Repository files navigation

The code has been tested in the following environment:

# Create and activate the environmentcondacreate-ntimemarpython=3.10-ycondaactivatetimemar# Install dependenciespipinstall-rrequirements.txt

The trained model checkpoints will be saved in a directory such aslog/vq_stock/version_0/checkpoints.

pythontrain_vqvae.py

Set the path to the trained VQ-VAE model in theload_dir field of the configuration fileconfigs/train_var_stock.yaml.

pythontrain_ar.py

Specify the path to the trained VQ-VAE model in thevar_path: load_dir field ofconfigs/train_sample_stock.yaml. The generated data will be saved in a directory likeoutput/stock/generated_samples/24/.

pythonsample.py

Evaluate the quality of the generated data.

Pass the path to the generated data as the second argument to theeval() function ineval.py.

# Inside eval.pyeval("stock","output/stock/generated_samples/24/20xx-xx-xx_xx-xx-xx/manual_generate.npy")pythoneval.py