thedatumorg / tsb-uad Goto Github PK

I tried to install the package with "pip install -r requirements.txt" but it seems there is a problem with numpy versions which different packages required. for example this error:
The conflict is caused by:
The user requested numpy==1.18.5
arch 5.0.1 depends on numpy>=1.17
h5py 2.10.0 depends on numpy>=1.7
hurst 0.0.5 depends on numpy>=1.10
matrixprofile 1.1.10 depends on numpy>=1.16.2
pandas 1.3.3 depends on numpy>=1.17.3
pmdarima 1.8.2 depends on numpy~=1.19.0

when I tried to fixed the error with another numpy version, I faced with similar errors

!

Hi, I tried to run the benchmark of yahoo a1 real data with lstm.
I wrote the code based on your example.

results = tsb_uad(data, label, 'LSTM', metric='all')

Then, an error came out,

No module named 'TSB_UAD.vus.affiliation'

I think maybe you need to add __init__.py in TSB_UAD/vus/affiliation.

In addition, versions of requires in setup.py and requirements.txt are quiet different. For now, I deleted install_requires in setup.py. Some of these versions are very old, can you tell me how to modify them better?

I am trying to apply series2graph (from TSB_UAD.models.series2graph import Series2Graph) to my data set. However, once I run the following code :

label = df[:max_length,1].astype(int)

Result is :

IndexError Traceback (most recent call last)
Cell In[21], line 1
----> 1 label = df[:max_length,1].astype(int)

IndexError: index 1 is out of bounds for axis 1 with size 1

When I apply the dataset uploaded to https://github.com/TheDatumOrg/TSB-UAD/tree/main/data/benchmark/ECG, no errors occur.

Please help me solve this issue.

Thanks in advance

Hey!
I was wondering if you had any code for downloading and processing the datasets so I could reproduce your benchmarks.
Thanks!