When i tried to run again section "Training and Evaluating M4C-Captioner" according to M4C_Captioner/README.md, I suffered from dealing with this error. It said:

.... "/content/drive/MyDrive/CODE/AnchorCaptioner_KN/AnchorCaptioner/pythia/datasets/image_database.py", line 54, in _load_npy self.db = np.load(imdb_path, allow_pickle=True) File "/usr/local/lib/python3.10/dist-packages/numpy/lib/npyio.py", line 427, in load fid = stack.enter_context(open(os_fspath(file), "rb")) [Errno 2] No such file or directory: ' ../AnchorCaptioner/pythia/../data/imdb/m4c_textcaps/imdb_train_global_rule.npy'

The command i use: !python tools/run.py --tasks captioning --datasets m4c_textcaps --model m4c_captioner \ --config configs/captioning/m4c_textcaps/m4c_captioner.yml \ --save_dir save/m4c_captioner \ --run_type val --evalai_inference 1 \ --resume_file m4c_textcaps_m4c_captioner.ckpt

generate predictions on val

!python tools/run.py --tasks captioning --datasets m4c_textcaps --model m4c_captioner \ --config configs/captioning/m4c_textcaps/m4c_captioner.yml \ --save_dir save/m4c_captioner \ --run_type val --evalai_inference 1 \ --resume_file data/models/m4c_textcaps_m4c_captioner.ckpt

Here is the table of contents of my dataset:

data ├── imdb │ ├── m4c_textcaps │ | |── imdb_train_global_rule.npy │ | ├── imdb_val_filtered_by_image_id_global_rule.npy └── m4c_captioner_vocabs ├── models

I tried placing the /data folder inside AnchorCaptioner or /pythia but it didn't work.

Is there a problem with my folder structure?

guanghuixu / AnchorCaptioner

Problem with /data directory and imdb_train_global_rule.npy #17

generate predictions on val