search

lanagarmire / deepimpute

An accurate and efficient deep learning method for single-cell RNA-seq data imputation
MIT License
84 stars 27 forks source link

Getting an error after training #5

Closed LouisFaure closed 5 years ago

LouisFaure commented 5 years ago

Hi, thanks for this package, I am really looking forward to apply this on my dataset, sadly I am getting an error after training:

---- Hyperparameters summary ---- Layer-0 ['label: dense', 'activation: relu', 'nb_neurons: 128'] Layer-1 ['label: dropout', 'activation: dropout', 'rate: 0.2'] Layer-2 ['label: dense', 'activation: softplus'] Batch size 64 Learning rate 0.0001 Loss function <function wMSE at 0x7f5c28dff9d8> Traceback (most recent call last): File "DeepImpute.py", line 8, in imputed = deepImpute(data, NN_lim='auto', n_cores=16, cell_subset=1) File "/home/lfaure/anaconda3/lib/python3.6/site-packages/deepimpute/deepImpute.py", line 18, in deepImpute return multi.predict(data, imputed_only=imputed_only,restore_pos_values=restore_pos_values) TypeError: predict() got an unexpected keyword argument 'restore_pos_values'

Any idea of what could be the problem?

Best Wishes,

Louis Faure

Puumanamana commented 5 years ago

Hi Louis,

Thanks for reporting the error. I found the bug and it should work just fine now. Let me know if the problem persist of if you get any more errors.

Best, Cédric

LouisFaure commented 5 years ago

Hi Cédric,

Thanks a lot for the fast fix and answer! Error is fixed now. I am closing the issue.

BW,

Louis