horizon0408 comments

Repositories
Issues
Comments

Results 2 comments of


                                            horizon0408

Dimension 1 in both shapes must be equal, but are 4096 and 64. Shapes are [64,4096] and [64,64]. From merging shape 0 with other shapes. for 'concat/concat_dim' (op: 'Pack') with input shapes: [64,4096], [64,64].

TF changed the API. You may try to use `concated = tf.concat([e5, a2], 1)` to see if it works.

Inference results from tensorflow version

To get the predictions in numpy, I just run ``` disparity = full_res_prediction_np if args.fullRes else raw_prediction_np np.save(dest[:-4], np.squeeze(disparity)) ``` in your `inference.py` The inputs are not rescaled. I run...