I want to execute a 100x loop, get rownames from a dataframe A and add its rownames as values in a column to a new dataframe B each time, ending with 100 columns in B. I am beating my head against the wall here trying to do this, as I seem to overwrite B every way I have tried it.
data <- bs.exp.sorted.1198.input
outsft = dataframe
outsamps = dataframe
samples<- data.frame(matrix(0, nrow = 60, ncol = 1))
currentsamps =dataframe
itnumb=100
for(i in 1:itnumb){
subsamp <- data[sample(nrow(data), 60), ]
powers =c(c(1:10), seq(from=12,to=20,by=2))
sft =pickSoftThreshold(bs.exp.sorted.1198.input, powerVector = powers, verbose=5)
outsft <- rbind(outsft, sft$fitIndices[,2])
currentsamps <- as.data.frame(rownames(subsamp))
output <- add_column(samples, currentsamps)
}
output always contains a column of 0s (which I had to include to get it to work... And whatever the last set of rownames were in the loop. I want to save ALL the rownames in one big dataframe. This should be a simple task, but it has me stumped, and at this point, infuriated. Any suggestions? Thanks.
How about this:
initial_rows = 100
initial_df = data.frame( col1 = rnorm( initial_rows ),
row.names = 1:initial_rows )
stores_names = rownames( initial_df )
chunk_size = 10
n_repetitions = 10
result = data.frame( matrix( NA_character_,
nrow = chunk_size, ncol = n_repetitions ) )
for ( i_rep in seq_len(n_repetitions) ) {
current_names = sample( stores_names, chunk_size )
result[ , i_rep ] = current_names
}
> result
X1 X2 X3 X4 X5 X6 X7 X8 X9 X10
1 72 62 24 35 80 48 6 79 17 96
2 66 4 59 39 32 67 95 96 89 26
3 34 21 96 75 22 87 8 49 98 50
4 40 33 74 90 39 2 86 78 15 3
5 87 12 70 72 62 98 69 4 1 32
6 98 43 7 38 18 57 83 47 35 41
7 28 70 6 84 30 4 54 46 4 35
8 94 100 80 36 84 7 56 36 73 9
9 60 88 28 11 73 32 3 23 85 67
10 69 69 68 69 4 17 60 63 62 47
Is this what you wanted?