报错ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type numpy.ndarray)

[yyamx]

如何解决错误

[duanshengliu]

贴一下具体的代码，不然不知道是什么情况

[yyamx]

贴一下具体的代码，不然不知道是什么情况

File "E:\End-to-end-for-chinese-plate-recognition-master\License-plate-recognition\Unet.py", line 96, in unet_train model.fit(X_train, y_train, epochs=100, batch_size=15)#epochs和batch_size看个人情况调整，batch_size不要过大，否则内存容易溢出 File "D:\anaconda3\envs\pytorch\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler raise e.with_traceback(filtered_tb) from None File "D:\anaconda3\envs\pytorch\lib\site-packages\tensorflow\python\framework\constant_op.py", line 102, in convert_to_eager_tensor return ops.EagerTensor(value, ctx.device_name, dtype) ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type numpy.ndarray).

[duanshengliu]

你的TF版本是多少，是2.x吗？

[yyamx]

你的TF版本是多少，是2.x吗？

是的 2.8，必须降低版本嘛

[duanshengliu]

你的TF版本是多少，是2.x吗？

是的 2.8，必须降低版本嘛

我之前训练用的是TF1.15，你可以试一下

[yyamx]

Traceback (most recent call last): File "E:/End-to-end-for-chinese-plate-recognition-master/License-plate-recognition/train.py", line 6, in unet_train()#训练后得到unet.h5,用于车牌定位 File "E:\End-to-end-for-chinese-plate-recognition-master\License-plate-recognition\Unet.py", line 96, in unet_train model.fit(X_train, y_train, epochs=100, batch_size=15)#epochs和batch_size看个人情况调整，batch_size不要过大，否则内存容易溢出 File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 1406, in fit batch_size=batch_size) File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 1300, in _standardize_user_data exception_prefix='model input') File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 138, in _standardize_input_data ' dimensions, but got array with shape ' + str(array.shape)) ValueError: Error when checking model input: expected input_1 to have 4 dimensions, but got array with shape (2475, 1) 更改之后的错误，这个如何解决

[duanshengliu]

Traceback (most recent call last): File "E:/End-to-end-for-chinese-plate-recognition-master/License-plate-recognition/train.py", line 6, in unet_train()#训练后得到unet.h5,用于车牌定位 File "E:\End-to-end-for-chinese-plate-recognition-master\License-plate-recognition\Unet.py", line 96, in unet_train model.fit(X_train, y_train, epochs=100, batch_size=15)#epochs和batch_size看个人情况调整，batch_size不要过大，否则内存容易溢出 File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 1406, in fit batch_size=batch_size) File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 1300, in _standardize_user_data exception_prefix='model input') File "D:\anaconda3\envs\hyperlpr36\lib\site-packages\tensorflow\contrib\keras\python\keras\engine\training.py", line 138, in _standardize_input_data ' dimensions, but got array with shape ' + str(array.shape)) ValueError: Error when checking model input: expected input_1 to have 4 dimensions, but got array with shape (2475, 1) 更改之后的错误，这个如何解决

我刚刚使用TF2.8测试了没有问题，可以训练，你的数据集是图片吗，shape不对？

[yyamx]

是的Img是jpg格式，label是png格式，大小必须是512？我可否直接改成320

[yyamx]

训练图片大小有要求没

[duanshengliu]

是的Img是jpg格式，label是png格式，大小必须是512？我可否直接改成320

是的，需要512*512的

[yyamx]

是的Img是jpg格式，label是png格式，大小必须是512？我可否直接改成320

是的，需要512*512的

就是说训练之前的数据集准备中就必须把它reshape成512x512的，我要是想320x320的话，具体的unet网络代码是否需要修改

[duanshengliu]

是的Img是jpg格式，label是png格式，大小必须是512？我可否直接改成320

是的，需要512*512的

就是说训练之前的数据集准备中就必须把它reshape成512x512的，我要是想320x320的话，具体的unet网络代码是否需要修改 是的