i can't run the code without QCsamples

[YanniWang-g]

I got the error: the settings of training： task: train meta_data: ./allearly_MSRT_knn_log.csv sample_data: ./batch1_6_alldata_order_batch_noqc.csv train_data: all save: ./result_noqc ae_encoder_units: [1000, 1000] ae_decoder_units: [1000, 1000] disc_b_units: [250, 250] disc_o_units: [250, 250] bottle_num: 500 dropouts: (0.3, 0.1, 0.3, 0.3) lambda_b: 1.0 lambda_o: 1.0 lr_rec: 0.0002 lr_disc_b: 0.005 lr_disc_o: 0.0005 epoch: (1000, 10, 700) use_batch_for_order: True batch_size: 64 load: None visdom_env: main visdom_port: 8097 num_workers: 12 use_log: False use_batch: None sample_size: None random_seed: 1234 device: None

Traceback (most recent call last): File "main.py", line 83, in main() File "main.py", line 30, in main random_seed=opts.random_seed) File "/storage1/lilab/student/ynwang/meta_data/batch_effact/knn_1499features/NormAE/datasets.py", line 136, in get_metabolic_data meta_df, y_df = pre_transfer(meta_df, y_df) File "/storage1/lilab/student/ynwang/meta_data/batch_effact/knn_1499features/NormAE/transfer.py", line 24, in call x = self.scaler.fit_transform(values) File "/home/ynwang/miniconda3/envs/python3.7/lib/python3.7/site-packages/sklearn/base.py", line 852, in fit_transform return self.fit(X, **fit_params).transform(X) File "/home/ynwang/miniconda3/envs/python3.7/lib/python3.7/site-packages/sklearn/preprocessing/_data.py", line 806, in fit return self.partial_fit(X, y, sample_weight) File "/home/ynwang/miniconda3/envs/python3.7/lib/python3.7/site-packages/sklearn/preprocessing/_data.py", line 847, in partial_fit reset=first_call, File "/home/ynwang/miniconda3/envs/python3.7/lib/python3.7/site-packages/sklearn/base.py", line 566, in _validate_data X = check_array(X, **check_params) File "/home/ynwang/miniconda3/envs/python3.7/lib/python3.7/site-packages/sklearn/utils/validation.py", line 817, in check_array % (n_features, array.shape, ensure_min_features, context) ValueError: Found array with 0 feature(s) (shape=(598, 0)) while a minimum of 1 is required by StandardScaler.

[S-KD]

I received the same error without QC samples. I narrow down the source of this error to lines 94-97 in datasets.py. Unfortunately, simply commenting these lines leads to new errors because later visual.py expect qc_pca. It appears that in the current form tool can't be used without QC samples.

@luyiyun could you please update the NormAE, so it can be used without QC samples?