zoo
zoo copied to clipboard
Data directory
Hi, if I want to train the model from scratch with my own dataset, how/where do I modify the code to set the directory?
Hey, thanks for checking out Larq! If your dataset is in TensorFlow Datasets, the easiest way is to define a Zookeeper @component
for it in larq_zoo/training/datasets.py
; you can then call it using the CLI as described in the docs.
Otherwise, the easiest way is to write your own training code (see Keras guide) and only importing the model from Larq Zoo, for example:
import larq_zoo as lqz
dataset = ... # https://www.tensorflow.org/guide/data
model = lqz.literature.BiRealNet() # https://docs.larq.dev/zoo/#available-models
model.compile(...)
model.fit(...)
Let me know if you have more questions!