Jaemin Cho

It could be related to high `num_workers` or HDF5. Could you try lowering the `num_workers` in dataloader, and/or replacing HDF5 with another file storing system (e.g., tsv, npy, pickle, lmdb,...

Glad that you found the issue! In my future projects, I probably should replace HDF5 with other formats.

I have not seen that cuDNN error. This might be an cuDNN compatibility issue for GTX 1660 Ti discussed [here](https://devtalk.nvidia.com/default/topic/1048627/cuda-setup-and-installation/does-the-latest-gtx-1660-model-support-cuda-/). Could you run other PyTorch code in your computer? Could...

좋은 정보 감사합니다!

Did you check arguments of `tf.contrib.seq2seq.dynamic_decode` as below? `impute_finished = True` `maximum_sequence_length=dec_sentence_length (int, maximum length of current batch)` ``` tf.contrib.seq2seq.dynamic_decode( inference_decoder, output_time_major=False, impute_finished=True, maximum_iterations=dec_sentence_length) ```

My pleasure! Right, as you said some codes are incompatible with TF 1.2, and I plan to update them (+ with new features, when I feel not so lazy :P...

And it seems like the current auto-downloaded checkpoint is not compatible with the current colab code. I originally thought it was due to the version mismatch above, but maybe there's...

It seems the 0.14.3.zip is downloaded in the wrong path (`/content/dalle-pytorch-pretrained/` instead of `/content/`) so that `DALLE-pytorch` directory is not created via `!unzip /content/0.14.3.zip -d /content/dalle-pytorch-pretrained` You have to add...

Late submission deadline updated (for watchers)

Hi, it is `language-evaluation` repo (https://github.com/bckim92/language-evaluation). You can install the package via ```bash pip install git+https://github.com/bckim92/language-evaluation.git python -c "import language_evaluation; language_evaluation.download('coco')" ``` I'll update the readme accordingly.