Tabnet model serialization error

🐛 Bug Description

When I save the model to a file then reload it for further training, there was this error of 'too many values to unpack'.
[9248:MainThread](2022-03-28 13:45:09,824) INFO - qlib.TabNet - [pytorch_tabnet.py:160] - Pretrain... [9248:MainThread](2022-03-28 13:45:10,259) INFO - qlib.TabNet - [pytorch_tabnet.py:134] - epoch: 0 [9248:MainThread](2022-03-28 13:45:10,259) INFO - qlib.TabNet - [pytorch_tabnet.py:135] - pre-training... [9248:MainThread](2022-03-28 13:45:12,005) ERROR - qlib.workflow - [utils.py:38] - An exception has been raised[ValueError: too many values to unpack (expected 2)]. File ".\rolling_workflow.py", line 333, in <module> best_score, best_epoch, loss_type = model.fit(dataset=dataset) File "C:\Users\Hense\AppData\Roaming\Python\Python38\site-packages\qlib\contrib\model\pytorch_tabnet.py", line 161, in fit self.pretrain_fn(dataset, self.pretrain_file) File "C:\Users\Hense\AppData\Roaming\Python\Python38\site-packages\qlib\contrib\model\pytorch_tabnet.py", line 136, in pretrain_fn self.pretrain_epoch(x_train) File "C:\Users\Hense\AppData\Roaming\Python\Python38\site-packages\qlib\contrib\model\pytorch_tabnet.py", line 327, in pretrain_epoch (vec, sparse_loss) = self.tabnet_model(feature, priors) ValueError: too many values to unpack (expected 2)

To Reproduce

Steps to reproduce the behavior:

1. save model by 'model.to_pickle()
2. load the model by pickle
3. train the model by model.fit()

Expected Behavior

model continue fitting.

Screenshot

Environment

Note: User could run cd scripts && python collect_info.py all under project directory to get system information
and paste them here directly.

• Qlib version: 0.8.0
• Python version: 3.8.12
• OS (Windows, Linux, MacOS): Windows 10 64bit
• Commit number (optional, please provide it if you are using the dev version):

Additional Notes