Question

当我使用lstm来完成多类标签时，我使用y_train作为我的标签输入。

以下代码：

y_train = yuantrain['LOCF']
labels = to_categorical(np.array(y_train),286)

错误代码提示：

Traceback (most recent call last):
File "<ipython-input-55-4ae3f21f520f>", line 1, in <module>
labels = to_categorical(np.array(y_train))
File "C:\ProgramData\Anaconda2\lib\site-packages\keras\utils\np_utils.py",
line 24, in to_categorical

categorical = np.zeros((n, num_classes))

MemoryError

下面的

是y_train（5）：

In [65]: 
y_train[0:5]
Out[65]: 
0    620245
1       282
2    500004
3    620193
4     60119
Name: LOCF, dtype: int64

我有一个尝试：

# biao qian chu li
# 处理标签为二进制，以及思考二进制的解码
labels = yuantrain["LOCF"].drop_duplicates()
#labels is the y_train's unique label
num_labels = len(labels) # (all of the unique labels)
one_hot = np.zeros((num_labels, num_labels), int)
np.fill_diagonal(one_hot, 1)
label_dict = dict(zip(labels, one_hot)) 
y_train = yuantrain['LOCF']
y_train = y_train.apply(lambda y: label_dict[y])

但我觉得这对我来说并不容易，我只能按自己的方式进行解码。

当我使用我的热门方式时，keras也会出现如下错误：

Train...
ERROR (theano.gof.opt): SeqOptimizer apply                  <theano.tensor.opt.FusionOptimizer object at 0x000000000D87B0F0>
ERROR (theano.gof.opt): Traceback:
ERROR (theano.gof.opt): Traceback (most recent call last):
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\opt.py", line 235, in apply
sub_prof = optimizer.optimize(fgraph)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\opt.py", line 87, in optimize
ret = self.apply(fgraph, *args, **kwargs)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\tensor\opt.py", line 7289, in apply
new_outputs = self.optimizer(node)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\tensor\opt.py", line 7122, in local_fuse
tv = gof.op.get_test_value(ii)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\op.py", line 987, in get_test_value
return PureOp._get_test_value(v_var)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\op.py", line 580, in _get_test_value
detailed_err_msg = utils.get_variable_trace_string(v)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\utils.py", line 134, in get_variable_trace_string
return sio.getvalue()
File "C:\ProgramData\Anaconda2\lib\StringIO.py", line 271, in getvalue
self.buf += ''.join(self.buflist)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe8 in position 27: ordinal not in range(128)
ERROR (theano.gof.opt): SeqOptimizer apply <theano.tensor.opt.FusionOptimizer object at 0x000000000D87B0F0>
ERROR (theano.gof.opt): Traceback:
ERROR (theano.gof.opt): Traceback (most recent call last):
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\opt.py", line 235, in apply
sub_prof = optimizer.optimize(fgraph)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\opt.py", line 87, in optimize
ret = self.apply(fgraph, *args, **kwargs)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\tensor\opt.py", line 7289, in apply
new_outputs = self.optimizer(node)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\tensor\opt.py", line  7122, in local_fuse
tv = gof.op.get_test_value(ii)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\op.py", line  987, in get_test_value
return PureOp._get_test_value(v_var)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\op.py", line 580, in _get_test_value
detailed_err_msg = utils.get_variable_trace_string(v)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\utils.py", line  134, in get_variable_trace_string
return sio.getvalue()
File "C:\ProgramData\Anaconda2\lib\StringIO.py", line 271, in getvalue
self.buf += ''.join(self.buflist)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe8 in position 27: ordinal not in range(128)
Train on 100000 samples, validate on 77963 samples
Epoch 1/5
Traceback (most recent call last):
File "<ipython-input-67-5ce4b6739b03>", line 1, in <module>
runfile('E:/XIAMENproject/Prediction_Guo/count.py',   wdir='E:/XIAMENproject/Prediction_Guo')
File "C:\ProgramData\Anaconda2\lib\site-packages\spyder\utils\site\sitecustomize.py", line 866, in runfile
execfile(filename, namespace)
File "C:\ProgramData\Anaconda2\lib\site-packages\spyder\utils\site\sitecustomize.py", line 87, in execfile
exec(compile(scripttext, filename, 'exec'), glob, loc)
File "E:/XIAMENproject/Prediction_Guo/count.py", line 150, in <module>
validation_data=(x_val, y_val))
File "C:\ProgramData\Anaconda2\lib\site-packages\keras\models.py", line 870, in fit
initial_epoch=initial_epoch)
File "C:\ProgramData\Anaconda2\lib\site-packages\keras\engine\training.py", line 1507, in fit
initial_epoch=initial_epoch)
File "C:\ProgramData\Anaconda2\lib\site-packages\keras\engine\training.py", line 1156, in _fit_loop
outs = f(ins_batch)
File "C:\ProgramData\Anaconda2\lib\site-packages\keras\backend\theano_backend.py", line 1196, in __call__
return self.function(*inputs)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\compile\function_module.py", line 805, in __call__
self.maker.inputs[i].variable)
File "C:\ProgramData\Anaconda2\lib\site-packages\theano\gof\utils.py", line 134, in get_variable_trace_string
return sio.getvalue()
File "C:\ProgramData\Anaconda2\lib\StringIO.py", line 271, in getvalue
self.buf += ''.join(self.buflist)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe8 in position 27:  ordinal not in range(128)

Answer 1

我将标签ID重置为连续ID，然后使用to_categorical。

当我使用to_分类，但有一个MemoryError

1 个答案: