search

avisingh599 / visual-qa

[Reimplementation Antol et al 2015] Keras-based LSTM/CNN models for Visual Question Answering
https://avisingh599.github.io/deeplearning/visual-qa/
MIT License
481 stars 186 forks source link

Make an installation guide for the dependencies #3

Open avisingh599 opened 8 years ago

avisingh599 commented 8 years ago

Temporary solution: this guide

dhruvbatra commented 8 years ago

Also, the help for running the various scripts is incomplete (in some cases incorrect). For instance, main page says "own_image.py" can be run via

python extract_features.py --caffe /path/to/caffe

But it also requires path to VGG model and prototxt.

avisingh599 commented 8 years ago

Thanks for reporting this. @arushk1 can you have a look at this?

arushk1 commented 8 years ago

The VGG prototxt file is already included in the scripts folder, and according to the instructions, I have asked users to download the caffemodel file and place it in the scripts folder. In any case, I added an update for setting a custom path to the VGG caffemodel file here : ttps://github.com/arushk1/visual-qa/, and submitted a PR. @dhruvbatra @avisingh599

avisingh599 commented 8 years ago

I am not able to see the PR.