Kitiro / DHG

Directed Hierarchical Graph for Zero-shot Learning on ImageNet

Directed Hierarchical Graph for Zero-shot Learning on ImageNet

Requirements

• python 3
• pytorch 0.4.0
• nltk

Instructions

Materials Preparation

There is a folder materials/, which contains some meta data and programs already.

Glove Word Embedding

1. Download: http://nlp.stanford.edu/data/glove.6B.zip
2. Unzip it, find and put glove.6B.300d.txt to materials/.

Build Word-Net Graphs

1. cd materials/
2. Run python make_induced_graph.py, get imagenet-induced-graph.json
3. Run python make_dense_graph.py, get imagenet-dense-graph.json # 200082 edges
4. Run python make_dense_grouped_graph.py, get imagenet-dense-grouped-graph.json

Build NELL Graphs

1. cd materials/
2. find the construct section in 'construct_multi_weight_graph.ipynb'.

Obtain Pretrained ResNet101

cd materials/, run python process_resnet.py, get fc-weights.json and resnet101-base.pth

ImageNet and AwA2

Download ImageNet by your own. and AwA2, create the softlinks (command ln -s): materials/datasets/imagenet and materials/datasets/awa2, to the root directory of the dataset.

An ImageNet root directory should contain image folders, each folder with the wordnet id of the class.

An AwA2 root directory should contain the folder JPEGImages.

Training

Make a directory save/ for saving models.

In most programs, use --gpu to specify the devices to run the code (default: use gpu 0).

Train Graph Networks

• Only-word-vectors: Run python train_gcn.py, get results in save/gcn
• DHG: Run python train_gcn_att.py, get results in save/gcn-att
• DHG-r: Run python train_gcn_att_r.py, get results in save/gcn-att-r, in which edges are grouped by directions.

In the results folder:

• *.pth is the state dict of Graph Networks model
• *.pred is the prediction file, which can be loaded by torch.load(). It is a python dict, having two keys: wnids - the wordnet ids of the predicted classes, pred - the predicted fc weights

Finetune ResNet

Run python train_resnet_fit.py with the args:

• --pred: the .pred file for finetuning
• --train-dir: the directory contains 1K imagenet training classes, each class with a folder named by its wordnet id
• --save-path: the folder you want to save the result, e.g. save/resnet-fit-xxx

python train_resnet_fit.py --pred save/gcn-dense-att/epoch-3000.pred --train-dir materials/datasets/imagenet --save-path save/resnet-fit

(In the paper's setting, --train-dir is the folder composed of 1K classes from fall2011.tar, with the missing class "teddy bear" from ILSVRC2012.)

Testing

ImageNet

Run python evaluate_imagenet.py with the args:

• --cnn: path to resnet101 weights, e.g. materials/resnet101-base.pth or save/resnet-fit-xxx/x.pth
• --pred: the .pred file for testing
• --test-set: load test set in materials/imagenet-testsets.json, choices: [2-hops, 3-hops, all]

python evaluate_imagenet.py --cnn materials/resnet101-base.pth --pred save/gcn-dense-att/epoch-3000.pred --test-set 2-hops

• (optional) --keep-ratio for the ratio of testing data, --consider-trains to include training classes' classifiers, --test-train for testing with train classes images only.

AwA2

Run python evaluate_awa2.py with the args:

• --cnn: path to resnet101 weights, e.g. materials/resnet101-base.pth or save/resnet-fit-xxx/x.pth
• --pred: the .pred file for testing
• (optional) --consider-trains to include training classes' classifiers