Tensorflow ValueError:形状（无，无）和（无，28，28，12）不兼容
İ我正在处理一个分类为12类的图像数据集。İam使用VGG16的迁移学习。然而，İ遇到了一个错误：形状（无，无）和（无，28，28，12）不兼容。 我的代码： 错误：Tensorflow ValueError:形状（无，无）和（无，28，28，12）不兼容,tensorflow,machine-learning,keras,deep-learning,transfer-learning,Tensorflow,Machine Learning,Keras,Deep Learning,Transfer Learning,İ我正在处理一个分类为12类的图像数据集。İam使用VGG16的迁移学习。然而，İ遇到了一个错误：形状（无，无）和（无，28，28，12）不兼容。 我的代码： 错误：ValueError:形状（无，无）和（无，28，28，12）不兼容[！[在此处输入图像描述] 错误详细信息： ValueError Traceback (most recent call last) <ipython-input-39-938295cc34c4&g
ValueError:形状（无，无）和（无，28，28，12）不兼容ValueError Traceback (most recent call last)
<ipython-input-39-938295cc34c4> in <module>()
2 checkpoint = ModelCheckpoint("vgg16_1.h5", monitor='val_acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1)
3 early = EarlyStopping(monitor='val_acc', min_delta=0, patience=40, verbose=1, mode='auto')
----> 4 model_final.fit_generator(generator= train_images , steps_per_epoch= 2, epochs= 100, validation_data= val_images , validation_steps=1, callbacks=[checkpoint,early])
5 model_final.save_weights("vgg16_1.h5")
10 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
984 except Exception as e: # pylint:disable=broad-except
985 if hasattr(e, "ag_error_metadata"):
--> 986 raise e.ag_error_metadata.to_exception(e)
987 else:
988 raise
ValueError: in user code:
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:830 train_function *
return step_function(self, iterator)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:813 run_step *
outputs = model.train_step(data)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:771 train_step *
loss = self.compiled_loss(
/usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:201 __call__ *
loss_value = loss_obj(y_t, y_p, sample_weight=sw)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:142 __call__ *
losses = call_fn(y_true, y_pred)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:246 call *
return ag_fn(y_true, y_pred, **self._fn_kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper **
return target(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:1631 categorical_crossentropy
y_true, y_pred, from_logits=from_logits)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper
return target(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/backend.py:4827 categorical_crossentropy
target.shape.assert_is_compatible_with(output.shape)
/usr/local/lib/python3.7/dist-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, None) and (None, 28, 28, 12) are incompatible
valueError Traceback (most recent call last)
<ipython-input-56-5210d7f2da32> in <module>()
2 checkpoint = ModelCheckpoint("vgg16_1.h5", monitor='val_acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1)
3 early = EarlyStopping(monitor='val_acc', min_delta=0, patience=40, verbose=1, mode='auto')
----> 4 model_final.fit_generator(generator= train_images, steps_per_epoch= 2, epochs= 100, validation_data= val_images, validation_steps=1, callbacks=[checkpoint,early])
5 model_final.save_weights("vgg16_1.h5")
10 frames
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
984 except Exception as e: # pylint:disable=broad-except
985 if hasattr(e, "ag_error_metadata"):
--> 986 raise e.ag_error_metadata.to_exception(e)
987 else:
988 raise
ValueError: in user code:
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:830 train_function *
return step_function(self, iterator)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:813 run_step *
outputs = model.train_step(data)
/usr/local/lib/python3.7/dist-packages/keras/engine/training.py:771 train_step *
loss = self.compiled_loss(
/usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:201 __call__ *
loss_value = loss_obj(y_t, y_p, sample_weight=sw)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:142 __call__ *
losses = call_fn(y_true, y_pred)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:246 call *
return ag_fn(y_true, y_pred, **self._fn_kwargs)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper **
return target(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/losses.py:1631 categorical_crossentropy
y_true, y_pred, from_logits=from_logits)
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper
return target(*args, **kwargs)
/usr/local/lib/python3.7/dist-packages/keras/backend.py:4827 categorical_crossentropy
target.shape.assert_is_compatible_with(output.shape)
/usr/local/lib/python3.7/dist-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, None) and (None, 28, 28, 12) are incompatible
valueError回溯（最近一次调用）
在（）
2检查点=模型检查点（“vgg16_1.h5”，监视器=”val_acc'，verbose=1，save_best_only=True，save_weights_only=False，mode='auto'，period=1）
3早到=早到（监视器='val_acc'，最小增量=0，耐心=40，冗余=1，模式='auto'）
---->4模型最终拟合生成器（生成器=训练映像，每个历元的步骤=2，历元=100，验证数据=val映像，验证步骤=1，回调=[检查点，早期]）
5型号最终保存重量（“vgg16型号1.h5”）
10帧
/包装器中的usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py（*args，**kwargs）
984例外情况为e:#pylint:disable=broad Exception
985如果hasattr（e，“ag错误元数据”）：
-->986将e.ag\u错误\u元数据引发到\u异常（e）
987其他：
988提高
ValueError:在用户代码中：
/usr/local/lib/python3.7/dist包/keras/engine/training.py:830列车功能*
返回步骤_函数（self、迭代器）
/usr/local/lib/python3.7/dist包/keras/engine/training.py:813运行步骤*
输出=型号列车步进（数据）
/usr/local/lib/python3.7/dist包/keras/engine/training.py:771训练步骤*
损耗=自编的损耗(
/usr/local/lib/python3.7/dist-packages/keras/engine/compile\u-utils.py:201\u\u-call\u*
损耗值=损耗对象（y\u t，y\u p，样品重量=sw）
/usr/local/lib/python3.7/dist-packages/keras/loss.py:142调用*
损失=催缴股款fn（y_真，y_pred）
/usr/local/lib/python3.7/dist-packages/keras/loss.py:246调用*
返回ag_fn（y_true，y_pred，**self.\u fn_kwargs）
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206包装器**
返回目标（*args，**kwargs）
/usr/local/lib/python3.7/dist-packages/keras/loss.py:1631-categorical\u
y_true，y_pred，from_logits=from_logits）
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206包装器
返回目标（*args，**kwargs）
/usr/local/lib/python3.7/dist-packages/keras/backend.py:4827 category\u
target.shape.assert\u与（output.shape）兼容
/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/tensor\u-shape.py:1161-assert\u与
raise VALUERROR（“形状%s和%s不兼容”%（自身、其他））
ValueError:形状（无，无）和（无，28，28，12）不兼容
- 在使用生成器时，您使用的是字符串
而不是变量pathpath - 此外，列车路径、验证路径和测试路径也应不同
- 您尚未为VGG19型号指定输入张量
#train_dir_path is path to your training images
train_data = train_generator.flow_from_directory(directory=train_dir_path,target_size=IMAGE_SHAPE , color_mode="rgb" , class_mode='categorical', batch_size=BATCH_SIZE , shuffle = True )
#valid_dir_path is path to your validation images
valid_data = validation_generator.flow_from_directory(directory=valid_dir_path, target_size=IMAGE_SHAPE , color_mode="rgb" , class_mode='categorical' , batch_size=BATCH_SIZE , shuffle = True )
- VGG模型的输出在进入稠密层之前应该是平坦的
IMAGE\u SHAPE=（224224224）
批量大小=32
列车=图像数据生成器（）
train_generator=tf.keras.preprocessing.image.ImageDataGenerator（重缩放=1./255，填充模式=‘最近’）
#train_dir_path是指向训练图像的路径
列车数据=列车生成器。来自列车目录的流程（目录=列车目录路径，目标尺寸=图像形状，颜色模式=“rgb”，类模式=“分类”，批次尺寸=批次尺寸，随机播放=真）
valid=ImageDataGenerator（）
验证_生成器=tf.keras.preprocessing.image.ImageDataGenerator（重缩放=1./255）
#valid_dir_path是验证图像的路径
有效的\u数据=验证\u生成器。来自\u目录的流（目录=有效的\u目录\u路径，目标\u大小=图像\u形状，颜色\u mode=“rgb”，class\u mode='classifical'，批次\u大小=批次\u大小，随机播放=True）
测试=ImageDataGenerator（）
test_generator=tf.keras.preprocessing.image.ImageDataGenerator（重缩放=1./255）
#test_dir_path是指向测试映像的路径
test\u data=test\u generator.flow\u from\u directory（directory=test\u dir\u path，target\u size=IMAGE\u SHAPE，color\u mode=“rgb”，class\u mode='classifical'，batch\u size=1，shuffle=False）
测试数据重置（）
从keras.applications.vgg19导入vgg19
vggmodel=VGG19（weights='imagenet'，include_top=True，input_tensor=tensorflow.keras.layers.input（shape=（224224,3）））
对于（vggmodel.layers）[:32]中的层：
打印（层）
layers.trainable=False
X=vggmodel.layers[-12]。输出
X=tensorflow.keras.layers.Flatten（）（X）
预测=密集（12，activation=“softmax”）（X）
模型\最终=模型（vggmodel.input，预测）
模型_final.compile（优化器=优化器.Adam（lr=0.0003），loss='classifical_crossentropy'，metrics=[“准确性”]）
对于图像批次，在列数据中标记批次：
打印（图像\批处理形状）
打印（标签和批处理形状）
打破
从keras.callbacks导入模型检查点，EarlyStopping
检查点=模型检查点（“vgg16\u 1.h5”，监视器=”val\u acc'，详细=1，保存最佳值仅=真，保存权重仅=假，模式=”自动'，周期=1）
早期=早期停止（监视器='val_acc'，耐心=40，详细=1，模式='auto'）
模型最终拟合生成器（生成器=训练数据，每个历元的步骤=2，历元=100，验证数据=有效数据，验证步骤=1，回调=[检查点，早期]）
模型最终保存重量（“vgg16\U 1.h5”）
- 在使用生成器时，您使用的是字符串
而不是变量pathpath - 此外，列车路径、验证路径和测试路径也应不同
- 您没有指定
valueError Traceback (most recent call last) <ipython-input-56-5210d7f2da32> in <module>() 2 checkpoint = ModelCheckpoint("vgg16_1.h5", monitor='val_acc', verbose=1, save_best_only=True, save_weights_only=False, mode='auto', period=1) 3 early = EarlyStopping(monitor='val_acc', min_delta=0, patience=40, verbose=1, mode='auto') ----> 4 model_final.fit_generator(generator= train_images, steps_per_epoch= 2, epochs= 100, validation_data= val_images, validation_steps=1, callbacks=[checkpoint,early]) 5 model_final.save_weights("vgg16_1.h5") 10 frames /usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs) 984 except Exception as e: # pylint:disable=broad-except 985 if hasattr(e, "ag_error_metadata"): --> 986 raise e.ag_error_metadata.to_exception(e) 987 else: 988 raise ValueError: in user code: /usr/local/lib/python3.7/dist-packages/keras/engine/training.py:830 train_function * return step_function(self, iterator) /usr/local/lib/python3.7/dist-packages/keras/engine/training.py:813 run_step * outputs = model.train_step(data) /usr/local/lib/python3.7/dist-packages/keras/engine/training.py:771 train_step * loss = self.compiled_loss( /usr/local/lib/python3.7/dist-packages/keras/engine/compile_utils.py:201 __call__ * loss_value = loss_obj(y_t, y_p, sample_weight=sw) /usr/local/lib/python3.7/dist-packages/keras/losses.py:142 __call__ * losses = call_fn(y_true, y_pred) /usr/local/lib/python3.7/dist-packages/keras/losses.py:246 call * return ag_fn(y_true, y_pred, **self._fn_kwargs) /usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper ** return target(*args, **kwargs) /usr/local/lib/python3.7/dist-packages/keras/losses.py:1631 categorical_crossentropy y_true, y_pred, from_logits=from_logits) /usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:206 wrapper return target(*args, **kwargs) /usr/local/lib/python3.7/dist-packages/keras/backend.py:4827 categorical_crossentropy target.shape.assert_is_compatible_with(output.shape) /usr/local/lib/python3.7/dist-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, None) and (None, 28, 28, 12) are incompatible#train_dir_path is path to your training images train_data = train_generator.flow_from_directory(directory=train_dir_path,target_size=IMAGE_SHAPE , color_mode="rgb" , class_mode='categorical', batch_size=BATCH_SIZE , shuffle = True ) #valid_dir_path is path to your validation images valid_data = validation_generator.flow_from_directory(directory=valid_dir_path, target_size=IMAGE_SHAPE , color_mode="rgb" , class_mode='categorical' , batch_size=BATCH_SIZE , shuffle = True )