MLP with TensorFlow Input Shape not Matching Output Shape - python-3.x

Question

I'm trying to train a multilayer perceptron based on the Iris dataset using TensorFlow in Pycharm with Jupyter Notebook. Every time I run my code it fails on the model.fit() function and gives the following error.
ValueError: Shapes (None, 1) and (None, 3) are incompatible.
I've tried playing around with different values for the hyperparameters but there's obviously something I'm not getting. Any help/pointers that anyone could provide would be much appreciated.
Here's my data setup and preprocessing:
import pandas as pd
import numpy as np
# read iris data into pandas dataframe
iris = pd.read_csv("data/IRIS.csv", header=0)
# apply label to index column
iris.index.name = "id"
# create copy of iris dataframe in which to store normalised values and keep original dataframe for comparison later on
iris_unnormalized = iris
iris_normalized = iris.copy()
# isolate columns with numerical values
iris_num = iris.select_dtypes(include=[np.number])
# find max value in each column
col_maxes = iris_num.max()
# find overall max value among all columns
iris_num_max = col_maxes.max()
# divide all numerical values by overall max value in order to normalize data to a value between 0 and 1
iris_num_norm = iris_num / iris_num_max
# reassign normalised values back to their corresponding columns
iris_normalized[iris_num_norm.columns] = iris_num_norm
# specify seed for reproducibility
np.random.seed(1671)
training = iris_normalized.sample(frac = 0.8)
test = iris_normalized.drop(training.index)
# initialize the training input and output list
# same for testing set
X_train = []
Y_train = []
X_test = []
Y_test = []
# loop through the dataframe and separate inputs and outputs for training and testing
for index, row in training.iterrows():
X_train.append([row['sepal length cm'], row['sepal width cm'], row['petal length cm'], row['petal width cm']])
Y_train.append([row['species']])
for index, row in test.iterrows():
X_test.append([row['sepal length cm'], row['sepal width cm'], row['petal length cm'], row['petal width cm']])
Y_test.append([row['species']])
X_train = np.array(X_train).astype('float32')
Y_train = np.array(Y_train)
X_test = np.array(X_test).astype('float32')
Y_test = np.array(Y_test)
print(X_train.shape, "training samples") # Output: (120, 4) training samples
print(X_test.shape, "test samples") # Output: (30, 4) test samples
Here's where I try to create the neural network:
import tensorflow as tf
from tensorflow import keras
NB_CLASSES = 3 # number of iris varieties
N_HIDDEN = 128
BATCH_SIZE = 10
VERBOSE = 1
VALIDATION_SPLIT = 0.2 # how much of training set to hold for validation
EPOCHS = 200
model = tf.keras.models.Sequential(
[
keras.layers.Dense(N_HIDDEN, input_shape=(10,4,), batch_size=BATCH_SIZE, name="dense_layer1", activation="relu"),
keras.layers.Dense(N_HIDDEN, input_shape=(4,), batch_size=BATCH_SIZE, name="dense_layer2", activation="relu"),
keras.layers.Dense(NB_CLASSES, input_shape=(4,), batch_size=BATCH_SIZE, name="dense_layer3", activation="softmax"),
]
)
model.summary()
################### model summary output: #####################
Layer (type) Output Shape Param #
=================================================================
dense_layer1 (Dense) (10, 10, 128) 640
_________________________________________________________________
dense_layer2 (Dense) (10, 10, 128) 16512
_________________________________________________________________
dense_layer3 (Dense) (10, 10, 3) 387
=================================================================
Total params: 17,539
Trainable params: 17,539
Non-trainable params: 0
# compiling the model
model.compile(optimizer='SGD', loss='categorical_crossentropy', metrics=['accuracy'])
#train the model
model.fit(X_train, Y_train,
batch_size=BATCH_SIZE,
epochs = EPOCHS,
verbose = VERBOSE,
validation_split = VALIDATION_SPLIT)
The data setup and normalization work fine but when I run the code to create the neural network I get the below error:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_8888/2571387568.py in <module>
38
39 #train the model
---> 40 model.fit(X_train, Y_train,
41 batch_size=BATCH_SIZE,
42 epochs = EPOCHS,
\venv\lib\site-packages\keras\engine\training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
1182 _r=1):
1183 callbacks.on_train_batch_begin(step)
-> 1184 tmp_logs = self.train_function(iterator)
1185 if data_handler.should_sync:
1186 context.async_wait()
\venv\lib\site-packages\tensorflow\python\eager\def_function.py in __call__(self, *args, **kwds)
883
884 with OptionalXlaContext(self._jit_compile):
--> 885 result = self._call(*args, **kwds)
886
887 new_tracing_count = self.experimental_get_tracing_count()
\venv\lib\site-packages\tensorflow\python\eager\def_function.py in _call(self, *args, **kwds)
931 # This is the first call of __call__, so we have to initialize.
932 initializers = []
--> 933 self._initialize(args, kwds, add_initializers_to=initializers)
934 finally:
935 # At this point we know that the initialization is complete (or less
\venv\lib\site-packages\tensorflow\python\eager\def_function.py in _initialize(self, args, kwds, add_initializers_to)
757 self._graph_deleter = FunctionDeleter(self._lifted_initializer_graph)
758 self._concrete_stateful_fn = (
--> 759 self._stateful_fn._get_concrete_function_internal_garbage_collected( # pylint: disable=protected-access
760 *args, **kwds))
761
\venv\lib\site-packages\tensorflow\python\eager\function.py in _get_concrete_function_internal_garbage_collected(self, *args, **kwargs)
3064 args, kwargs = None, None
3065 with self._lock:
-> 3066 graph_function, _ = self._maybe_define_function(args, kwargs)
3067 return graph_function
3068
\venv\lib\site-packages\tensorflow\python\eager\function.py in _maybe_define_function(self, args, kwargs)
3461
3462 self._function_cache.missed.add(call_context_key)
-> 3463 graph_function = self._create_graph_function(args, kwargs)
3464 self._function_cache.primary[cache_key] = graph_function
3465
\venv\lib\site-packages\tensorflow\python\eager\function.py in _create_graph_function(self, args, kwargs, override_flat_arg_shapes)
3296 arg_names = base_arg_names + missing_arg_names
3297 graph_function = ConcreteFunction(
-> 3298 func_graph_module.func_graph_from_py_func(
3299 self._name,
3300 self._python_function,
\venv\lib\site-packages\tensorflow\python\framework\func_graph.py in func_graph_from_py_func(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes, acd_record_initial_resource_uses)
1005 _, original_func = tf_decorator.unwrap(python_func)
1006
-> 1007 func_outputs = python_func(*func_args, **func_kwargs)
1008
1009 # invariant: `func_outputs` contains only Tensors, CompositeTensors,
\venv\lib\site-packages\tensorflow\python\eager\def_function.py in wrapped_fn(*args, **kwds)
666 # the function a weak reference to itself to avoid a reference cycle.
667 with OptionalXlaContext(compile_with_xla):
--> 668 out = weak_wrapped_fn().__wrapped__(*args, **kwds)
669 return out
670
\venv\lib\site-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:
\venv\lib\site-packages\keras\engine\training.py:853 train_function *
return step_function(self, iterator)
\venv\lib\site-packages\keras\engine\training.py:842 step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
\venv\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:1286 run
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
\venv\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:2849 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
\venv\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:3632 _call_for_each_replica
return fn(*args, **kwargs)
\venv\lib\site-packages\keras\engine\training.py:835 run_step **
outputs = model.train_step(data)
\venv\lib\site-packages\keras\engine\training.py:788 train_step
loss = self.compiled_loss(
\venv\lib\site-packages\keras\engine\compile_utils.py:201 __call__
loss_value = loss_obj(y_t, y_p, sample_weight=sw)
\venv\lib\site-packages\keras\losses.py:141 __call__
losses = call_fn(y_true, y_pred)
\venv\lib\site-packages\keras\losses.py:245 call **
return ag_fn(y_true, y_pred, **self._fn_kwargs)
\venv\lib\site-packages\tensorflow\python\util\dispatch.py:206 wrapper
return target(*args, **kwargs)
\venv\lib\site-packages\keras\losses.py:1665 categorical_crossentropy
return backend.categorical_crossentropy(
\venv\lib\site-packages\tensorflow\python\util\dispatch.py:206 wrapper
return target(*args, **kwargs)
\venv\lib\site-packages\keras\backend.py:4839 categorical_crossentropy
target.shape.assert_is_compatible_with(output.shape)
\venv\lib\site-packages\tensorflow\python\framework\tensor_shape.py:1161 assert_is_compatible_with
raise ValueError("Shapes %s and %s are incompatible" % (self, other))
ValueError: Shapes (None, 1) and (None, 3) are incompatible