Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Example "Fine-tuning for style recognition" is broken due to missing links to flickr images #5593

Closed
joelthchao opened this issue May 5, 2017 · 3 comments

Comments

@joelthchao
Copy link

Issue summary

Example Fine-tuning CaffeNet for Style Recognition on “Flickr Style” Data is broken.
Can't download any images by the scripts (examples/finetune_flickr_style/assemble_data.py)
Root cause should be missing link of images.

Steps to reproduce

python examples/finetune_flickr_style/assemble_data.py
Downloading 80000 images with 3 workers...
Writing train/val for 0 successfully downloaded images.

Links to images are broken (e.g. http://farm3.staticflickr.com/2887/12524447173)

Your system configuration

Operating system: OSX yosemite
Python or MATLAB version (for pycaffe and matcaffe respectively): python 3.5

@joelthchao
Copy link
Author

If anyone needs to find substitution download links, please refer to my code.

@kashiani
Copy link

kashiani commented Feb 4, 2018

@joelthchao I encountered to the same issue, then refered to your solution but there are some mismatches, because of using 2.7 python version ( like print and so on) ,,,after modifying them, there is an error like File "examples/finetune_flickr_style/assemble_data.py", line 47, in main images_per_style = int(sys.argv[4]) ValueError: invalid literal for int() with base 10: '--label=5' , I guess that it does not contain digits value . Do you have any opinion to sort out this error

@joelthchao
Copy link
Author

The purpose and entry of assembly_data.py are different between my script and the caffe's. You may use my code to download the images. (instruction) Then use these image to continue your work.

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

No branches or pull requests

2 participants