时间戳

首页 > 解决方案 > 当我只有两个输入时，为什么 Keras 将我的 input_shape 视为三维？

r - 当我只有两个输入时，为什么 Keras 将我的 input_shape 视为三维？

问题描述

这是我目前遇到的 ValueError：

ValueError: Input 0 of layer sequential_10 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 9]

应该注意的是，我正在使用 Keras、Caret 和 Tidyverse 处理 R。

我目前正在 Kaggle 的 Titanic 数据集上构建神经网络模型。

从抛出的 ValueError 中，我了解到我描述X.train数据形状的方式存在问题。虽然，我不确定如何塑造这些数据以使我的模型运行顺畅。

以下是我开始构建模型的方式：

#Build Model
input_shape <- shape(ncol(X.train), nrow(X.train))
                     
model <- keras_model_sequential()

model %>%
    layer_batch_normalization(input_shape = input_shape) %>% #Normalization Layer
    layer_dense(units = 256, activation = 'relu') %>%  #First Layer
    layer_batch_normalization() %>%
    layer_dropout(rate = 0.3) %>% 
    layer_dense(units = 256, activation = 'relu') %>% #Second layer
    layer_batch_normalization() %>%
    layer_dropout(rate = 0.3) %>%
    layer_dense(units = 256, activation = 'relu') %>% #Third layer
    layer_batch_normalization() %>%
    layer_dropout(rate = 0.3) %>%
    layer_dense(units = 1, activation = 'sigmoid') %>% #Output Layer
    compile(
      loss = 'binary_crossentropy',
      optimizer = 'adam',
      metrics = c('accuracy'))

这是我遇到的完整错误：

ValueError: Input 0 of layer sequential_10 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 9]


Detailed traceback: 
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py", line 108, in _method_wrapper
    return method(self, *args, **kwargs)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py", line 1098, in fit
    tmp_logs = train_function(iterator)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 780, in __call__
    result = self._call(*args, **kwds)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 823, in _call
    self._initialize(args, kwds, add_initializers_to=initializers)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 697, in _initialize
    *args, **kwds))
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 2855, in _get_concrete_function_internal_garbage_collected
    graph_function, _, _ = self._maybe_define_function(args, kwargs)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 3213, in _maybe_define_function
    graph_function = self._create_graph_function(args, kwargs)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/function.py", line 3075, in _create_graph_function
    capture_by_value=self._capture_by_value),
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/framework/func_graph.py", line 986, in func_graph_from_py_func
    func_outputs = python_func(*func_args, **func_kwargs)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py", line 600, in wrapped_fn
    return weak_wrapped_fn().__wrapped__(*args, **kwds)
  File "/usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/framework/func_graph.py", line 973, in wrapper
    raise e.ag_error_metadata.to_exception(e)

Traceback:

1. model %>% fit(X.train, y.train, epochs = 500, batch_size = 200, 
 .     validation_split = 0.3, callbacks = list(callback_early_stopping(monitor = "val_loss", 
 .         mode = "auto", patience = 5, min_delta = 0.001, restore_best_weights = TRUE)))
2. withVisible(eval(quote(`_fseq`(`_lhs`)), env, env))
3. eval(quote(`_fseq`(`_lhs`)), env, env)
4. eval(quote(`_fseq`(`_lhs`)), env, env)
5. `_fseq`(`_lhs`)
6. freduce(value, `_function_list`)
7. withVisible(function_list[[k]](value))
8. function_list[[k]](value)
9. fit(., X.train, y.train, epochs = 500, batch_size = 200, validation_split = 0.3, 
 .     callbacks = list(callback_early_stopping(monitor = "val_loss", 
 .         mode = "auto", patience = 5, min_delta = 0.001, restore_best_weights = TRUE)))
10. fit.keras.engine.training.Model(., X.train, y.train, epochs = 500, 
  .     batch_size = 200, validation_split = 0.3, callbacks = list(callback_early_stopping(monitor = "val_loss", 
  .         mode = "auto", patience = 5, min_delta = 0.001, restore_best_weights = TRUE)))
11. do.call(object$fit, args)
12. (structure(function (...) 
  . {
  .     dots <- py_resolve_dots(list(...))
  .     result <- py_call_impl(callable, dots$args, dots$keywords)
  .     if (convert) 
  .         result <- py_to_r(result)
  .     if (is.null(result)) 
  .         invisible(result)
  .     else result
  . }, class = c("python.builtin.method", "python.builtin.object"
  . ), py_object = <environment>))(batch_size = 200L, epochs = 500L, 
  .     verbose = 1L, callbacks = list(<environment>, <environment>), 
  .     validation_split = 0.3, shuffle = TRUE, class_weight = NULL, 
  .     sample_weight = NULL, initial_epoch = 0L, x = <environment>, 
  .     y = <environment>)
13. py_call_impl(callable, dots$args, dots$keywords)
Error in py_call_impl(callable, dots$args, dots$keywords): ValueError: in user code: /usr/local/share/.virtualenvs/r-reticulate/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py:806 train_function * return step_function(self, iterator) /usr/local/share/.virtual
keyboard_arrow_right

谢谢你，我很感激你的反馈。

标签： rtensorflowkeraserror-handlingneural-network

解决方案

推荐阅读