Invalid argument: Incompatible shapes: [128,84] vs. [32,84]

Question

Invalid argument: Incompatible shapes: [128,84] vs. [32,84]

Opened this issue 6 years ago · 5 comments

tracing error：

/tools/../lib/fast_rcnn/train.py", line 113, in _modified_smooth_l1
inside_mul = tf.multiply(bbox_inside_weights, tf.subtract(bbox_pred, bbox_targets))

Answer 1 · 2019-04-28T14:47:02.000Z

Hi, I have the same problem, have you solved it?

Answer 2 · 2019-04-29T21:50:21.000Z

No. I also found other errors in the code. I implemented this SIN in Pytorch but cannot reproduce the claimed mAP.

Answer 3 · 2019-08-27T21:33:44.000Z

@zhl95 can you plz share your pytorch implementation?

Answer 4 · 2019-09-19T16:12:56.000Z

No. I also found other errors in the code. I implemented this SIN in Pytorch but cannot reproduce the claimed mAP.

Can you please share your pytorch implementation

Answer 5 · 2020-06-27T10:12:29.000Z

excuse me,do you have implemented the SIN model,I want to exchange some questions with you，Looking forward to your answer， thank you！@zhl95 @SayantiBardhan @monjurulkarim