Develop Reference

Weird Error in Regression Model using Artificial Neural Networks in Tensorflow

I am running this code.
import math
import pandas as pd
import tensorflow as tf
import matplotlib.pyplot as plt
from tensorflow.keras import Model
from tensorflow.keras import Sequential
from tensorflow.keras.optimizers import Adam
from sklearn.preprocessing import StandardScaler
from tensorflow.keras.layers import Dense, Dropout
from sklearn.model_selection import train_test_split
from tensorflow.keras.losses import MeanSquaredLogarithmicError
numerics = ['int16', 'int32', 'int64', 'float16', 'float32', 'float64']
df_new = df_deduped.select_dtypes(include=numerics)
X = df_new[['Duration',
'Customers_Affected',
'Customers_Served',
'Total_Minutes',
'CostOfOutage',
'PercentAffected']]
y = df_new[['HT_Outages']]
print(X.head(1))
print(y.head(1))
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42, test_size=0.2)
object= StandardScaler()
# standardization
X_train_scaled = object.fit_transform(X_train)
X_test_scaled = object.fit_transform(X_test)
hidden_units1 = 160
hidden_units2 = 480
hidden_units3 = 256
learning_rate = 0.01
# Creating model using the Sequential in tensorflow
def build_model_using_sequential():
model = Sequential([
Dense(hidden_units1, kernel_initializer='normal', activation='relu'),
Dropout(0.2),
Dense(hidden_units2, kernel_initializer='normal', activation='relu'),
Dropout(0.2),
Dense(hidden_units3, kernel_initializer='normal', activation='relu'),
Dense(1, kernel_initializer='normal', activation='linear')
])
return model
# build the model
model = build_model_using_sequential()
# loss function
msle = MeanSquaredLogarithmicError()
model.compile(
loss=msle,
optimizer=Adam(learning_rate=learning_rate),
metrics=[msle]
)
# train the model
history = model.fit(
x_train_scaled,
y_train,
epochs=25,
batch_size=64,
validation_split=0.2
)
def plot_history(history, key):
plt.plot(history.history[key])
plt.plot(history.history['val_'+key])
plt.xlabel("Epochs")
plt.ylabel(key)
plt.legend([key, 'val_'+key])
plt.show()
# Plot the history
plot_history(history, 'mean_squared_logarithmic_error')
Everything is fine up to this point. When I run the following line of code...
x_test['prediction'] = model.predict(x_test_scaled)
I get this error...
ValueError Traceback (most recent call last)
~\AppData\Local\Temp\1\ipykernel_23656\555877386.py in <module>
6 #df.head()
7
----> 8 x_test['prediction'] = model.predict(x_test_scaled)
~\Anaconda3\lib\site-packages\keras\utils\traceback_utils.py in error_handler(*args, **kwargs)
68 # To get the full stack trace, call:
69 # `tf.debugging.disable_traceback_filtering()`
---> 70 raise e.with_traceback(filtered_tb) from None
71 finally:
72 del filtered_tb
~\Anaconda3\lib\site-packages\keras\engine\training.py in tf__predict_function(iterator)
13 try:
14 do_return = True
---> 15 retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
16 except:
17 do_return = False
ValueError: in user code:
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\engine\training.py", line 2137, in predict_function *
return step_function(self, iterator)
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\engine\training.py", line 2123, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\engine\training.py", line 2111, in run_step **
outputs = model.predict_step(data)
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\engine\training.py", line 2079, in predict_step
return self(x, training=False)
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\utils\traceback_utils.py", line 70, in error_handler
raise e.with_traceback(filtered_tb) from None
File "C:\Users\RS\Anaconda3\lib\site-packages\keras\engine\input_spec.py", line 277, in assert_input_compatibility
raise ValueError(
ValueError: Exception encountered when calling layer 'sequential_23' (type Sequential).
Input 0 of layer "dense_44" is incompatible with the layer: expected axis -1 of input shape to have value 18, but received input with shape (None, 1)
Call arguments received by layer 'sequential_23' (type Sequential):
• inputs=tf.Tensor(shape=(None, 1), dtype=float32)
• training=False
• mask=None
I am trying to follow the example from this link.
https://www.analyticsvidhya.com/blog/2021/08/a-walk-through-of-regression-analysis-using-artificial-neural-networks-in-tensorflow/?

ValueError: Input 0 of layer "sequential" is incompatible with the layer: expected shape=(None, 81), found shape=(None, 77)

I am trying to train a neural network but I am getting the following error:
ValueError: Input 0 of layer "sequential" is incompatible with the layer: expected shape=(None, 81), found shape=(None, 77)
I tried to find the solution to this but am unable to do so. Can someone please help me?
Here Is the code of the same
from sklearn.preprocessing import StandardScaler
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout
from tensorflow.keras.wrappers.scikit_learn import KerasRegressor
from tensorflow.keras.callbacks import EarlyStopping
# Scaling the data
ss = StandardScaler()
X_train_sc = ss.fit_transform(X_train)
X_test_sc = ss.transform(X_test)
# Creating our model's structure
model = Sequential()
model.add(Dense(64, activation='relu', input_shape=(81,)))
model.add(Dropout(0.18))
model.add(Dense(32, activation='relu'))
model.add(Dropout(0.15))
model.add(Dense(1, activation='sigmoid'))
es = EarlyStopping(monitor='val_loss', patience=5)
# Compiling the model
model.compile(loss='bce',
optimizer='adam',
metrics=['binary_accuracy'])
# Fitting the model
history = model.fit(X_train_sc,
y_train,
batch_size = 256,
validation_data =(X_test_sc, y_test),
epochs = 500,
verbose = 0,
callbacks=[es])
As suggested I have edited the code to:
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout
from tensorflow.keras.wrappers.scikit_learn import KerasRegressor
from tensorflow.keras.callbacks import EarlyStopping
import tensorflow as tf
samples = 500
X_train_sc = tf.random.normal((samples, 81))
y_train = tf.random.uniform((samples, ), maxval=2, dtype=tf.int32)
# Creating our model's structure
model = Sequential()
model.add(Dense(64, activation='relu', input_shape=(81,)))
model.add(Dropout(0.18))
model.add(Dense(32, activation='relu'))
model.add(Dropout(0.15))
model.add(Dense(1, activation='sigmoid'))
es = EarlyStopping(monitor='val_loss', patience=5)
# Compiling the model
model.compile(loss='bce',
optimizer='adam',
metrics=['binary_accuracy'])
# Fitting the model
history = model.fit(X_train_sc,
y_train,
batch_size = 32,
epochs = 2,
verbose = 0)
But when I try to find the accuracy I get the same error as shown below:
# Scoring
train_score = model.evaluate(X_train_sc,
y_train,
verbose=1)
test_score = model.evaluate(X_test_sc,
y_test,
verbose=1)
labels = model.metrics_names
print('')
print(f'Training Accuracy: {train_score[1]}')
print(f'Testing Accuracy: {test_score[1]}')
16/16 [==============================] - 0s 2ms/step - loss: 0.6613 - binary_accuracy: 0.6040
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_7572/1082894889.py in <module>
3 y_train,
4 verbose=1)
----> 5 test_score = model.evaluate(X_test_sc,
6 y_test,
7 verbose=1)
~\Anaconda3\lib\site-packages\keras\utils\traceback_utils.py in error_handler(*args, **kwargs)
65 except Exception as e: # pylint: disable=broad-except
66 filtered_tb = _process_traceback_frames(e.__traceback__)
---> 67 raise e.with_traceback(filtered_tb) from None
68 finally:
69 del filtered_tb
~\Anaconda3\lib\site-packages\tensorflow\python\framework\func_graph.py in autograph_handler(*args, **kwargs)
1145 except Exception as e: # pylint:disable=broad-except
1146 if hasattr(e, "ag_error_metadata"):
-> 1147 raise e.ag_error_metadata.to_exception(e)
1148 else:
1149 raise
ValueError: in user code:
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\engine\training.py", line 1525, in test_function *
return step_function(self, iterator)
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\engine\training.py", line 1514, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\engine\training.py", line 1507, in run_step **
outputs = model.test_step(data)
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\engine\training.py", line 1471, in test_step
y_pred = self(x, training=False)
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler
raise e.with_traceback(filtered_tb) from None
File "C:\Users\sadik\Anaconda3\lib\site-packages\keras\engine\input_spec.py", line 264, in assert_input_compatibility
raise ValueError(f'Input {input_index} of layer "{layer_name}" is '
ValueError: Input 0 of layer "sequential_4" is incompatible with the layer: expected shape=(None, 81), found shape=(None, 77)

The problem is that your input data does not have the same shape that you defined in your first layer. Make sure the features dimension of your data corresponds to the input shape in the model's first layer. Here is an example:
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout
from tensorflow.keras.wrappers.scikit_learn import KerasRegressor
from tensorflow.keras.callbacks import EarlyStopping
import tensorflow as tf
samples = 500
# Create random dummy data
X_train_sc = tf.random.normal((samples, 81))
y_train = tf.random.uniform((samples, ), maxval=2, dtype=tf.int32)
X_test_sc = tf.random.normal((samples, 81))
y_test = tf.random.uniform((samples, ), maxval=2, dtype=tf.int32)
# Creating our model's structure
model = Sequential()
model.add(Dense(64, activation='relu', input_shape=(81,)))
model.add(Dropout(0.18))
model.add(Dense(32, activation='relu'))
model.add(Dropout(0.15))
model.add(Dense(1, activation='sigmoid'))
# Compiling the model
model.compile(loss='bce',
optimizer='adam',
metrics=['binary_accuracy'])
# Fitting the model
history = model.fit(X_train_sc,
y_train,
batch_size = 32,
epochs = 2,
verbose = 0)
So, if your feature dimension is 77 then change input_shape=(81,) to input_shape=(77,).

ValueError: Unknown loss function: custom_loss_function. Please ensure this object is passed to the `custom_objects` argument

I'm trying to train my model with this custom loss function:
1
where S(pn;ω) is the predicted value (y_pred) and MOSn is the target (y_true), so I wrote it this way:
import keras.backend as K
def custom_loss_function(y_true,y_pred):
for i in range(1,n+1):
l= K.abs(y_pred-y_true)
l = K.mean(l, axis=-1)
return l
then I built my model:
#Model definition
from keras import models
from keras import layers
def build_model():
model = models.Sequential()
model.add(layers.Conv2D(32, (5, 5), activation='relu', input_shape=(32, 32, 1)))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(32, (5, 5), activation='relu'))
model.add(layers.MaxPooling2D((10, 10)))
model.add(layers.Dense(250, activation='relu'))
model.add(layers.Dense(250, activation='relu'))
model.add(layers.Dense(1))
model.compile(optimizer='rmsprop', loss='custom_loss_function', metrics=['mae'])
return model
model = build_model()
but when I run the training process:
num_epochs = 20
history = model.fit(train_data, train_labels, epochs=num_epochs, batch_size=None, verbose=0)
I get this error:
ValueError Traceback (most recent call last)
<ipython-input-16-eb44c7fa4ec7> in <module>()
1 num_epochs = 20
----> 2 history = model.fit(train_data, train_labels, epochs=num_epochs, batch_size=None, verbose=0)
9 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
992 except Exception as e: # pylint:disable=broad-except
993 if hasattr(e, "ag_error_metadata"):
--> 994 raise e.ag_error_metadata.to_exception(e)
995 else:
996 raise
ValueError: in user code:
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:853 train_function *
return step_function(self, iterator)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:842 step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:1286 run
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:2849 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:3632 _call_for_each_replica
return fn(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:835 run_step **
outputs = model.train_step(data)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:789 train_step
y, y_pred, sample_weight, regularization_losses=self.losses)
/usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:184 __call__
self.build(y_pred)
/usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:133 build
self._losses = tf.nest.map_structure(self._get_loss_object, self._losses)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/nest.py:869 map_structure
structure[0], [func(*x) for x in entries],
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/nest.py:869 <listcomp>
structure[0], [func(*x) for x in entries],
/usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:273 _get_loss_object
loss = losses_mod.get(loss)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:2136 get
return deserialize(identifier)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:2095 deserialize
printable_module_name='loss function')
/usr/local/lib/python3.7/dist-packages/keras/utils/generic_utils.py:709 deserialize_keras_object
.format(printable_module_name, object_name))
ValueError: Unknown loss function: custom_loss_function. Please ensure this object is passed to the `custom_objects` argument. See https://www.tensorflow.org/guide/keras/save_and_serialize#registering_the_custom_object for details.
I read the details about the 'custom object' argument and tried to apply it but still can't figure it out, How exactly can I pass my custom function to the 'custom_objects' argument ?

In python, you can pass a function as a self-contained object, or Callable. In this case, you can pass the argument loss=custom_loss_function without the single-quotes. Here's more info about callables and custom loss functions in Tensorflow.

Python giving errors. How to solve it?

Here's the code:
import numpy as np
np.random.seed(123) #for reproducibility
from keras.models import sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D
from keras.utils import np_utils
from keras.datasets import mnist
(X_train, y_train), (X_test, y_test) = mnist.load_data()
X_train = X_train.reshape(X_train.shape[0], 1, 28, 28)
X_test = X_test.reshape(X_test.shape[0], 1, 28, 28)
X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255
y_train = np_utils.to_categorical(y_train, 10)
y_test = np_utils.to_categorical(y_test, 10)
model = Sequential()
model.add(Convolution2D(32, 3, 3, activation='relu', input_shape=(28,28,1)))
model.add(Convolution2D(32, 3, 3, activation='relu'))
model.add(MaxPooling2D(pool_size=(2,2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
model.compile(loss='categorical_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.fit(X_train, Y_train,
batch_size=32, epochs=10, verbose=1)
score = model.evaluate(X_test, Y_test, verbose=0)
Here's the error:
Epoch 1/10
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-71-db342978fb82> in <module>()
37
38 model.fit(X_train, Y_train,
---> 39 batch_size=32, epochs=10, verbose=1)
40 score = model.evaluate(X_test, Y_test, verbose=0)
9 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
992 except Exception as e: # pylint:disable=broad-except
993 if hasattr(e, "ag_error_metadata"):
--> 994 raise e.ag_error_metadata.to_exception(e)
995 else:
996 raise
ValueError: in user code:
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:853 train_function *
return step_function(self, iterator)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:842 step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:1286 run
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:2849 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/distribute/distribute_lib.py:3632 _call_for_each_replica
return fn(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:835 run_step **
outputs = model.train_step(data)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:787 train_step
y_pred = self(x, training=True)
/usr/local/lib/python3.7/dist-packages/keras/engine/base_layer.py:1020 __call__
input_spec.assert_input_compatibility(self.input_spec, inputs, self.name)
/usr/local/lib/python3.7/dist-packages/keras/engine/input_spec.py:254 assert_input_compatibility
' but received input with shape ' + display_shape(x.shape))
ValueError: Input 0 of layer sequential_29 is incompatible with the layer: expected axis -1 of input shape to have value 1 but received input with shape (32, 1, 28, 28)

Change
X_train = X_train.reshape(X_train.shape[0], 1, 28, 28)
X_test = X_test.reshape(X_test.shape[0], 1, 28, 28)
To,
X_train = X_train.reshape(X_train.shape[0], 28, 28, 1)
X_test = X_test.reshape(X_test.shape[0], 28, 28, 1)

ValueError: Input 0 of layer sequential_37 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 15]

I've made all the attempts I knew.
Also all combinations of input_dim = 15 already. If anyone can help me?
print(x_train.shape)
print(y_train.shape)
print(x_test.shape)
print(y_test.shape)
(233941, 15)
(233941, 1)
(100261, 15)
(100261,)
I've already done the test using input_dim = (233941, 15) and input_dim = (233941, 1). But I still can't find the problem.
Can my problem be in the dataset division?
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Dense, Dropout, LSTM
model = Sequential()
model.add(LSTM(100, input_dim=15, return_sequences=True))
model.add(Dropout(0.3))
model.add(LSTM(50, return_sequences = True))
model.add(Dropout(0.3))
#3 camada
model.add(LSTM(50, return_sequences = True))
model.add(Dropout(0.3))
model.add(LSTM(units = 50))
model.add(Dropout(0.3))
model.add(Dense(1, activation='sigmoid'))
# Compile model
model.compile(optimizer = 'adam', loss = 'mean_squared_error',
metrics = ['mean_absolute_error'])
# Fit the model
model.fit(x_train,y_train,epochs=100, validation_data=(x_test,y_test))
Epoch 1/100
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-79-2e6c4d489e38> in <module>()
21 metrics = ['mean_absolute_error'])
22 # Fit the model
---> 23 model.fit(x_train,y_train,epochs=100, validation_data=(x_test,y_test))
10 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
971 except Exception as e: # pylint:disable=broad-except
972 if hasattr(e, "ag_error_metadata"):
--> 973 raise e.ag_error_metadata.to_exception(e)
974 else:
975 raise
ValueError: in user code:
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:806 train_function *
return step_function(self, iterator)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:796 step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1211 run
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2585 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2945 _call_for_each_replica
return fn(*args, **kwargs)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:789 run_step **
outputs = model.train_step(data)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:747 train_step
y_pred = self(x, training=True)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/base_layer.py:976 __call__
self.name)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/input_spec.py:180 assert_input_compatibility
str(x.shape.as_list()))
ValueError: Input 0 of layer sequential_37 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 15]

As Marco Certliani mentioned in the comments you need to format your input properly for a RNN, which as the error you got mentioned is 3 dimensional.
Here is a representation of what the input tensor should look like:
Meaning that your 3D tensor will have a shape of (batch_size, time_step, input_dimension).