hansbuehler/deephedging

Implementation of the vanilla Deep Hedging engine

deephedging's Issues

Financial Market Base Model

Dr. Hans, with the current development of LLM and GPT, do you believe that there exists a Base model for Financial Market? It will be trained with market data and info, and capture the relationships between different markets, e.g. Equity, FX, and etc. Once we have this, we could build other functionalities on top of it, like hedging.

Thanks

Tensorflow 2.11 update breaks use of Keras optimizers, quick fix included.

The current code is incompatible with Tensorflow 2.11 due to updates of the optimizer API:s.

The Adam and RMSprop optimizers no longer has the get_weights() method.

See some info here. A quick fix is to simply pass to the legacy namespace.

This implies changing one row in trainer.py and one row in trainer_serialize.ipynb.

In trainer.py change:
optimzier = config.train("optimizer", "RMSprop", help="Optimizer" )
to:
optimzier = config.train("optimizer", tf.keras.optimizers.legacy.RMSprop(), help="Optimizer" )

And in trainer_serialize.ipynb equivalently set the optimizer as:

...
# trainer
from tensorflow.keras.optimizers.legacy import Adam
config.trainer.train.optimizer = Adam()
...

Apparently the old optimizers are to be kept indefinitely thus this should be a relatively stable solution.