Li Xiucheng
Li Xiucheng
Hi @rose-jinyang, sorry for the late reply. The [script](https://github.com/boathit/t2vec/blob/master/experiment/experiment.jl) holds the code where I used in the Jupyter notebook, and cannot be run directly in the terminal. The experiments actually...
You can freeze one or more layers simply by using `tl.Fn`. Suppose you have a model `m0 = tl.Serial(m1, m2, m3)` with three layers `m1, m2, m3` and wish to...
That is easy. You can decompose a loaded model into a collection of sub-models and reconstruct the model using `tl.Fn` layer. ```python m0 = tl.Serial( tl.Dense(3), tl.Dense(2), tl.Dense(1) ) ##...
@friesel Did you try [`save_to_file`](https://github.com/google/trax/blob/master/trax/layers/base.py#L350) function?
I have never conducted experiments on the Chengdu dataset and we only used the Harbin dataset in our paper. But as far as I know, the GPS points of the...
I will try to test the models on the Chengdu dataset and figure out why DeepGTT fails.
The availability of our whole dataset is subject to the grant from our partner. If you only intend to reproduce our experiments reported in the paper, the 5 days data...
@xanderdunn you have to use like `Lazy.take(20, fibs)` see this thread https://github.com/one-more-minute/Lazy.jl/issues/27