search

molimat / YOLOv4-Counter-in-TF

An implementation of a tracker using yolov4, tf and sort_tracker. (still in development)
MIT License
24 stars 12 forks source link

detectvideo_counter.py #1

Closed byq-luo closed 4 years ago

byq-luo commented 4 years ago

Traceback (most recent call last): File "detectvideo_counter.py", line 189, in app.run(main) File "D:\ProgramData\Anaconda3\envs\drone\lib\site-packages\absl\app.py", line 299, in run _run_main(main, args) File "D:\ProgramData\Anaconda3\envs\drone\lib\site-packages\absl\app.py", line 250, in _run_main sys.exit(main(argv)) File "detectvideo_counter.py", line 115, in main pred_bbox = utils.postprocess_bbbox(pred_bbox, ANCHORS, STRIDES, XYSCALE) File "C:\Users\Administrator\Downloads\Compressed\tcc-d16ccad36f42818bb0dcbe116af70b802ee62f6f\core\utils.py", line 308, in postprocess_bbbox xy_grid = np.tile(tf.expand_dims(xy_grid, axis=0), [1, 1, 1, 3, 1]) File "<__array_function__ internals>", line 6, in tile File "D:\ProgramData\Anaconda3\envs\drone\lib\site-packages\numpy\lib\shape_base.py", line 1248, in tile c = _nx.array(A, copy=False, subok=True, ndmin=d) File "D:\ProgramData\Anaconda3\envs\drone\lib\site-packages\tensorflow_core\python\framework\ops.py", line 736, in array " array.".format(self.name)) NotImplementedError: Cannot convert a symbolic Tensor (ExpandDims:0) to a numpy array.

molimat commented 4 years ago

Hello, actually this file was just a test, i was not aware that someone was seeing it yet. You should use the _newtrack file which is correctly implemented. I will keep working in this and you can follow up right here.

byq-luo commented 4 years ago

Hello, actually this file was just a test, i was not aware that someone was seeing it yet. You should use the _newtrack file which is correctly implemented. I will keep working in this and you can follow up right here.

Thank you for your great work, I am very interested and had followed you.