Issue with Personalized recommenders

[George3d6]

Running recommend on any personal recommender yields the following non-descriptive error:

ERROR: LoadError: BoundsError: attempt to access 16×16 Array{Float64,2} at index [0, Base.Slice(Base.OneTo(16))]
Stacktrace:
 [1] throw_boundserror(::Array{Float64,2}, ::Tuple{Int64,Base.Slice{Base.OneTo{Int64}}}) at ./abstractarray.jl:434
 [2] checkbounds at ./abstractarray.jl:362 [inlined]
 [3] macro expansion at ./multidimensional.jl:494 [inlined]
 [4] _getindex at ./multidimensional.jl:491 [inlined]
 [5] getindex(::Array{Float64,2}, ::Int64, ::Colon) at ./abstractarray.jl:883
 [6] predict(::Recommendation.UserKNN, ::Int64, ::Int64) at /home/george/.julia/v0.6/Recommendation/src/model/user_knn.jl:51
 [7] recommend(::Recommendation.UserKNN, ::Int64, ::Int64, ::Array{Int64,1}) at /home/george/.julia/v0.6/Recommendation/src/base_recommender.jl:19
 [8] macro expansion at /home/george/forecasting/individual_analysis/recommender.jl:42 [inlined]
 [9] anonymous at ./<missing>:?

The code was basically this:

da = DataAccessor(events, length(accounts), length(itemns))
recommender = UserKNN(da)
build(recommender)
...
for acc_index in accounts
    recommendations = recommend(recommender, acc_index, nr_recc, items_indexes)
end

The Baseline recommenders work just fine.

[takuti]

@George3d6 Sorry for delay. First of all, could you check if minimum account and item indices are 1, not 0; In Julia, "index" is one-origin.

[takuti]

@George3d6 Closing issue for now. Feel free to reopen it if you have further problems.