jxzly/kaggle-american-express-default-prediction-1st-solution

对经济学感兴趣，学过物理与天文，现在在做AI制药，想健康地活很久。

Interested in economics, studied physics and astronomy, now working on AI for drug discovery and want to live a long healthy life.

Google Scholar

Kaggle

kaggle-american-express-default-prediction-1st-solution's Issues

Encounter Problem when running `S2_manual_feature.py`

when the program run to this line

Kaggle-American-Express-Default-Prediction-1st-solution/S2_manual_feature.py

Line 94 in 03b86f5

df['rank'] = df.groupby('customer_ID')['S_2'].rank(ascending=False)

it encounters the error:

*** pandas.core.base.DataError: No numeric types to aggregate

FileNotFoundError: [Errno 2] No such file or directory: './output/NN_with_series_feature/fold0.ckpt'

Hi,
I am trying out this project on a AWS instance.
I have completed steps S1 - S5 without any errors or warnings.
But with the file S6_NN_main.py I get a file not found error as follows.

Traceback (most recent call last): File "/home/ubuntu/final-year-easy/S6_NN_main.py", line 41, in <module> NN_train_and_predict([df,f,y,series_idx.values[:y.shape[0]]],[df,f,series_idx.values[y.shape[0]:]],Amodel,nn_config,use_series_oof=False,run_id='NN_with_series_feature') File "/home/ubuntu/final-year-easy/utils.py", line 409, in NN_train_and_predict state_dict = torch.load(output_path + 'fold%s.ckpt'%fold, torch.device('cuda' if torch.cuda.is_available() else 'cpu') ) File "/home/ubuntu/.local/lib/python3.10/site-packages/torch/serialization.py", line 791, in load with _open_file_like(f, 'rb') as opened_file: File "/home/ubuntu/.local/lib/python3.10/site-packages/torch/serialization.py", line 271, in _open_file_like return _open_file(name_or_buffer, mode) File "/home/ubuntu/.local/lib/python3.10/site-packages/torch/serialization.py", line 252, in __init__ super().__init__(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: './output/NN_with_series_feature/fold0.ckpt'

I haven't changed anything other than the "folds = 5" to "folds = 3" value in nn_config.

I have a Tesla T4 GPU with CUDA enabled

DataError: No numeric types to aggregate

for the file from S2_manual_feature.py.
I got following errors.

Could you help me take a look?