IndexError：运行python 3.9.1时元组索引超出范围

如何解决IndexError：运行python 3.9.1时元组索引超出范围

运行我的代码时出错

dataset_total = pd.concat((dataset['Open'],dataset_test['Open']),axis = 0)
inputs = dataset_total[len(dataset_total) - len(dataset_test) - 60:].values
inputs = inputs.reshape(-1,1)
inputs = sc.transform(inputs)
X_test = []
for i in range(60,80):
   X_test.append(inputs[i-60:i,0])
X_test = np.array(X_test)
X_test = np.reshape(X_test,(X_test.shape[0],X_test.shape[1],1))
predicted_forex_price = regressor.predict(X_test)
predicted_forex_price = sc.inverse_transform(predicted_forex_price)

结果是：

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:8: VisibleDeprecationWarning：从不规则的嵌套创建一个 ndarray 序列（这是一个列表或元组的列表或元组或 ndarrays 不同的长度或形状）已被弃用。如果你打算做这个，你必须在创建 ndarray 时指定 'dtype=object'

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-110-0e4e370b525c> in <module>()
      7 X_test.append(inputs[i-60:i,0])
      8 X_test = np.array(X_test)
----> 9 X_test = np.reshape(X_test,1))
     10 predicted_forex_price = regressor.predict(X_test)
     11 predicted_forex_price = sc.inverse_transform(predicted_forex_price)

IndexError: tuple index out of range

解决方法

你的切片长度不一样，所以 X_test 不是二维数组，而是一维数组，每个条目都是一个形状不一致的数组。

为了方便起见，这里演示了使用较小数组的问题：

inputs = np.arange(3)
X_test = [inputs[i:i + 2] for i in range(3)]

print(X_test)
# [array([0,1]),array([1,2]),array([2])]

X_test = np.array(X_test)
print(X_test)
# [array([0,1]) array([1,2]) array([2])]

np.reshape(X_test,(X_test.shape[0],X_test.shape[1],1))
# ---------------------------------------------------------------------------
# IndexError                                Traceback (most recent call last)
# <ipython-input-21-769dc2c0479b> in <module>()
#       6 print(X_test)
#       7 # [array([0,2]) array([2])]
# ----> 8 np.reshape(X_test,1))

# IndexError: tuple index out of range

要解决此问题，您需要确保 X_test 的原始构造包含所有长度相同的输入子集。例如：

X_test = [inputs[i:i + 2] for i in range(2)]
X_test = np.array(X_test)
np.reshape(X_test,1))
# array([[[0],#         [1]],#        [[1],#         [2]]])