我想训练自己的模型但是报错了

[sherrygux]

我想训练512的模型，把尺寸改了但是有如下错误
2023-05-02 13:07:23.003633: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'Placeholder/_0' with dtype string and shape [1]
[[{{node Placeholder/_0}}]]
2023-05-02 13:07:23.003876: I tensorflow/core/common_runtime/executor.cc:1197] [/device:CPU:0] (DEBUG INFO) Executor start aborting (this does not indicate an error and you can ignore this message): INVALID_ARGUMENT: You must feed a value for placeholder tensor 'Placeholder/_0' with dtype string and shape [1]
[[{{node Placeholder/_0}}]]
Traceback (most recent call last):
File "/pfs/data5/home/kit/wbk/ea7309/ssd512_py3.8/TensorFlow2.0_SSD/train.py", line 94, in
main()
File "/pfs/data5/home/kit/wbk/ea7309/ssd512_py3.8/TensorFlow2.0_SSD/train.py", line 52, in main
loss_l, loss_c = loss_fn(y_true=labels, y_pred=predictions)
File "/pfs/data5/home/kit/wbk/ea7309/ssd512_py3.8/TensorFlow2.0_SSD/core/loss.py", line 46, in call
pos_idx = tf.broadcast_to(pos_idx, shape=loc_data.shape)
File "/opt/bwhpc/common/jupyter/tensorflow/2023-03-24/lib/python3.9/site-packages/tensorflow/python/ops/gen_array_ops.py", line 864, in broadcast_to
return broadcast_to_eager_fallback(
File "/opt/bwhpc/common/jupyter/tensorflow/2023-03-24/lib/python3.9/site-packages/tensorflow/python/ops/gen_array_ops.py", line 910, in broadcast_to_eager_fallback
_result = _execute.execute(b"BroadcastTo", 1, inputs=_inputs_flat,
File "/opt/bwhpc/common/jupyter/tensorflow/2023-03-24/lib/python3.9/site-packages/tensorflow/python/eager/execute.py", line 52, in quick_execute
tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node _wrapped__BroadcastTo_device/job:localhost/replica:0/task:0/device:GPU:0}} Incompatible shapes: [8,8732,1] vs. [8,24656,4] [Op:BroadcastTo]

[calmiLovesAI]

我现在不用TensorFlow了，你可以用我最新写的这个代码：https://github.com/calmisential/DeepLearning.pytorch，里面有SSD算法

[sherrygux]

我现在不用TensorFlow了，你可以用我最新写的这个代码：[https://github.com/calmisential/DeepLearning.pytorch，里面有SSD算法]我打开不了这个网页

[calmiLovesAI]

这个：
https://github.com/calmisential/DeepLearning.pytorch

[calmiLovesAI]

SSD的配置文件是：
https://github.com/calmisential/DeepLearning.pytorch/blob/main/configs/ssd_cfg.py
你可以在其中修改相关训练参数
接着把train.py里面的CONFIG修改为ssd的配置文件路径，然后运行train.py就可以开始训练了。
ssd要训出好的效果很难，建议你直接用我这个仓库里面的YOLOv8来做目标检测

[sherrygux]

如果我使用的是512*512的图是直接改ssd_ cfg的尺寸就可以了嘛

[calmiLovesAI]

如果我使用的是512*512的图是直接改ssd_ cfg的尺寸就可以了嘛

输入只能是300*300，SSD512的网络结构与SSD300不一样，后续我有时间会把SSD512加上去

[sherrygux]

好的，谢谢

[calmiLovesAI]

好的，谢谢

我更新了代码，现在可以输入512*512尺寸的图片了

[sherrygux]

你有git push吗？为什么我没看到有更改呢？

[calmiLovesAI]

你有git push吗？为什么我没看到有更改呢？

这是commit