lookup = preprocessing.StringLookup(vocabulary=np.unique(titanic_features[name].dropna()))
    one_hot = preprocessing.CategoryEncoding(max_tokens=lookup.vocab_size())

    x = lookup(input)
    x = one_hot(x)
    preprocessed_inputs.append(x)

    
preprocessed_inputs_cat = layers.Concatenate()(preprocessed_inputs)

titanic_preprocessing = tf.keras.Model(inputs, preprocessed_inputs_cat)

I have the following error:

ValueError: Graph disconnected: cannot obtain value for tensor Tensor("Cabin_17:0", shape=(None, 1), dtype=string) at layer "string_lookup_214". The following previous layers were accessed without issue: []

So I removed the 'Cabin' column (as you can see in my code), but the problem is still here. I have checked all other arrays, datas etc. in my code, there is not any trace of 'Cabin'.

I really cannot see what's going wrong here.

Thanks a lot

Aymeric

question from:https://stackoverflow.com/questions/65872290/graph-disconnected-cannot-obtain-value-for-tensor-tensor

Answer 1

Waitting for answers