如何用我们自己的输入层替换预训练的输入层 tensorflow model...？
[英]How to replace the input layer of a pre-trained tensorflow model with our own input layer...?
问题描述
I have a pre-trained model with input shape of shape=(None,4096, 12).
我有一个预训练的 model，输入形状为 shape=(None,4096, 12)。 I want to use this trained model with my own input layer having shape=(None, 1250, 5).我想将这个训练有素的 model 与我自己的形状为 (None, 1250, 5) 的输入层一起使用。 i have tried the solution posted here .我已经尝试过此处发布的解决方案。 but i got error但我得到了错误
enter ValueError Traceback (most recent call last)
/tmp/ipykernel_34/2421370102.py in <module>
2 weights = [layer.get_weights() for layer in mod.layers[1:]]
3 for layer, weight in zip(new_model.layers[1:], weights):
----> 4 layer.set_weights(weight)
/opt/conda/lib/python3.7/site-packages/keras/engine/base_layer.py in
set_weights(self, weights)
1799 raise ValueError(
1800 'Layer weight shape %s not compatible with
provided weight '
-> 1801 'shape %s' % (ref_shape, weight_shape))
1802 weight_value_tuples.append((param, weight))
1803 weight_index += 1
ValueError: Layer weight shape (16, 5, 64) not compatible with
provided weight shape (16, 12, 64)
0 个解决方案
声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.
相关问题
用不同的通道替换预训练的 model 的输入层？ - Replacing the input layer of a pre-trained model with different channels?
到logits层的Tensorflow推理-适用于预训练模型上的批次 - Tensorflow inference upto logits layer - for batches on pre-trained model
如何使用我们自己的数据集定制预训练模型？ - How to customize a pre-trained model with our own dataset?
如何使用TensorFlow Java API移除预训练模型的输出层？ - How can I remove the output layer of pre-trained model with TensorFlow java api?
如何在 Keras 中更改预训练 CNN model 中层的 output？ - How to change output of a layer in a pre-trained CNN model in Keras?
从预训练的 model 中移除顶层，迁移学习，tensorflow (load_model) - Remove top layer from pre-trained model, transfer learning, tensorflow (load_model)
如何连接预训练的嵌入层和输入层 - how to concatenate pre trained embedding layer and Input layer
如何使用预训练的 model 进行双输入迁移学习 - how to use pre-trained model for dual input transfer learning
在预训练的 model 开头添加归一化层 - Add normalization layer at the begining of a pre-trained model
如何在TensorFlow中使用预训练模型 - How to use pre-trained model in TensorFlow
相关标签