Specify the ___location of pretrained resnet model.

Author: ruotianluo

README.md

@@ -8,8 +8,10 @@ There's something difference compared to neuraltalk2.
 # Requirements
 Python 2.7 (no [coco-caption](https://github.com/tylin/coco-caption) version for python 3), pytorch
 
-# Pretrained FC model.
-Download pretrained model from [link](https://drive.google.com/drive/folders/0B7fNdx_jAqhtOVBabHRCQzJ1Skk?usp=sharing). You also need pretrained resnet which can be downloaded from [pytorch-resnet](https://github.com/ruotianluo/pytorch-resnet.git).
+# Pretrained models.
+You need pretrained resnet both for training and evaluation. The models can be downloaded from [here](https://drive.google.com/open?id=0B7fNdx_jAqhtbVYzOURMdDNHSGM), and should be placed in `data/imagenet_weights`.
+
+We also provide pretrained fc model, and you can download it from [here](https://drive.google.com/drive/folders/0B7fNdx_jAqhtOVBabHRCQzJ1Skk?usp=sharing).
 
 Then you can follow [this section](#markdown-header-caption-images-after-training).
 
@@ -29,9 +31,6 @@ $ python scripts/prepro_labels.py --input_json .../dataset_coco.json --output_js
 $ python scripts/prepro_feats_npy.py --input_json .../dataset_coco.json --output_dir data/cocotalk --images_root ...
 ```
 
-(prepro_feats_npy.py uses the resnet converted from caffe
-[pytorch-resnet](https://github.com/ruotianluo/pytorch-resnet.git). )
-
 You need to download [dataset_coco.json](http://cs.stanford.edu/people/karpathy/deepimagesent/caption_datasets.zip) from Karpathy's homepage.
 
 This is telling the script to read in all the data (the images and the captions), allocate the images to different splits according to the split json file, extract the resnet101 features (both fc feature and last conv feature) of each image, and map all words that occur <= 5 times to a special `UNK` token. The resulting `json` and `h5` files are about 200GB and contain everything we want to know about the dataset.

dataloaderraw.py

@@ -23,7 +23,7 @@
 import misc.resnet as resnet
 
 resnet = resnet.resnet101()
-resnet.load_state_dict(torch.load('/home-nfs/rluo/rluo/model/pytorch-resnet/resnet101.pth'))
+resnet.load_state_dict(torch.load('./data/imagenet_weights/resnet101.pth'))
 my_resnet = myResnet(resnet)
 my_resnet.cuda()
 my_resnet.eval()

scripts/prepro_feats_npy.py

@@ -102,7 +102,7 @@ def main(params):
   parser.add_argument('--images_root', default='', help='root ___location in which images are stored, to be prepended to file_path in input json')
   parser.add_argument('--att_size', default=14, type=int, help='14x14 or 7x7')
   parser.add_argument('--model', default='resnet101', type=str, help='resnet101, resnet152')
-  parser.add_argument('--model_root', default='/home-nfs/rluo/rluo/model/pytorch-resnet/', type=str, help='model root')
+  parser.add_argument('--model_root', default='./data/imagenet_weights', type=str, help='model root')
 
   args = parser.parse_args()
   params = vars(args) # convert to ordinary dict