tensorflow-ocr
tensorflow-ocr copied to clipboard
where is the data
I am running train_letter.py and return this "No such file or directory: 'fonts.list'"
Same question here. Also /data/fonts/ directory missing as well.
/data/fonts/DroidSans.ttf missing
sorry for the very late reply. the mechanism to automatically find the fonts was repaired.
i am running on windows, the newest version still not ok , message :
F:\OCR\tensorflow-ocr-master>python train_letters.py
extensions loaded
using OVERFIT DEBUG DATA!!!
Building fonts.list
'locate' 不是内部或外部命令,也不是可运行的程序
或批处理文件。
Traceback (most recent call last):
File "train_letters.py", line 4, in
and i search this code : def find_fonts(): if platform == "darwin": os.system("mdfind -name '.ttf' | grep '.ttf$' | iconv -f utf-8 -t ascii > fonts.list") else: os.system("locate '.ttf' | grep '.ttf$' > fonts.list") return readlines("fonts.list")
no windows....
added windows: sorry, how do I find fonts on Windows? falling back to Menlo.ttf