Cristóbal Valenzuela

This is still in beta! If you want access please send an email to [email protected]

The Openpose model expect base64 images and return an array of detected humans with body parts (x,y) recognized. But if you have a motion capture system why would you need...

Was the checkpoint generated from the same script?

Not sure what could be the issue. A new repo with some fixes was recently published here: https://github.com/ml5js/training-lstm, you could try that insted

great, thanks! we do have public account in Colab we can use to host shared notebooks. I will go over your notebook over the weekend!

This might be relevant to see: https://github.com/zaidalyafeai/Fast-Style-Transfer-Keras-TF.js Althought the repo says: > This is done using tensorflow.js check the file fast-style.html. Note that I have and editted version of the...

I'm setting `NODE_ENV` to `development` because I have another process that uses that to minify and preprocess files.

great! thanks for pointing out this resources. I guess we can close this for now

Seems to be an unicode issue with your text. Try replacing line [91 in train.py](https://github.com/ml5js/training-lstm/blob/master/train.py#L91): ```python data_loader = TextLoader(args.data_dir, args.batch_size, args.seq_length) ``` with ```python data_loader = TextLoader(args.data_dir, args.batch_size, args.seq_length, "ISO-8859-1")...

Yes, it is possible. Here are the arguments you need to pass to the script: ```python parser.add_argument('--init_from', type=str, default=None, help="""continue training from saved model at this path. Path must contain...