Skip to content
This repository has been archived by the owner on Oct 18, 2022. It is now read-only.

无法识别pics目录里面带猫的图片 #1

Open
sunjl opened this issue Dec 21, 2017 · 1 comment
Open

无法识别pics目录里面带猫的图片 #1

sunjl opened this issue Dec 21, 2017 · 1 comment

Comments

@sunjl
Copy link

sunjl commented Dec 21, 2017

下载VOC2007和PASCAL_VOC_SSD300_weight_checkpoint-epoch-16-loss-3.3392.hdf5文件,运行SSD.py,可以识别自行车、人物,图片上都能正确标注区域,但是所有带猫的图片没有标注。

我自己fork了你的项目,https://github.com/sunjl/keras_ssd,修改了SSD.py文件:

OUTPUT_DIR = './output'
MODEL_H5_NAME = 'weight.hdf5'
# MODEL_H5_NAME = 'PASCAL_VOC_SSD300_weight_checkpoint-epoch-16-loss-3.3392.hdf5'
model.save_weights(os.path.join(OUTPUT_DIR, MODEL_H5_NAME))
model.load_weights(os.path.join(OUTPUT_DIR, MODEL_H5_NAME))

自己训练模型,无法识别pics任一图片,怎么回事?
运行环境:Ubuntu 16.04,CUDA 8.0,Python 3.5,Keras 2.0.9,Tensorflow 1.3.0
GTX960显卡,4G显存不够用,所以把SSD.py文件中的batch改为1

@sunjl sunjl closed this as completed Dec 22, 2017
@sunjl sunjl reopened this Dec 22, 2017
@wikke
Copy link
Owner

wikke commented Dec 28, 2017

1、我的ipynb里也没有把猫框出来。也许设置阈值就可以出来吧,不过可能候选就很多了,也失去意义。
2、我是阿里云比较高的配置训练了几个小时吧。batch是1本身随机性就很大,需要更小的learning rate和更多的epochs。建议你升级配置,训练更长的时间。

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants