# GaussianMixtures question
**URL:** https://discourse.julialang.org/t/gaussianmixtures-question/61526
**Category:** Statistics
**Created:** [May 20, 2021, 3:59pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526 "2021-05-20T15:59:49Z")
**Posts on this page:** 9
**Page:** 1
### Author:  [@SergeantMike67](https://discourse.julialang.org/u/SergeantMike67)
#### Post date: [May 20, 2021, 3:59pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/1 "2021-05-20T15:59:49Z")
I have a dataset that results from calculating the CpG observed over expected (an estimation of cytosine methylation in the genome) and I am trying to fit the data into a two mean, 1D Gaussian Mixture model using Gaussian Mixtures. Example input data below:
740736×2 Matrix{Any}:
“RDRX01000001” 0.251146
“RDRX01000002” 0.279088
“RDRX01000003” 0.428743
“RDRX01000004” 0.327771
“RDRX01000005” 0.350942
“RDRX01000006” 0.90064
“RDRX01000007” 0.267866
“RDRX01000008” 0.360507
“RDRX01000009” 0.600434
⋮
“RDRX01799874” 1.40488
“RDRX01799875” 1.65668
“RDRX01799876” 1.66154
“RDRX01799877” 1.04891
“RDRX01799878” 0.987179
“RDRX01799879” 1.29231
“RDRX01799880” 1.72998
“RDRX01799881” 1.08387
The histogram looks like:
I constructed the gmm with:
`g = GMM(2,1,kind=:full)`
and received this output, which I take as a success
GMM{Float64} with 2 components in 1 dimensions and full covariance
Mix 1: weight 0.500000
mean: [0.0]
covariance: 1×1 Matrix{Float64}:
1.0
Mix 2: weight 0.500000
mean: [0.0]
covariance: 1×1 Matrix{Float64}:
1.0
however when I run the training function:
`em!(g,x1)`
where x1 is the matrix of the above data
I get the following error:
ERROR: Inconsistent size gmm and x
Stacktrace:
[1] error(s::String)
@ Base .\error.jl:33
[2] em!(gmm::GMM{Float64, Vector{LinearAlgebra.UpperTriangular{Float64, Matrix{Float64}}}}, x::Matrix{Any}; nIter::Int64, varfloor::Float64, sparse::Int64, debug::Int64)
@ GaussianMixtures ~.julia\packages\GaussianMixtures\1pQcF\src\train.jl:238
[3] em!(gmm::GMM{Float64, Vector{LinearAlgebra.UpperTriangular{Float64, Matrix{Float64}}}}, x::Matrix{Any})
@ GaussianMixtures ~.julia\packages\GaussianMixtures\1pQcF\src\train.jl:238
[4] top-level scope
@ REPL[16]:1
What am I missing here?
---
### Author:  [@ElOceanografo](https://discourse.julialang.org/u/ElOceanografo)
#### Post date: [May 20, 2021, 4:28pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/2 "2021-05-20T16:28:24Z")
It looks like your input matrix is n x 2, and the first column is strings. Try it with `em!(g, x1[:, 2])`.
As an aside, if your data `x1` include both labels and continuous values, they should probably be stored in a `DataFrame`, not a matrix…
---
### Author:  [@SergeantMike67](https://discourse.julialang.org/u/SergeantMike67)
#### Post date: [May 20, 2021, 4:51pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/3 "2021-05-20T16:51:55Z")
I didn’t think GaussianMixtures worked with a dataframe. That is the original format of the data but I converted it to matrix format per the type shown on the Github page.
---
### Author:  [@SergeantMike67](https://discourse.julialang.org/u/SergeantMike67)
#### Post date: [May 20, 2021, 5:02pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/4 "2021-05-20T17:02:49Z")
ElOceanografo,
Thank you for the response but when I run the
`em! (g, x1[:, 2])`
or
`em! (g, graphdata[:, 2])`
where graphdata is the original DataFrame,
I get this error
em!(g, x1[:, 2])
ERROR: MethodError: no method matching em!(::GMM{Float64, Vector{LinearAlgebra.UpperTriangular{Float64, Matrix{Float64}}}}, ::Vector{Any})
Closest candidates are:
em!(::GMM, ::DataOrMatrix{T} where T; nIter, varfloor, sparse, debug) at C:\Users\Mikey.julia\packages\GaussianMixtures\1pQcF\src\train.jl:237
em!(::VGMM, ::Any; nIter) at C:\Users\Mikey.julia\packages\GaussianMixtures\1pQcF\src\bayes.jl:273
Stacktrace:
[1] top-level scope
@ REPL[34]:1
---
### Author:  [@ElOceanografo](https://discourse.julialang.org/u/ElOceanografo)
#### Post date: [May 20, 2021, 10:16pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/5 "2021-05-20T22:16:45Z")
I think this is because `em!` expects the data to be a 2D `Matrix`, even if it only has a single column. This works:
```julia
using DataFrames, GaussianMixtures
df = DataFrame(label = rand(["a", "b", "c", "d"], 200), x = [randn(50); randn(150).+4])
g = GMM(2,1,kind=:full)
g.μ[1] = 1.0
em!(g, reshape(df.x, :, 1))
```
Note that I changed one of the means before fitting it on the data–otherwise, both means seem to stay identical to each other and the EM algorithm never splits the data into two groups.
---
### Author:  [@cchderrick](https://discourse.julialang.org/u/cchderrick)
#### Post date: [May 20, 2021, 10:28pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/6 "2021-05-20T22:28:58Z")
> [@SergeantMike67](#):
>
> graphdata[:, 2]
Another way to index a matrix out instead of a 1-D vector is `graphdata[:, 2:2]`
or if using DataFrame `Matrix(df[!,2:2])` also seems to work.
or `Matrix(df[!,[:x]])`
---
### Author:  [@SergeantMike67](https://discourse.julialang.org/u/SergeantMike67)
#### Post date: [May 21, 2021, 2:02pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/7 "2021-05-21T14:02:59Z")
Thank you ElOceanographo, that worked. The calculated means seem to be off, but I that isn’t likely from the package.
---
### Author:  [@ElOceanografo](https://discourse.julialang.org/u/ElOceanografo)
#### Post date: [May 21, 2021, 6:06pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/8 "2021-05-21T18:06:12Z")
From the graph you posted, it looks like your mixture isn’t _that_ Gaussian (i.e., values all \> 0 and a fairly skewed histogram) so it isn’t surprising that the means are off from where they “should” be. If it’s worth the effort, you might be able to fit a custom non-Gaussian mixture model using Turing or something.
---
### Author:  [@SergeantMike67](https://discourse.julialang.org/u/SergeantMike67)
#### Post date: [May 21, 2021, 7:11pm UTC](https://discourse.julialang.org/t/gaussianmixtures-question/61526/9 "2021-05-21T19:11:20Z")
What I am actually looking for is two gaussian peaks, one \<=0.5 and and another \>1.0. This particular organism may have that as it looks like there is a peak at around 0.3-0.4 but I certainly could be wrong on my visual assessment.