Python 更改预训练模型的批量大小

我试图从中加载模型，以便对图像进行推断。重点是我想一次预测多个图像，但当我尝试时，我得到以下错误：

代码：

错误：

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<timed exec> in <module>

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in __call__(self, *args, **kwargs)
   1079       TypeError: For invalid positional/keyword argument combinations.
   1080     """
-> 1081     return self._call_impl(args, kwargs)
   1082 
   1083   def _call_impl(self, args, kwargs, cancellation_manager=None):

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _call_impl(self, args, kwargs, cancellation_manager)
   1119       raise TypeError("Keyword arguments {} unknown. Expected {}.".format(
   1120           list(kwargs.keys()), list(self._arg_keywords)))
-> 1121     return self._call_flat(args, self.captured_inputs, cancellation_manager)
   1122 
   1123   def _filtered_call(self, args, kwargs):

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _call_flat(self, args, captured_inputs, cancellation_manager)
   1208                      arg_name, arg,
   1209                      self._func_graph.inputs[i].shape,
-> 1210                      arg.shape))
   1211       elif (self._signature is not None and
   1212             isinstance(self._signature[i], tensor_spec.TensorSpec)):

ValueError: The argument 'ImageTensor' (value Tensor("Const_3:0", shape=(2, 342, 513, 3), dtype=uint8)) is not compatible with the shape this function was traced with. Expected shape (1, None, None, 3), but got shape (2, 342, 513, 3).

If you called get_concrete_function, you may need to pass a tf.TensorSpec(..., shape=...) with a less specific shape, having None on axes which can vary.

---------------------------------------------------------------------------
ValueError回溯（最近一次调用上次）
在里面
~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow\u core/python/eager/function.py在调用中（self，*args，**kwargs）
1079 TypeError:用于无效的位置/关键字参数组合。
1080     """
->1081返回自我。调用impl（args，kwargs）
1082
1083 def_call_impl（self、args、kwargs、cancellation_manager=None）：
~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow\u core/python/eager/function.py in\u call\u impl（self、args、kwargs、cancellation\u manager）
1119 raise TypeError（“关键字参数{}未知。应为{}.”格式(
1120列表（kwargs.keys（）），列表（self.\u arg\u关键字）））
->1121返回self.\u调用\u平面（参数、self.captured\u输入、取消\u管理器）
1122
1123 def_filtered_调用（self、args、kwargs）：
~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow\u core/python/eager/function.py在调用平面中（self、args、捕获的输入、取消管理器）
1208 arg_名称，arg，
1209 self.\u func\u graph.inputs[i].形状，
->1210参数形状）
1211 elif（自我）签名不是无和
1212 isinstance（自我签名[i]，张量规范张量规范）：
ValueError:参数“ImageTensor”（值张量（“Const_3:0”，shape=（2342513，3），dtype=uint8））与跟踪此函数的形状不兼容。应为形状（1，无，无，3），但得到形状（2342513，3）。
如果调用get_concrete_函数，则可能需要传递一个tf.TensorSpec（…，shape=…）和一个不太具体的形状，在轴上没有任何形状可以变化。

---

这表明模型的固定批量大小为1。您知道如何更改吗？

我认为您对错误原因的解释不正确。据我所知，批量大小与优化器有关，而不是与模型有关。您可以传递要建模的任何批量大小。fit@MikeFurlender，我不想适合这个型号，因为它太贵了他已经预先训练过了，我想用不同于1的批次进行推断。你能给我们看一下预测代码吗？@sebastian sz我来编辑这个问题

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<timed exec> in <module>

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in __call__(self, *args, **kwargs)
   1079       TypeError: For invalid positional/keyword argument combinations.
   1080     """
-> 1081     return self._call_impl(args, kwargs)
   1082 
   1083   def _call_impl(self, args, kwargs, cancellation_manager=None):

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _call_impl(self, args, kwargs, cancellation_manager)
   1119       raise TypeError("Keyword arguments {} unknown. Expected {}.".format(
   1120           list(kwargs.keys()), list(self._arg_keywords)))
-> 1121     return self._call_flat(args, self.captured_inputs, cancellation_manager)
   1122 
   1123   def _filtered_call(self, args, kwargs):

~/miniconda3/envs/tex/lib/python3.7/site-packages/tensorflow_core/python/eager/function.py in _call_flat(self, args, captured_inputs, cancellation_manager)
   1208                      arg_name, arg,
   1209                      self._func_graph.inputs[i].shape,
-> 1210                      arg.shape))
   1211       elif (self._signature is not None and
   1212             isinstance(self._signature[i], tensor_spec.TensorSpec)):

ValueError: The argument 'ImageTensor' (value Tensor("Const_3:0", shape=(2, 342, 513, 3), dtype=uint8)) is not compatible with the shape this function was traced with. Expected shape (1, None, None, 3), but got shape (2, 342, 513, 3).

If you called get_concrete_function, you may need to pass a tf.TensorSpec(..., shape=...) with a less specific shape, having None on axes which can vary.