我一直在尝试使用TensorFlow通过U-Net进行图像分割
代码：
inputs = tf.keras.layers.Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS))
s = tf.keras.layers.Lambda(lambda x: x / 255)(inputs)
c1 = tf.keras.layers.Conv2D(16, (3, 3), activation='relu', kernel_initializer='he_normal', padding='same')(s)
c1 = tf.keras.layers.Dropout(0.1)(c1)
c1 = tf.keras.layers.Conv2D(16, (3, 3), activation='relu', kernel_initializer='he_normal', padding='same')(c1)
p1 = tf.keras.layers.MaxPooling2D((2, 2))(c1)
我也添加了更多的层。但是我认为问题出在输入层。
checkpointer = tf.keras.callbacks.ModelCheckpoint('unet_test.h5', verbose=1, save_best_only=True)
callbacks = [
tf.keras.callbacks.EarlyStopping(patience=2, monitor='val_loss'),
tf.keras.callbacks.TensorBoard(log_dir='logs')]
results = model.fit(X_train, Y_train, validation_split=0.1, batch_size=16, epochs=25, callbacks=callbacks)
当我尝试适应时，这是错误消息：
Epoch 1/25
WARNING:tensorflow:Model was constructed with shape (None, 128, 128, 3) for input Tensor("input_23:0", shape=(None, 128, 128, 3), dtype=float32), but it was called on an input with incompatible shape (None, 128, 3).
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-55-180608a30979> in <module>
3
4
----> 5 results = model.fit(X_train, Y_train, validation_split=0.1, batch_size=16, epochs=25)
~/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py in _method_wrapper(self, *args, **kwargs)
64 def _method_wrapper(self, *args, **kwargs):
65 if not self._in_multi_worker_mode(): # pylint: disable=protected-access
---> 66 return method(self, *args, **kwargs)
67
68 # Running inside `run_distribute_coordinator` already.
~/.local/lib/python3.8/site-packages/tensorflow/python/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)
846 batch_size=batch_size):
847 callbacks.on_train_batch_begin(step)
--> 848 tmp_logs = train_function(iterator)
849 # Catch OutOfRangeError for Datasets of unknown size.
850 # This blocks until the batch has finished executing.
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py in __call__(self, *args, **kwds)
578 xla_context.Exit()
579 else:
--> 580 result = self._call(*args, **kwds)
581
582 if tracing_count == self._get_tracing_count():
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py in _call(self, *args, **kwds)
625 # This is the first call of __call__, so we have to initialize.
626 initializers = []
--> 627 self._initialize(args, kwds, add_initializers_to=initializers)
628 finally:
629 # At this point we know that the initialization is complete (or less
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py in _initialize(self, args, kwds, add_initializers_to)
503 self._graph_deleter = FunctionDeleter(self._lifted_initializer_graph)
504 self._concrete_stateful_fn = (
--> 505 self._stateful_fn._get_concrete_function_internal_garbage_collected( # pylint: disable=protected-access
506 *args, **kwds))
507
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/function.py in _get_concrete_function_internal_garbage_collected(self, *args, **kwargs)
2444 args, kwargs = None, None
2445 with self._lock:
-> 2446 graph_function, _, _ = self._maybe_define_function(args, kwargs)
2447 return graph_function
2448
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/function.py in _maybe_define_function(self, args, kwargs)
2775
2776 self._function_cache.missed.add(call_context_key)
-> 2777 graph_function = self._create_graph_function(args, kwargs)
2778 self._function_cache.primary[cache_key] = graph_function
2779 return graph_function, args, kwargs
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/function.py in _create_graph_function(self, args, kwargs, override_flat_arg_shapes)
2655 arg_names = base_arg_names + missing_arg_names
2656 graph_function = ConcreteFunction(
-> 2657 func_graph_module.func_graph_from_py_func(
2658 self._name,
2659 self._python_function,
~/.local/lib/python3.8/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)
979 _, original_func = tf_decorator.unwrap(python_func)
980
--> 981 func_outputs = python_func(*func_args, **func_kwargs)
982
983 # invariant: `func_outputs` contains only Tensors, CompositeTensors,
~/.local/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py in wrapped_fn(*args, **kwds)
439 # __wrapped__ allows AutoGraph to swap in a converted function. We give
440 # the function a weak reference to itself to avoid a reference cycle.
--> 441 return weak_wrapped_fn().__wrapped__(*args, **kwds)
442 weak_wrapped_fn = weakref.ref(wrapped_fn)
443
~/.local/lib/python3.8/site-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
966 except Exception as e: # pylint:disable=broad-except
967 if hasattr(e, "ag_error_metadata"):
--> 968 raise e.ag_error_metadata.to_exception(e)
969 else:
970 raise
ValueError: in user code:
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py:571 train_function *
outputs = self.distribute_strategy.run(
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/distribute/distribute_lib.py:951 run **
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/distribute/distribute_lib.py:2290 call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/distribute/distribute_lib.py:2649 _call_for_each_replica
return fn(*args, **kwargs)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/training.py:531 train_step **
y_pred = self(x, training=True)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py:927 __call__
outputs = call_fn(cast_inputs, *args, **kwargs)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/network.py:717 call
return self._run_internal_graph(
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/network.py:888 _run_internal_graph
output_tensors = layer(computed_tensors, **kwargs)
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py:885 __call__
input_spec.assert_input_compatibility(self.input_spec, inputs,
/home/gokul/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/input_spec.py:176 assert_input_compatibility
raise ValueError('Input ' + str(input_index) + ' of layer ' +
ValueError: Input 0 of layer conv2d_176 is incompatible with the layer: expected ndim=4, found ndim=3. Full shape received: [None, 128, 3]
这是模型摘要：
Model: "model_5"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_23 (InputLayer) [(None, 128, 128, 3) 0
__________________________________________________________________________________________________
lambda_22 (Lambda) (None, 128, 128, 3) 0 input_23[0][0]
__________________________________________________________________________________________________
conv2d_176 (Conv2D) (None, 128, 128, 16) 448 lambda_22[0][0]
__________________________________________________________________________________________________
dropout_86 (Dropout) (None, 128, 128, 16) 0 conv2d_176[0][0]
__________________________________________________________________________________________________
conv2d_177 (Conv2D) (None, 128, 128, 16) 2320 dropout_86[0][0]
__________________________________________________________________________________________________
max_pooling2d_46 (MaxPooling2D) (None, 64, 64, 16) 0 conv2d_177[0][0]
__________________________________________________________________________________________________
conv2d_178 (Conv2D) (None, 64, 64, 32) 4640 max_pooling2d_46[0][0]
__________________________________________________________________________________________________
dropout_87 (Dropout) (None, 64, 64, 32) 0 conv2d_178[0][0]
__________________________________________________________________________________________________
conv2d_179 (Conv2D) (None, 64, 64, 32) 9248 dropout_87[0][0]
__________________________________________________________________________________________________
max_pooling2d_47 (MaxPooling2D) (None, 32, 32, 32) 0 conv2d_179[0][0]
__________________________________________________________________________________________________
conv2d_180 (Conv2D) (None, 32, 32, 64) 18496 max_pooling2d_47[0][0]
__________________________________________________________________________________________________
dropout_88 (Dropout) (None, 32, 32, 64) 0 conv2d_180[0][0]
__________________________________________________________________________________________________
conv2d_181 (Conv2D) (None, 32, 32, 64) 36928 dropout_88[0][0]
__________________________________________________________________________________________________
max_pooling2d_48 (MaxPooling2D) (None, 16, 16, 64) 0 conv2d_181[0][0]
__________________________________________________________________________________________________
conv2d_182 (Conv2D) (None, 16, 16, 128) 73856 max_pooling2d_48[0][0]
__________________________________________________________________________________________________
dropout_89 (Dropout) (None, 16, 16, 128) 0 conv2d_182[0][0]
__________________________________________________________________________________________________
conv2d_183 (Conv2D) (None, 16, 16, 128) 147584 dropout_89[0][0]
__________________________________________________________________________________________________
max_pooling2d_49 (MaxPooling2D) (None, 8, 8, 128) 0 conv2d_183[0][0]
__________________________________________________________________________________________________
conv2d_184 (Conv2D) (None, 8, 8, 256) 295168 max_pooling2d_49[0][0]
__________________________________________________________________________________________________
dropout_90 (Dropout) (None, 8, 8, 256) 0 conv2d_184[0][0]
__________________________________________________________________________________________________
conv2d_185 (Conv2D) (None, 8, 8, 256) 590080 dropout_90[0][0]
__________________________________________________________________________________________________
conv2d_transpose_22 (Conv2DTran (None, 16, 16, 128) 131200 conv2d_185[0][0]
__________________________________________________________________________________________________
concatenate_22 (Concatenate) (None, 16, 16, 256) 0 conv2d_transpose_22[0][0]
conv2d_183[0][0]
__________________________________________________________________________________________________
conv2d_186 (Conv2D) (None, 16, 16, 128) 295040 concatenate_22[0][0]
__________________________________________________________________________________________________
dropout_91 (Dropout) (None, 16, 16, 128) 0 conv2d_186[0][0]
__________________________________________________________________________________________________
conv2d_187 (Conv2D) (None, 16, 16, 128) 147584 dropout_91[0][0]
__________________________________________________________________________________________________
conv2d_transpose_23 (Conv2DTran (None, 32, 32, 64) 32832 conv2d_187[0][0]
__________________________________________________________________________________________________
concatenate_23 (Concatenate) (None, 32, 32, 128) 0 conv2d_transpose_23[0][0]
conv2d_181[0][0]
__________________________________________________________________________________________________
conv2d_188 (Conv2D) (None, 32, 32, 64) 73792 concatenate_23[0][0]
__________________________________________________________________________________________________
dropout_92 (Dropout) (None, 32, 32, 64) 0 conv2d_188[0][0]
__________________________________________________________________________________________________
conv2d_189 (Conv2D) (None, 32, 32, 64) 36928 dropout_92[0][0]
__________________________________________________________________________________________________
conv2d_transpose_24 (Conv2DTran (None, 64, 64, 32) 8224 conv2d_189[0][0]
__________________________________________________________________________________________________
concatenate_24 (Concatenate) (None, 64, 64, 64) 0 conv2d_transpose_24[0][0]
conv2d_179[0][0]
__________________________________________________________________________________________________
conv2d_190 (Conv2D) (None, 64, 64, 32) 18464 concatenate_24[0][0]
__________________________________________________________________________________________________
dropout_93 (Dropout) (None, 64, 64, 32) 0 conv2d_190[0][0]
__________________________________________________________________________________________________
conv2d_191 (Conv2D) (None, 64, 64, 32) 9248 dropout_93[0][0]
__________________________________________________________________________________________________
conv2d_transpose_25 (Conv2DTran (None, 128, 128, 16) 2064 conv2d_191[0][0]
__________________________________________________________________________________________________
concatenate_25 (Concatenate) (None, 128, 128, 32) 0 conv2d_transpose_25[0][0]
conv2d_177[0][0]
__________________________________________________________________________________________________
conv2d_192 (Conv2D) (None, 128, 128, 16) 4624 concatenate_25[0][0]
__________________________________________________________________________________________________
dropout_94 (Dropout) (None, 128, 128, 16) 0 conv2d_192[0][0]
__________________________________________________________________________________________________
conv2d_193 (Conv2D) (None, 128, 128, 16) 2320 dropout_94[0][0]
__________________________________________________________________________________________________
conv2d_194 (Conv2D) (None, 128, 128, 1) 17 conv2d_193[0][0]
==================================================================================================
Total params: 1,941,105
Trainable params: 1,941,105
有人可以在这里帮助我吗？在过去的几个小时里似乎一直被困在这里：（