search

floft / codats

Source code for "Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data" (KDD 2020)
https://arxiv.org/abs/2005.10996
52 stars 18 forks source link

code #2

Open Obsinaan opened 2 years ago

Obsinaan commented 2 years ago

Please,could you suggest how to run your code? thanks in advance!

oh-yu commented 1 year ago

If you apply this code to another problem, our commits in fork repo branch “feature/ecodataset” might be helpful. Execution steps are almost the same as the original.

1.local data(.csv) to .tfrecords python3 -m datasets.preprocess_for_ecodataset --household_id=2 2.visualize data python3 -m datasets.view_datasets --source=ecodataset_2 --normalize=none --maxexample=2 3.training, evaluation python3 main.py --logdir=example-logs --modeldir=example-models --method=dann --dataset=ecodataset --sources=1 --target=2 --uid=0 --debugnum=0 --gpumem=0 4.see the logs tensorboard --logdir ./example-logs

Obsinaan commented 1 year ago

Thanks! Helpful!

from normalization import calc_normalization, calc_normalization_jagged, apply_normalization, apply_normalization_jagged ModuleNotFoundError: No module named 'normalization'

We couldn't found 'normalization' module? pls any helps?

oh-yu commented 1 year ago

same execution as below?

pwd 
=> /something/codats

python3 -m datasets.view_datasets ...

hanhanzhengzheng commented 10 months ago

May I ask your TensorFlow, TensorFlow Addions, TensorFlow probability，What version are them?