Update for most recent versions of NamedGraphs and DataGraphs

Author: mtfishman

.github/workflows/CI.yml

@@ -39,7 +39,7 @@ jobs:
         shell: julia --project=dev {0}
         run: |
           using Pkg;
-          pkg"dev https://github.com/mtfishman/DataGraphs.jl.git ."
+          pkg"dev ."
       - name: Run the tests
         shell: julia --project=dev {0}
         run: |

Project.toml

@@ -1,30 +1,35 @@
 name = "ITensorNetworks"
 uuid = "2919e153-833c-4bdc-8836-1ea460a35fc7"
 authors = ["Matthew Fishman <mfishman@flatironinstitute.org> and contributors"]
-version = "0.1.0"
+version = "0.2.0"
 
 [deps]
+Compat = "34da2185-b29b-5c13-b0c7-acf172513d20"
 DataGraphs = "b5a273c3-7e6c-41f6-98bd-8d7f1525a36a"
 Dictionaries = "85a47980-9c8c-11e8-2b9f-f7ca1fa99fb4"
 DocStringExtensions = "ffbed154-4ef7-542d-bbb7-c09d3a79fcae"
 Graphs = "86223c79-3864-5bf0-83f7-82e725a168b6"
 ITensors = "9136182c-28ba-11e9-034c-db9fb085ebd5"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
-MultiDimDictionaries = "87ff4268-a46e-478f-b30a-76b83dd64e3c"
 NamedGraphs = "678767b0-92e7-4007-89e4-4527a8725b19"
 Requires = "ae029012-a4dd-5104-9daa-d747884805df"
+SimpleTraits = "699a6c99-e7fa-54fc-8d76-47d257e15c1d"
 SplitApplyCombine = "03a91e81-4c3e-53e1-a0a4-9c0c8f19dd66"
 Suppressor = "fd094767-a336-5f1f-9728-57cf17d0bbfb"
 
 [compat]
+Compat = "3, 4"
+DataGraphs = "0.1.1"
 Dictionaries = "0.3.15"
-Graphs = "1.6.0"
+DocStringExtensions = "0.8, 0.9"
+Graphs = "1.6"
 ITensors = "0.3"
-MultiDimDictionaries = "0.0.1"
-NamedGraphs = "0.0.1"
-Requires = "1.3.0"
-Suppressor = "0.2.1"
-julia = "1.6"
+NamedGraphs = "0.1.3"
+Requires = "1.3"
+SimpleTraits = "0.9"
+SplitApplyCombine = "1.2"
+Suppressor = "0.2"
+julia = "1.7"
 
 [extras]
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"

README.md

@@ -4,28 +4,25 @@
 
 #' ## Installation
 #'
-#' This package relies on a few unregistered packages. To install, you will need to do:
-#'
+#' You can install this package through the Julia package manager:
 #' ```julia
-#' julia> using Pkg
-#' 
-#' julia> Pkg.add(url="https://github.com/mtfishman/MultiDimDictionaries.jl")
-#' 
-#' julia> Pkg.add(url="https://github.com/mtfishman/NamedGraphs.jl")
-#' 
-#' julia> Pkg.add(url="https://github.com/mtfishman/DataGraphs.jl")
-#' 
-#' julia> Pkg.add(url="https://github.com/mtfishman/ITensorNetworks.jl")
+#' julia> ] add ITensorNetworks
 #' ```
 
+#+ echo=false; term=false
+
+using Random
+using ITensors
+Random.seed!(ITensors.index_id_rng(), 1234);
+
 #' ## Examples
 #'
 #' Here are is an example of making a tensor network on a chain graph (a tensor train or matrix product state):
 #+ term=true
 
 using ITensors
 using ITensorNetworks
-tn = ITensorNetwork(named_grid((4,)); link_space=2)
+tn = ITensorNetwork(named_grid(4); link_space=2)
 tn[1]
 tn[2]
 neighbors(tn, 1)
@@ -38,16 +35,16 @@ neighbors(tn, 4)
 
 tn = ITensorNetwork(named_grid((2, 2)); link_space=2)
 tn[1, 1]
-neighbors(tn, 1, 1)
-neighbors(tn, 1, 2)
-tn_1 = tn[1, :]
-tn_2 = tn[2, :]
+neighbors(tn, (1, 1))
+neighbors(tn, (1, 2))
+tn_1 = subgraph(v -> v[1] == 1, tn)
+tn_2 = subgraph(v -> v[1] == 2, tn)
 
 #' Networks can also be merged/unioned:
 #+ term=true
 
 using ITensorUnicodePlots
-s = siteinds("S=1/2", named_grid((3,)))
+s = siteinds("S=1/2", named_grid(3))
 tn1 = ITensorNetwork(s; link_space=2)
 tn2 = ITensorNetwork(s; link_space=2)
 @visualize tn1;
@@ -59,9 +56,9 @@ Z̃ = contract(Z, (1, 1) => (2, 1));
 @visualize Z̃;
 
 #' ## Generating this README
-#'
+
 #' This file was generated with [weave.jl](https://github.com/JunoLab/Weave.jl) with the following commands:
-#' ```julia
-#' using ITensorNetworks, Weave
-#' weave(joinpath(pkgdir(ITensorNetworks), "examples", "README.jl"); doctype="github", out_path=pkgdir(ITensorNetworks))
-#' ```
+#+ eval=false
+
+using ITensorNetworks, Weave
+weave(joinpath(pkgdir(ITensorNetworks), "examples", "README.jl"); doctype="github", out_path=pkgdir(ITensorNetworks))

examples/README.jl

@@ -1,50 +1,51 @@
+using Compat
 using ITensors
+using Metis
 using ITensorNetworks
+
 using ITensorNetworks:
-  delta_network,
-  formsubgraphs,
-  partition,
-  flatten_thicken_bonds,
-  flattened_inner_network,
   construct_initial_mts,
   update_all_mts,
-  get_single_site_expec,
-  iterate_single_site_expec,
-  contract_boundary_mps
-using KaHyPar
-using Compat
+  get_single_site_expec
 
-#nxn GRID
 n = 4
-g = named_grid((n, n))
+dims = (n, n)
+g = named_grid(dims)
+# g = named_comb_tree(dims)
 s = siteinds("S=1/2", g)
-chi = 3
+chi = 2
 
-#Random Tensor Network, Flatten it too
-psi = randomITensorNetwork(s; link_space=chi)
-psiflat, combiners = flatten_thicken_bonds(deepcopy(psi))
+ψ = randomITensorNetwork(s; link_space=chi)
 
-v = (1, 1)
+ψψ = norm_sqr_network(ψ; flatten=true, map_bra_linkinds=prime)
+combiners = linkinds_combiners(ψψ)
+ψψ = combine_linkinds(ψψ, combiners)
 
-#Apply Sz to site v and flatten that
-psiflatO = flatten_thicken_bonds(psi; ops=["Sz"], vops=[v], s=s, combiners=combiners)
+# Apply Sz to site v
+v = one.(dims)
+Oψ = copy(ψ)
+Oψ[v] = apply(op("Sz", s[v]), ψ[v])
+ψOψ = inner_network(ψ, Oψ; flatten=true, map_bra_linkinds=prime)
+ψOψ = combine_linkinds(ψOψ, combiners)
 
-#Get the value of sz on v via exact contraction
-actual_sz = ITensors.contract(psiflatO)[1] / ITensors.contract(psiflat)[1]
+# Get the value of sz on v via exact contraction
+contract_seq = contraction_sequence(ψψ)
+actual_sz = contract(ψOψ; sequence=contract_seq)[] / contract(ψψ; sequence=contract_seq)[]
 
 println("Actual value of Sz on site " * string(v) * " is " * string(actual_sz))
 
-nsites = 1
-println("First " * string(nsites) * " sites form a subgraph")
-dg_subgraphs = formsubgraphs(g, Int(n * n / nsites))
-mts = construct_initial_mts(psiflat, dg_subgraphs; init=(I...) -> allequal(I) ? 1 : 0)
-niters = 5
+niters = 20
 
-iterate_single_site_expec(psiflat, psiflatO, mts, dg_subgraphs, niters, v)
+nsites = 1
+println("\nFirst " * string(nsites) * " sites form a subgraph")
+mts = construct_initial_mts(ψψ, nsites; init=(I...) -> @compat allequal(I) ? 1 : 0)
+@show get_single_site_expec(ψψ, mts, ψOψ, v)
+mts = update_all_mts(ψψ, mts, niters)
+@show get_single_site_expec(ψψ, mts, ψOψ, v)
 
 nsites = 4
-println("Now " * string(nsites) * " sites form a subgraph")
-dg_subgraphs = formsubgraphs(g, Int(n * n / nsites))
-mts = construct_initial_mts(psiflat, dg_subgraphs; init=(I...) -> allequal(I) ? 1 : 0)
-
-iterate_single_site_expec(psiflat, psiflatO, mts, dg_subgraphs, niters, v)
+println("\nNow " * string(nsites) * " sites form a subgraph")
+mts = construct_initial_mts(ψψ, nsites; init=(I...) -> @compat allequal(I) ? 1 : 0)
+@show get_single_site_expec(ψψ, mts, ψOψ, v)
+mts = update_all_mts(ψψ, mts, niters)
+@show get_single_site_expec(ψψ, mts, ψOψ, v)

examples/TTN/ttn_basics.jl

@@ -1,3 +1,4 @@
+using NamedGraphs
 using ITensors
 using ITensorNetworks
 using Random
@@ -13,7 +14,7 @@ s = siteinds("S=1/2", g)
 χ = 10
 ψ = randomITensorNetwork(s; link_space=χ)
 
-tn = norm_network(ψ)
+tn = norm_sqr_network(ψ)
 
 # Contraction sequence for exactly computing expectation values
 # contract_edges = map(t -> (1, t...), collect(keys(cartesian_to_linear(dims))))
@@ -26,7 +27,7 @@ using OMEinsumContractionOrders
 
 println("greedy")
 seq_greedy = @time contraction_sequence(tn; alg="greedy")
-res_greedy = @time contract(tn; alg=res_greedy)
+res_greedy = @time contract(tn; sequence=seq_greedy)
 
 println("tree_sa")
 seq_tree_sa = @time contraction_sequence(tn; alg="tree_sa")

examples/belief_propagation/bpexample.jl

@@ -1,11 +1,11 @@
 using ITensors
 using ITensorNetworks
 using ITensorUnicodePlots
+using NamedGraphs
 
 χ, d = 5, 2
 dims = (4, 4)
-vertices = vec([(i, j) for i in 1:dims[1], j in 1:dims[2]])
-g = NamedDimGraph(grid(dims), vertices)
+g = named_grid(dims)
 
 # Network of indices
 is = IndsNetwork(g; link_space=χ, site_space=d)

examples/contraction_sequence/contraction_sequence.jl

@@ -25,7 +25,7 @@ e = edge_data(is)
 
 @visualize ψ̃
 
-ψψ = ⊗(ψ̃, ψ; new_dim_names=("bra", "ket"))
+ψψ = ("bra" => ψ̃) ⊗ ("ket" => ψ)
 
 @visualize ψψ
 
@@ -38,6 +38,6 @@ inner_res = contract(ψψ; sequence)[]
 
 @show inner_res
 
-sub = ψψ[[("bra", 1), ("ket", 1), ("bra", 2), ("ket", 2)]]
+sub = subgraph(ψψ, [(1, "bra"), (1, "ket"), (2, "bra"), (2, "ket")])
 
 @visualize sub

examples/examples.jl

@@ -1,12 +1,12 @@
 using Graphs
-using ITensorNetworks
 using KaHyPar
 using Metis
+using ITensorNetworks
 
 g = grid((16,))
-npartitions = 2
+npartitions = 4
 
 kahypar_partitions = partition(g, npartitions; backend="KaHyPar")
 metis_partitions = partition(g, npartitions; backend="Metis")
-@show kahypar_partitions
-@show metis_partitions
+@show kahypar_partitions, length(unique(kahypar_partitions))
+@show metis_partitions, length(unique(metis_partitions))