Extracting the Predicted Reults of Neural network in R
0
0
Entering edit mode
6 months ago

Hi, I'm new to machine learning. I have trained my data to separate my data as heart and eye. while testing it I cant able to extract the data which is predicted as eye and heart for example if my test data of row 1 is predicted as eye means I want all the data in the row of that. can anyone help me with it. I have given the code and my input data below

> library(neuralnet)
> library(readxl)
> Test_run <- read_excel("Test_run.xlsx")
> View(Test_run)                                                                                                                                   
> data =data.frame(Test_run)
> data$Heart=data$Disease=="Heart"
> data$Eye=data$Disease=="Eye"
> View(data)
> data.train.idx <- sample(x = nrow(data), size = nrow(data)*0.8)
> data.train <- data[data.train.idx,]
> data.valid <- data[-data.train.idx,]
> library(neuralnet)
> View(Test_run)
> View(data)
> data.net <- neuralnet(Eye+Heart ~ SIFT_converted_rankscore + Polyphen2_HDIV_rankscore + LRT_converted_rankscore + MutationTaster_converted_rankscore + MutationAssessor_score + FATHMM_converted_rankscore + PROVEAN_converted_rankscore + VEST3_score + MetaSVM_rankscore + MetaLR_rankscore + MCAP_rankscore + CADD_phred + DANN_score + fathmmMKL_coding_score + GERP_RS + phyloP100way_vertebrate_rankscore + phyloP20way_mammalian_rankscore + phastCons100way_vertebrate_rankscore + phastCons20way_mammalian_rankscore + SiPhy_29way_logOdds_rankscore + CB_MA + CB_Z + ClosestCRX + Cornea_MA + Cornea_Z + Exons + Iris_MA + Iris_Z + Lens_MA + Lens_Z + ONH_MA + ONH_Z + ON_MA + ON_Z + PeakCount + PeakSum + RPE_Choroid_MA + RPE_Choroid_Z + Retina_MA + Retina_Z + Sclera_MA + Sclera_Z + TM_MA + TM_Z + adipose_fpkm + adrenal_fpkm + brain_fpkm + breast_fpkm + colon_fpkm + heart_fpkm + human_retina_sra_fpkm + kidney_fpkm + liver_fpkm + lung_fpkm + lymph_node_fpkm + maxLength + ovary_fpkm + prostate_fpkm + skeletal_muscle_fpkm + testes_fpkm + thyroid_fpkm + white_blood_cells_fpkm + retDiff + Z_score, data=data.train, hidden=c(15,10), rep = 15, err.fct = "ce", linear.output = F, lifesign = "minimal", stepmax = 1000000, threshold = 0.001)
> plotdata.net, rep="best")
> data.prediction <- computedata.net, data.valid[-65:-67])
> idx <- apply(data.prediction$net.result, 1, which.max)
> predicted <- c('Eye', 'Heart')[idx]
> table(predicted,data.valid$Disease)

Output

predicted Eye Heart
        Eye    54     1
        Heart   0    22

Input file https://drive.google.com/file/d/1SmucxdaJklQxPejMtxLomqpOA7K2uQdp/view?usp=sharing

R Neural-network NGS expression • 158 views
ADD COMMENT

Login before adding your answer.

Traffic: 1810 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6