The matrix is organized as genes (probsets) as rows and treatments as columns. An example is shown below.
assay-pathology-probe,PIRACETAM-3d-2500mg/kg,ACETAMINOPHE-.25d-486mg/kg,CLEMASTINE-5d-5.4mg/kg,ARTEMETHER-1d-74mg/kg,FERROCENE-5d-330mg/kg,"1,4-DICHLORO-1d-735mg/kg",PIROXICAM-3d-14mg/kg,METOPROLOL-3d-1200mg/kg,LEVOFLOXACIN-5d-739mg/kg,ALLYL ALCOHO-5d-32mg/kg,STREPTOZOTOC-1d-138mg/kg,AMIODARONE-5d-147mg/kg,HYDROQUINONE-3d-800mg/kg,ROLIPRAM-1d-75mg/kg,ENROFLOXACIN-3d-2000mg/kg,PHENYLBUTAZO-5d-368mg/kg,INTERLEUKIN -1d-.125mg/kg,PROCARBAZINE-1d-27mg/kg,NIMESULIDE-5d-30mg/kg,SULFANILAMID-1d-636mg/kg,CERIVASTATIN-.25d-.05mg/kg,GLICLAZIDE-1d-1500mg/kg,ETOPOSIDE-3d-100mg/kg,AMITRIPTYLIN-3d-160mg/kg,INDINAVIR-1d-1000mg/kg,PICLAMILAST-1d-17mg/kg,PROBENECID-.25d-600mg/kg,HYDRAZINE-3d-45mg/kg,PROCARBAZINE-5d-27mg/kg,LEAD(IV) ACE-5d-200mg/kg,VINORELBINE-5d-1.5mg/kg,CHLORAMBUCIL-.25d-4.5mg/kg,CARMUSTINE-3d-16mg/kg,GENISTEIN-3d-375mg/kg,CLOFIBRIC AC-5d-448mg/kg,ETOPOSIDE-1d-100mg/kg,MEXILETINE-3d-350mg/kg,PRAZIQUANTEL-5d-1200mg/kg,VALPROIC ACI-3d-1500mg/kg,CARBACHOL-.25d-15mg/kg,BENZOTHIAZYL-3d-2600mg/kg,ATROPINE-1d-2.3mg/kg,ATROPINE-3d-2.3mg/kg,ETIDRONATE-1d-1005mg/kg,CYCLOSPORIN -1d-70mg/kg,SILDENAFIL-.25d-2.5mg/kg,CLOZAPINE-3d-95mg/kg,BUPROPION-3d-895mg/kg,VALPROIC ACI-3d-1340mg/kg,VALPROIC ACI-5d-1340mg/kg
LI_RG230_Osgepl1_P020517,0.00,-0.12,0.06,0.13,0.11,0.03,-0.09,0.03,0.02,0.12,-0.07,-0.06,0.08,-0.14,0.08,0.00,-0.14,0.01,0.01,0.15,0.01,-0.10,-0.12,0.08,-0.02,-0.06,-0.04,0.02,-0.07,0.05,-0.02,-0.05,0.06,0.11,0.17,-0.05,0.11,0.03,0.15,0.04,0.01,0.28,0.04,-0.12,0.06,-0.01,0.00,0.17,0.02,0.04
LI_RG230_Cyp11b1_P005901,0.00,-0.41,0.20,0.16,-0.18,-0.07,-0.44,-0.03,-0.20,-0.01,0.50,-0.20,0.23,-0.35,0.01,-0.04,0.14,0.33,-0.19,0.09,0.15,0.14,0.12,0.23,0.19,-0.20,-0.15,0.41,0.16,-0.02,-0.32,-0.01,0.20,-0.13,0.06,-0.07,0.07,-0.26,-0.05,0.06,-0.14,0.06,-0.08,0.23,0.03,0.23,0.33,0.03,0.40,0.73
LI_RG230_Hoxc10_P013913,-0.01,-0.22,-0.13,-0.13,0.01,0.07,-0.14,0.07,-0.36,-0.18,-0.39,-0.08,-0.08,-0.26,-0.20,-0.04,-0.24,-0.19,-0.09,-0.59,0.12,-0.14,-0.06,-0.17,-0.17,-0.32,-0.12,-0.26,0.07,-0.34,0.26,-0.31,-0.36,-0.42,0.15,0.07,-0.25,-0.18,-0.06,-0.02,-0.12,-0.15,-0.11,-0.66,0.22,-0.06,-0.16,-0.01,0.09,0.16
LI_RG230_Slc5a11_P027251,0.01,0.14,0.33,0.00,0.25,0.21,0.28,-0.25,0.28,0.49,0.17,0.39,0.20,-0.15,0.91,-1.10,0.01,-0.11,-0.19,0.28,-0.05,0.65,-0.26,0.26,0.09,-0.12,-0.08,0.18,0.14,-0.03,0.10,0.47,-0.09,-0.34,0.03,-0.32,0.91,-0.12,-0.71,-0.01,0.17,-0.27,-0.15,0.96,0.12,-0.46,-0.14,-0.06,0.24,0.33
LI_RG230_Vps53_predicted_P032985,0.01,0.03,-0.01,-0.00,-0.00,-0.02,-0.09,0.02,0.04,-0.03,-0.07,0.00,0.04,0.03,0.12,-0.06,-0.04,-0.03,0.01,0.03,0.04,-0.01,-0.05,0.09,0.05,-0.00,0.03,-0.02,0.00,0.02,0.04,0.07,0.03,-0.03,0.02,0.03,0.05,0.04,0.04,-0.09,0.07,0.02,-0.03,-0.11,0.03,-0.10,-0.03,0.03,0.01,0.01
LI_RG230__P009875,-0.08,0.08,-0.16,-0.07,0.18,0.13,0.01,-0.07,-0.09,-0.05,-0.18,0.19,0.32,0.14,0.39,-0.21,0.18,0.32,0.37,0.35,-0.21,0.48,-0.18,0.31,-0.07,-0.04,-0.01,-0.34,0.04,-0.11,0.08,0.20,0.29,-0.14,-0.04,-0.14,-0.72,0.17,0.45,0.11,-0.16,-0.09,0.21,0.23,0.10,0.13,0.13,0.32,-0.32,-0.18
LI_RG230_Nck2_predicted_P019937,0.08,-0.20,0.14,-0.00,-0.03,0.04,0.17,0.02,-0.10,-0.02,-0.09,-0.10,-0.20,-0.14,-0.05,-0.01,0.05,-0.03,-0.09,-0.16,-0.14,0.09,-0.05,0.07,0.16,-0.09,0.00,-0.07,-0.43,0.01,-0.11,-0.09,0.05,0.01,0.01,-0.02,0.06,-0.01,-0.07,0.01,0.06,-0.07,-0.15,-0.52,0.07,-0.47,-0.05,-0.13,-0.34,0.12
LI_RG230_Ssr2_predicted_P026704,-0.01,-0.00,-0.01,-0.01,-0.02,0.01,0.05,0.06,-0.02,-0.02,-0.13,-0.07,0.07,0.07,0.07,-0.07,0.07,0.01,-0.05,0.04,-0.15,0.03,-0.07,-0.05,0.02,-0.07,-0.09,0.14,0.02,-0.03,-0.15,0.00,-0.01,0.19,-0.07,-0.04,-0.02,0.13,0.04,-0.00,-0.04,-0.06,0.01,0.21,0.32,-0.14,0.08,0.01,-0.07,0.12
LI_RG230_Mtch2_predicted_P018351,-0.02,-0.17,-0.02,0.01,0.07,-0.03,-0.10,-0.08,0.00,-0.18,0.10,0.10,-0.01,-0.12,-0.02,0.07,-0.06,0.01,-0.01,0.02,0.17,-0.05,0.02,-0.03,-0.00,-0.08,-0.01,-0.01,0.03,0.00,-0.03,-0.06,-0.20,0.05,0.07,-0.05,-0.03,-0.04,-0.09,-0.04,0.03,-0.04,0.02,-0.24,-0.06,0.01,-0.00,-0.11,0.00,-0.06
LI_RG230__P030209,0.04,-0.05,-0.13,-0.07,-0.09,-0.08,0.08,-0.18,-0.03,-0.12,-0.03,0.02,-0.06,0.12,-0.09,-0.01,0.02,0.10,-0.00,0.02,0.17,-0.01,-0.13,-0.03,-0.02,-0.12,-0.02,-0.07,0.15,0.26,0.08,-0.02,0.03,-0.02,-0.08,-0.08,0.04,0.01,-0.09,0.02,-0.11,-0.09,-0.19,-0.30,-0.19,0.03,-0.02,-0.09,0.04,0.08
LI_RG230__P029648,-0.01,-0.08,-0.18,0.13,0.10,-0.05,-0.18,0.21,-0.12,-0.11,-0.12,-0.17,0.28,0.16,0.26,0.28,0.13,0.37,-0.09,-0.11,0.28,0.28,0.04,0.34,-0.07,0.15,-0.04,-0.33,-0.13,-0.04,0.12,-0.53,0.18,-0.03,-0.28,-0.28,0.03,-0.19,-0.08,-0.01,-0.07,-0.10,0.16,-0.55,-0.09,-0.03,0.05,0.01,-0.01,0.25
Matrix completion recovers a matrix M of size n1 ×n2 from a subset of its entries. Low rank matrix completion is well-studied with many theoretic results and practical applications (including the famous Netflix challenge). Matrix completion algorithms minimize the rank, the nuclear norm, or the Frobenius norm. Beyond plain completion, in real-world applications, oftentimes additional information is associated with the rows and columns. Take the Netflix challenge for example. The rows have associated user information, and the columns have associated movie information. Such information can be incorporated as additional features, and they in turn may be leveraged to improve completion performance or to predict for new rows or columns where there are no existing entries at all. The latter is often referred as the cold-start problem.
Torch (>1.0)
- Funk-SVD, we only use this to fetch the ratings. This is due to historical reason.
- PyTorch-Ignite
gpu=0
lr=1e-3
bs=128
epochs=350
for full in 0
do
for factors in 1000
do
output=output-bs-$bs-factors-$factors-epochs-$epochs-full-$full-$lr.txt
CUDA_VISIBLE_DEVICES=$gpu python ./train.py --factors $factors --lr $lr --bs $bs --epochs $epochs --full $full > $output
done
done