ValueError: Invalid filepath extension for saving. Please add either a `.keras` extension for the native Keras format (recommended) or a `.h5` extension.

[omolchanov]

model = Sequential() model.add(Dense(1, activation='relu', kernel_initializer='normal')) model.compile(loss='mean_absolute_error', optimizer='adam', metrics=['mean_absolute_error'])

estimator = KerasRegressor(model=model, epochs=1, batch_size=15, verbose=0) results = cross_val_score(estimator, X, y, cv=cv, scoring='neg_mean_squared_error',)

---

Error is raised here '_saving_utils.py'

"""Support for Pythons's Pickle protocol.""" with _get_temp_folder() as temp_dir: model.save(tempdir) b = BytesIO() with tarfile.open(fileobj=b, mode="w") as archive: for root, , filenames in tf_io.gfile.walk(temp_dir): for filename in filenames: dest = os.path.join(root, filename) with tf_io.gfile.GFile(dest, "rb") as f: info = tarfile.TarInfo(name=os.path.relpath(dest, temp_dir)) info.size = f.size() archive.addfile(tarinfo=info, fileobj=f) tf_io.gfile.remove(dest) b.seek(0) model_bytes = np.asarray(memoryview(b.read())) return (unpack_keras_model, (model_bytes,))

[adriangb]

Can you provide a MRE? Thanks!

[omolchanov]

Can you provide a MRE? Thanks!

Sure, thanks!

from pandas import read_csv
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
from scikeras.wrappers import KerasRegressor
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold

dataframe = read_csv("housing.csv", delim_whitespace=True, header=None)
dataset = dataframe.values

X = dataset[:,0:13]
Y = dataset[:,13]

def baseline_model():

    model = Sequential()
    model.add(Dense(13, input_shape=(13,), kernel_initializer='normal', activation='relu'))
    model.add(Dense(1, kernel_initializer='normal'))

    model.compile(loss='mean_squared_error', optimizer='adam')
    return model

estimator = KerasRegressor(model=baseline_model, epochs=100, batch_size=5, verbose=0)
kfold = KFold(n_splits=10)
results = cross_val_score(estimator, X, Y, cv=kfold, scoring='neg_mean_squared_error')
print("Baseline: %.2f (%.2f) MSE" % (results.mean(), results.std()))

[omolchanov]

Adrian, any luck with this broken saving logic ?

[adriangb]

I haven't had time to look into this yet. It'll probably be next week.

[omolchanov]

Hello Adrian. Sorry to bothering you with this, but do you have any progress with fixing this adapter ?

Thanks

[adriangb]

I'm looking into #315 and I feel like you may be using a newer version of Keras. What version of keras are you using?

[omolchanov]

Keras version is 3.0.5.

[adriangb]

Okay yes we don't support that version, see #315. I'm going to close this since it will be fixed via that issue.

[omolchanov]

Ok, thank you