I prepare dataset and instruction on the readme.md. However, when I run the train.py file, I has an error Incompatible shapes: [1,50,2,1] vs. [1,504,284,1] [[{{node mean_squared_error/SquaredDifference}}]].
I figure out that due to we call statement "sess.run([train_op, euc_loss], feed_dict=feed_dict_data)" and this statement will link to the function loss (tf.compat.v1.losses.mean_squared_error(est, gt)).
I prepare dataset and instruction on the readme.md. However, when I run the train.py file, I has an error Incompatible shapes: [1,50,2,1] vs. [1,504,284,1] [[{{node mean_squared_error/SquaredDifference}}]]. I figure out that due to we call statement "sess.run([train_op, euc_loss], feed_dict=feed_dict_data)" and this statement will link to the function loss (tf.compat.v1.losses.mean_squared_error(est, gt)).
Could you tell me how to solve this problem ?