Open jiar1019 opened 3 years ago
if you want to train MSCOCO, the directory construct will be formatted as follow:
.
├── annotation
│ ├── controllable
│ │ ├── cleaned_tst_region_descriptions.json
│ │ ├── int2word.npy
│ │ ├── public_split # 图片文件名,切分方式
│ │ │ ├── trn_names.npy
│ │ │ ├── tst_names.npy
│ │ │ └── val_names.npy
│ │ ├── regionfiles # object ID 对应的图片,冗余信息,在 json 文件中完全包括了
│ │ │ ├── trn_names.npy
│ │ │ ├── tst_names.npy
│ │ │ └── val_names.npy
│ │ │ ├── train2014 # 每个json 是一个 dict,level1 每个key 是 object ID,对应的 value 是相应
│ │ │ │ ├── COCO_train2014_000000000009.json
│ │ │ │ ├── COCO_train2014_000000000025.json
│ │ │ │ ├── COCO_train2014_000000000030.json
│ │ │ ├── val2014
│ │ │ │ ├── COCO_val2014_000000000042.json
│ │ │ │ ├── COCO_val2014_000000000073.json
│ │ │ │ ├── COCO_val2014_000000000074.json
│ │ │ │ ├── COCO_val2014_000000000133.json
│ │ │ │ ├── COCO_val2014_000000000136.json
│ │ │ │ ├── COCO_val2014_000000040011.json
│ │ └── word2int.json
├── dir.txt
├── ordered_feature
│ ├── MP
│ │ └── resnet101.ctrl
│ │ ├── trn_ft.npy
│ │ ├── tst_ft.npy
│ │ └── val_ft.npy
│ └── SA
│ └── X_101_32x8d
│ └── objrels
│ ├── train2014_COCO_train2014_000000000009.jpg.hdf5
│ ├── train2014_COCO_train2014_000000581909.jpg.hdf5
│ ├── train2014_COCO_train2014_000000581921.jpg.hdf5
│ ├── val2014_COCO_val2014_000000000074.jpg.hdf5
│ ├── val2014_COCO_val2014_000000000139.jpg.hdf5
│ └── val2014_COCO_val2014_000000581929.jpg.hdf5
└── results
if you want to train MSCOCO, the directory construct will be formatted as follow:
. ├── annotation │ ├── controllable │ │ ├── cleaned_tst_region_descriptions.json │ │ ├── int2word.npy │ │ ├── public_split # 图片文件名,切分方式 │ │ │ ├── trn_names.npy │ │ │ ├── tst_names.npy │ │ │ └── val_names.npy │ │ ├── regionfiles # object ID 对应的图片,冗余信息,在 json 文件中完全包括了 │ │ │ ├── trn_names.npy │ │ │ ├── tst_names.npy │ │ │ └── val_names.npy │ │ │ ├── train2014 # 每个json 是一个 dict,level1 每个key 是 object ID,对应的 value 是相应 │ │ │ │ ├── COCO_train2014_000000000009.json │ │ │ │ ├── COCO_train2014_000000000025.json │ │ │ │ ├── COCO_train2014_000000000030.json │ │ │ ├── val2014 │ │ │ │ ├── COCO_val2014_000000000042.json │ │ │ │ ├── COCO_val2014_000000000073.json │ │ │ │ ├── COCO_val2014_000000000074.json │ │ │ │ ├── COCO_val2014_000000000133.json │ │ │ │ ├── COCO_val2014_000000000136.json │ │ │ │ ├── COCO_val2014_000000040011.json │ │ └── word2int.json ├── dir.txt ├── ordered_feature │ ├── MP │ │ └── resnet101.ctrl │ │ ├── trn_ft.npy │ │ ├── tst_ft.npy │ │ └── val_ft.npy │ └── SA │ └── X_101_32x8d │ └── objrels │ ├── train2014_COCO_train2014_000000000009.jpg.hdf5 │ ├── train2014_COCO_train2014_000000581909.jpg.hdf5 │ ├── train2014_COCO_train2014_000000581921.jpg.hdf5 │ ├── val2014_COCO_val2014_000000000074.jpg.hdf5 │ ├── val2014_COCO_val2014_000000000139.jpg.hdf5 │ └── val2014_COCO_val2014_000000581929.jpg.hdf5 └── results
thank you for your reply but where are dir.txt and results?
dir.txt is not needed, it contains the output of the tree
command, which is my reply content.
dir.txt is not needed, it contains the output of the
tree
command, which is my reply content.
ok thank you sir but when I run ‘python configs/prepare_coco_imgsg_config.py $mtype’ it show no module nemed caption how to fix it?
dir.txt is not needed, it contains the output of the
tree
command, which is my reply content.ok thank you sir but when I run ‘python configs/prepare_coco_imgsg_config.py $mtype’ it show no module nemed caption how to fix it?
I'd suggest you reimplement this project, copy the core code, pad train logic and ignore this unimportant preprocess code. It won't take you much time. Here is my repo asg2cap implemented under my train framework
Can you tell me the directory format of the dataset