soeque1 commented on issue #13647: Update lip reading example URL: https://github.com/apache/incubator-mxnet/pull/13647#issuecomment-451321255 > I'm about 10-11 hours in on the preprocessing. And it is on the s7 folder, which if it has been going in order - we've got a really long time to go! Was I supposed to pass a parameter to the script to engage multi-processing? I thought it did that by default... > I'm curious, can the preprocessed data be loaded onto s3 or is there some licensing limitation? Also, if I break out of this process, do I start from scratch? (1) As you said, the default of preprocess in dowload_data.py and preprocess_data.py is one cpu. (2) We can upload preprocessed data beacuse of no licensing limitation. (3) You can choose 3 options. (3-1) - As you were just test the model, you can train the model now. (because data_loader collects the data in complete image folder, and than attach align_data. So 'image' is the key, you can train the incomplete data. (3-2) - Stop the preprocess. Then resume the process. If therer the processed image, it skips. (3-3) - If you just test the model, I uplolad the preprocessed files temporarily onto S3 (personal account). But, just about 10 days possible because of the cost.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
