Chu Qingqing

https://github.com/astaxie/bat/blob/master/httplib/httplib.go#L344 PipeReader already has `Close()`.

in windows: ':=' not mean raw json object

1

``` go bat POST http://www.baidu.com/ abc:='["123","456"]' ``` result in windows: ``` http POST / HTTP/1.1 Host: www.baidu.com Accept: application/json Accept-Encoding: gzip, deflate Content-Type: application/json User-Agent: bat/0.0.3 {"abc":"'[123,456]'"} ``` result in...

mpi_rc2_test中参数没有使用传入的命令行参数

2

例如fps的参数没有从enc_cmd传递到rc_cfg中： https://github.com/rockchip-linux/mpp/blob/c22b65ef229ca32666f15f451d8cb8e8ea7c675c/test/mpi_rc2_test.c#L434 这里是不是可以改成： ``` rc_cfg->fps_in_denorm = enc_cmd->fps_in_den; rc_cfg->fps_out_denorm = enc_cmd->fps_out_den; rc_cfg->fps_in_num = enc_cmd->fps_in_num; rc_cfg->fps_out_num = enc_cmd->fps_out_num; rc_cfg->fps_in_flex = enc_cmd->fps_in_flex; rc_cfg->fps_out_flex = enc_cmd->fps_out_flex; ```

无法完整复现mobilenet模型迁移到rknn

1

你好！ 我尝试完整复现mobilenet模型迁移到rknn的过程，即自行训练mobilenet模型、转成onnx、再转成rknn，最后推理。 但是我遇到了一个问题：在把我自行训练的onnx模型转换成rknn模型时，`examples/mobilenet/python/mobilenet.py`会使用onnx推理bell.jpg，结果偏差非常大： ``` -----TOP 5----- [148] score=0.06 class="n02071294 killer whale, killer, orca, grampus, sea wolf, Orcinus orca" [150] score=0.02 class="n02077923 sea lion" [258] score=0.02 class="n02111889 Samoyed, Samoyede" [129] score=0.01...