我是ML的新手。我正在尝试使用张量流实现多标签分类器
我在线上搜索过，但一直没有结束。也许我弄错了基础知识？
实现的代码如下：
import pandas as pd
import tensorflow as tf
train=pd.read_csv('data/train.csv',header=None)
print(train.columns)
columns=[294,295,296,297,298,299]
train_x=train.drop(columns,axis=1)
train_y=train[columns]
train_x=train_x.values
train_y=train_y.values
print(train_x.shape)
print(train_y.shape)
print(train_y)
train_dataset= tf.data.Dataset.from_tensor_slices((train_x,train_y))
print(train_dataset)
model = tf.keras.Sequential([
tf.keras.layers.Dense(units=294,activation='relu'),
tf.keras.layers.Dense(units=128, activation='relu'),
tf.keras.layers.Dense(units=6,activation='sigmoid')
])
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
model.fit(train_dataset, epochs=10)
输出：
Int64Index([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,
...
290, 291, 292, 293, 294, 295, 296, 297, 298, 299],
dtype='int64', length=300)
(1438, 294)
(1438, 6)
[[0 0 0 1 0 1]
[0 0 0 1 0 1]
[0 0 0 1 0 1]
...
[1 0 0 0 0 1]
[1 0 0 0 0 1]
[1 0 0 0 0 1]]
Traceback (most recent call last):
File "/home/shivam/Desktop/tftest/load.py", line 25, in <module>
model.fit(train_dataset, epochs=10)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training.py", line 819, in fit
use_multiprocessing=use_multiprocessing)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2.py", line 342, in fit
total_epochs=epochs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2.py", line 128, in run_one_epoch
batch_outs = execution_function(iterator)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 98, in execution_function
distributed_function(input_fn))
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 568, in __call__
result = self._call(*args, **kwds)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 615, in _call
self._initialize(args, kwds, add_initializers_to=initializers)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 497, in _initialize
*args, **kwds))
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 2389, in _get_concrete_function_internal_garbage_collected
graph_function, _, _ = self._maybe_define_function(args, kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 2703, in _maybe_define_function
graph_function = self._create_graph_function(args, kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 2593, in _create_graph_function
capture_by_value=self._capture_by_value),
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/framework/func_graph.py", line 978, in func_graph_from_py_func
func_outputs = python_func(*func_args, **func_kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 439, in wrapped_fn
return weak_wrapped_fn().__wrapped__(*args, **kwds)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 85, in distributed_function
per_replica_function, args=args)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 763, in experimental_run_v2
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 1819, in call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 2164, in _call_for_each_replica
return fn(*args, **kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/autograph/impl/api.py", line 292, in wrapper
return func(*args, **kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 433, in train_on_batch
output_loss_metrics=model._output_loss_metrics)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_eager.py", line 312, in train_on_batch
output_loss_metrics=output_loss_metrics))
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_eager.py", line 253, in _process_single_batch
training=training))
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_eager.py", line 167, in _model_loss
per_sample_losses = loss_fn.call(targets[i], outs[i])
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/losses.py", line 221, in call
return self.fn(y_true, y_pred, **self._fn_kwargs)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/losses.py", line 994, in binary_crossentropy
K.binary_crossentropy(y_true, y_pred, from_logits=from_logits), axis=-1)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/keras/backend.py", line 4615, in binary_crossentropy
return nn.sigmoid_cross_entropy_with_logits(labels=target, logits=output)
File "/home/shivam/anaconda3/envs/tf2/lib/python3.6/site-packages/tensorflow_core/python/ops/nn_impl.py", line 170, in sigmoid_cross_entropy_with_logits
(logits.get_shape(), labels.get_shape()))
ValueError: logits and labels must have the same shape ((294, 6) vs (6, 1))
Process finished with exit code 1
请有人告诉我我做错了吗？ 我尝试过SparseCategoricalCrossentropy，但也没有取得成功 也许我的数据集形状错误？
模型有问题吗？
答案 0 :(得分：1)
您正在解决一个多类问题，但是所使用的损失函数用于解决二元类问题。请如下更改compile语句。
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
注意： 如果您的目标是一键编码，请使用categorical_crossentropy。 一键编码的示例：
[1,0,0]
[0,1,0]
[0,0,1]
但是，如果您的目标是整数，请使用sparse_categorical_crossentropy。 整数编码的示例：
1
2
3
您可以在以下链接的keras中找到不同的损失函数-