Release v0.5.0

CHANGELOG.md

@@ -6,6 +6,47 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/) and this p
 
 Please see [here](https://github.com/hrntsm/Tunny/releases) for the data released for each version.
 
+## [0.5.0] -2022-09-03
+
+### Added
+
+- Hypervolume visualization
+  - It is useful for determining convergence in multi-objective optimization.
+- Clustering visualization
+  - Clustering of results during multi-objective optimization makes it easier to evaluate solutions.
+- BoTorch Sampler
+  - This sampler use Gaussian Process and support multi-objective optimization.
+- Quasi-MonteCarlo Sampler
+  - [Detail](https://optuna.readthedocs.io/en/latest/reference/samplers/generated/optuna.samplers.QMCSampler.html) about this sampler.
+- Support Constraint.
+  - Only TPE, GP, NSGAII can use constraint.
+- Sampler detail settings UI
+  - Previously it was necessary to change the JSON file of the settings, but now it can be changed in the UI
+- Enable Text Bake in the FishMarket component.
+- Allows selection of NSGA-II crossover methods.
+  - `Uniform`, `BLXAlpha`, `SPX`, `SBX`, `VSBX`, `UNDX`
+- Ability to set Popsize on CMA-ES restart
+- Run GC after trial when has geometry attribute or setting always run.
+  - **This change probably make optimize slower before**
+  - If you want to cut this setting, set the value of "GcAfterTrial" to 2 in Settings.json.
+- Show LICENSE button in Tunny UI.
+
+### Changed
+
+- When genepool is an input, it now creates variable names using nicknames.
+- The output of the Pareto solution was made to consider the constraints.
+- Multivariate in TPE sampler default option is false to true
+- Updated Optuna used to v3.0.0
+  - Use `suggest_int` and `suggest_float` instead of `suggest_uniform` for more accurate variable generation in optimization
+  - Random and Grid samplers now support multi-objective optimization
+  - The format of the db file in which the results are saved has changed. Please note that it is not compatible with the previous one.
+
+### Fixed
+
+- The PythonInstaller window now has no text on the progress bar.
+- When more than one Study exists, another Study Name is set and RunOpt no longer causes a Solver Error.
+- The error does not occur when the Brep of Geometry of Attribute is null.
+
 ## [0.4.0] -2022-07-09
 
 ### Added
@@ -54,7 +95,7 @@ Please see [here](https://github.com/hrntsm/Tunny/releases) for the data release
 
 - Use json to set sampler detail settings
   - Since the settings are saved in Json, the previous values set in the UI remain saved when the window is closed and reopened.
-- UI for above detail settings 
+- UI for above detail settings
 - If input "-10" in restore model number, Tunny output all result
 - Fish component
   - Special GH_Param component for Tunny result to handle its result easier.

README.md

@@ -65,6 +65,10 @@ Please see PYTHON_PACKAGE_LICENSE for more license information.
 
 https://user-images.githubusercontent.com/23289252/178105107-5e9dd9f7-5680-40d4-97b0-840a4f1f329c.mp4
 
+### :fish_cake: Select sampler flow chart
+
+<img width="75%" alt="image" src="https://user-images.githubusercontent.com/23289252/188254450-1e718d97-f81e-49a1-949b-e158837bc44f.png">
+
 ### :anchor: Component location
 
 Tunny can be found in the Tunny tab if it has been installed.
@@ -103,7 +107,10 @@ The nickname of the ConstructFishAttribute component input is stored paired with
 
 The Geometry input has a special meaning; what is entered here will be displayed as a Geometry when the results are sorted in the FishMarket component described below.
 
-<img width="40%" src="https://user-images.githubusercontent.com/23289252/178102569-16b64446-1a67-4eb6-80fa-a4c846c4b294.png">
+Constraint inputs are also special inputs.**(new in v0.5.0)** The values entered here are the constraints in the optimization.
+When this value is less than 0, the constraint is considered satisfied. Constraint conditions are supported by TPE, GP and NSGAII.
+
+<img width="60%" alt="image" src="https://user-images.githubusercontent.com/23289252/188254609-3c8432ba-3f1c-45f4-bd2a-9e3f08271c2b.png">
 
 ### :octopus: Other components
 
@@ -128,24 +135,26 @@ On the other hand, it is possible to save the learning status, and even after th
 On the other hand, Running the Dashboard function, which charts the results, starts a server that handles the results and allows you to see the optimization results in real time in your browser.  
 This feature is useful for analyzing post-optimization results, as it allows the user to not only see the results in real time, but also to view several figures at once.
 
-It is recommended that optimization be performed a small number of times, and after completion, the results should be reviewed to determine if continued optimization should be performed.
-
 #### :sailboat: Optimize Tab
 
-<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/178103175-02ade5f2-fa3e-4f34-9757-1b944d699785.png">
+<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/188254704-533139c7-169a-4aee-bcd5-0e98ff45f35e.png">
 
 Values that can be set and their meanings are as follows.
 
 - Sampler
   - Sets the algorithm to perform the optimization. The following types are available.
   - All are optimization algorithms provided by Optuna.
-    1. TPE (Bayesian optimization)
-    1. NSGA-II (Genetic algorithm)
-    1. CMA-ES (Evolution strategy)
-    1. Random
-    1. Grid
+    1. Bayesian optimization(TPE)
+    2. Bayesian optimization(GP) **(new in v0.5.0)**
+    3. Genetic algorithm(NSGA-II)
+    4. Evolution strategy(CMA-ES)
+    5. Quasi-MonteCarlo **(new in v0.5.0)**
+    6. Random
+    7. Grid
 - Number of trial
   - This number of trials will be performed.
+    - Bayesian optimization first performs random sampling to create a surrogate model. This is the setting for how many trials random sampling is performed.
+    - The original paper recommends “number of variables” \* 11-1.
   - If the grid sampler is selected, the calculation is performed by dividing each entered Variable by this number.
     - **Note** that the number of calculations is (Number of trial) to the power of (Number of Variable).
 - Timeout(sec)
@@ -163,7 +172,7 @@ Values that can be set and their meanings are as follows.
 
 #### :boat: Visualize Tab
 
-<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/178103288-6e7bc0fe-f730-4e23-8159-6d43df55f1ae.png">
+<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/188255757-8532948f-b784-45b7-b116-17ba79db9536.png">
 
 Values that can be set and their meanings are as follows.
 
@@ -181,12 +190,15 @@ Values that can be set and their meanings are as follows.
     6. param importance
     7. pareto front
     8. slice
-- Show selected type of plots
-  - Show the plot selected in Visualize type above.
+    9. hypervolume **(new in v0.5.0)**
+  - Show selected type of plots
+    - Show the plot selected in Visualize type above.
+- k-means clustering
+  - Cluster the results of the Pareto front when performing multi-objective optimization.Any number of clusters can be specified.
 
 #### :fishing_pole_and_fish: Output Tab
 
-<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/178103410-e2a589b2-ffd0-436e-b4b1-2deb4e7346ba.png">
+<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/188255933-62df2324-29b0-40e2-9568-6b825ed147dd.png">
 
 Values that can be set and their meanings are as follows.
 
@@ -197,11 +209,8 @@ Values that can be set and their meanings are as follows.
 - All trials
   - Output all trials.
 - Use model number
-
   - Multiple values can be entered separated by commas, such as "1,3,42".
-
 - Output
-
   - The model with the number entered here is restored from the optimization results file and is the output of the component.
   - The model number can be found on each plot.
   - In the example below, model number 424 with a value of -11.49 for the first objective function is selected.
@@ -213,36 +222,28 @@ Values that can be set and their meanings are as follows.
 
 #### :droplet: Settings Tab
 
-<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/178103758-3ca8fa88-4796-4b0c-8c17-f8a810a47caf.png">
-
-Tunny stores the optimize settings in json.  
-Detailed settings in optimization can now also be configured in Json.
-
-- **IMPORTANT**: The default hyperparameter for each optimization contain Optuna defaults.
-  - These are not necessarily the best values for your optimization.
-  - Pay particular attention to the initial population or trial.
-    - the recommended initial trial for TPE is "the number of Variable" x 11 - 1.
-- Open API page
-  - Open Optuna's API page with the meaning of each value in the advanced optimization settings.
-- Save settings to json
-  - Save the settings to Json.
-  - Settings are also saved when the window is closed with the X button.
-- Load settings from json
-  - Loads a settings file.
-- Open Settings.Json folder
-  - Open the folder where the settings files are stored.
-  - The file Settings.json is Tunny's settings file. Edit it with any text editor.
+<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/188255969-9355c626-36d3-4d9b-8f50-0a8bda79bf8e.png">
+
+**(new in v0.5.0)**
+Allows detailed optimization settings to be performed in the UI.
+See below for the meaning of each setting.
+
+[Guideline for selecting optimization algorithms](https://graceful-stag-dae.notion.site/Guidelines-for-Selecting-a-Optimization-Algorithm-8505b2e020ee4af2a77272f25acc6094)
 
 #### :turtle: File Tab
 
-<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/178103993-d4448d03-5aef-45bd-a8a0-497a0cf417a4.png">
+<img width="30%" alt="image" src="https://user-images.githubusercontent.com/23289252/188256059-82679c52-4909-4e29-a9ad-36b3bab5c577.png">
+
+- Result
+  - Open result file folder
+    - Open the folder where the file containing the optimization results is located.
+    - Results are saved as "Tunny_Opt_Result.db" by default.
+    - This can be made to be anything you want by rewriting the Setting.json file.
+  - Clear result file
+    - Deletes the optimization result file.
+- License **(new in v0.5.0)**
+  - You can check license to push each button.
 
-- Open result file folder
-  - Open the folder where the file containing the optimization results is located.
-  - Results are saved as "Tunny_Opt_Result.db" by default.
-  - This can be made to be anything you want by rewriting the Setting.json file.
-- Clear result file
-  - Deletes the optimization result file.
 
 ## :surfer: Contact information
 

Tunny/Component/ConstructFishAttribute.cs

@@ -12,7 +12,8 @@ namespace Tunny.Component
 {
     public class ConstructFishAttribute : GH_Component, IGH_VariableParameterComponent
     {
-        private readonly string _geomDescription = "Connect model geometries here. Not required. Large size models are not recommended as it affects the speed of analysis.\"Geometry\" is a special nickname that is visualized with the optimization results. Do not change it.";
+        private readonly string _geomDescription = "Connect model geometries here. Not required. Large size models are not recommended as it affects the speed of analysis.";
+        private readonly string _constraintDescription = "A value strictly larger than 0 means that a constraints is violated. A value equal to or smaller than 0 is considered feasible. ";
         private readonly string _attrDescription = "Attributes to each trial. Attribute name will be the nickname of the input, so change it to any value.";
         public override GH_Exposure Exposure => GH_Exposure.secondary;
 
@@ -26,11 +27,13 @@ public ConstructFishAttribute()
         protected override void RegisterInputParams(GH_InputParamManager pManager)
         {
             pManager.AddGeometryParameter("Geometry", "Geometry", _geomDescription, GH_ParamAccess.list);
+            pManager.AddNumberParameter("Constraint", "Constraint", _constraintDescription, GH_ParamAccess.item);
             pManager.AddGenericParameter("Attr1", "Attr1", _attrDescription, GH_ParamAccess.list);
             pManager.AddGenericParameter("Attr2", "Attr2", _attrDescription, GH_ParamAccess.list);
             Params.Input[0].Optional = true;
             Params.Input[1].Optional = true;
             Params.Input[2].Optional = true;
+            Params.Input[3].Optional = true;
         }
 
         protected override void RegisterOutputParams(GH_OutputParamManager pManager)
@@ -49,18 +52,33 @@ protected override void SolveInstance(IGH_DataAccess DA)
                 return;
             }
 
+            GetInputData(DA, paramCount, dict);
+            DA.SetData(0, dict);
+        }
+
+        private void GetInputData(IGH_DataAccess DA, int paramCount, Dictionary<string, object> dict)
+        {
             for (int i = 0; i < paramCount; i++)
             {
-                var list = new List<object>();
-                if (!DA.GetDataList(i, list))
+                string key = Params.Input[i].NickName;
+                if (i == 1)
                 {
-                    continue;
+                    double constraint = 0;
+                    if (DA.GetData(i, ref constraint))
+                    {
+                        dict.Add(key, constraint);
+                    }
+                }
+                else
+                {
+                    var list = new List<object>();
+                    if (!DA.GetDataList(i, list))
+                    {
+                        continue;
+                    }
+                    dict.Add(key, list);
                 }
-                string key = Params.Input[i].NickName;
-                dict.Add(key, list);
             }
-
-            DA.SetData(0, dict);
         }
 
         //FIXME: Should be modified to capture and check for change events.
@@ -79,13 +97,19 @@ private bool CheckIsNicknameDuplicated()
             return false;
         }
 
-        public bool CanInsertParameter(GH_ParameterSide side, int index) => side != GH_ParameterSide.Output && (Params.Input.Count == 0 || index != 0);
+        public bool CanInsertParameter(GH_ParameterSide side, int index) => side != GH_ParameterSide.Output && (Params.Input.Count == 0 || index >= 2);
 
-        public bool CanRemoveParameter(GH_ParameterSide side, int index) => side != GH_ParameterSide.Output && index != 0;
+        public bool CanRemoveParameter(GH_ParameterSide side, int index) => side != GH_ParameterSide.Output && index >= 2;
 
         public IGH_Param CreateParameter(GH_ParameterSide side, int index)
         {
-            return side == GH_ParameterSide.Input ? index == 0 ? SetGeometryParameterInput() : SetGenericParameterInput(index) : null;
+            return side == GH_ParameterSide.Input
+                ? index == 0
+                    ? SetGeometryParameterInput()
+                    : index == 1
+                        ? SetNumberParameterInput()
+                        : SetGenericParameterInput(index)
+                : null;
         }
 
         private IGH_Param SetGenericParameterInput(int index)
@@ -98,12 +122,23 @@ private IGH_Param SetGenericParameterInput(int index)
             return p;
         }
 
+        private IGH_Param SetNumberParameterInput()
+        {
+            var p = new Param_Number();
+            p.Name = p.NickName = $"Constraint";
+            p.Description = _constraintDescription;
+            p.Access = GH_ParamAccess.list;
+            p.MutableNickName = false;
+            p.Optional = true;
+            return p;
+        }
+
         private IGH_Param SetGeometryParameterInput()
         {
             var p = new Param_Geometry();
             p.Name = p.NickName = "Geometry";
             p.Description = _geomDescription;
-            p.Access = GH_ParamAccess.list;
+            p.Access = GH_ParamAccess.item;
             p.MutableNickName = false;
             p.Optional = true;
             return p;

Tunny/Component/FishMarket.cs

@@ -2,11 +2,13 @@
 using System.Collections.Generic;
 using System.Drawing;
 using System.Linq;
+using System.Windows.Forms;
 
 using Grasshopper.Kernel;
 using Grasshopper.Kernel.Data;
 using Grasshopper.Kernel.Types;
 
+using Rhino;
 using Rhino.Display;
 using Rhino.Geometry;
 
@@ -171,6 +173,31 @@ public override void DrawViewportWires(IGH_PreviewArgs args)
             }
         }
 
+        public override void BakeGeometry(RhinoDoc doc, List<Guid> obj_ids)
+        {
+            base.BakeGeometry(doc, obj_ids);
+
+            for (int i = 0; i < _fishes.Count; i++)
+            {
+                doc.Objects.AddText(_fishes[i].ToString(), _tagPlanes[i], _size, "Meiryo", false, false, TextJustification.TopLeft);
+            }
+        }
+
+        public override void AppendAdditionalMenuItems(ToolStripDropDown menu)
+        {
+            base.AppendAdditionalMenuItems(menu);
+            ToolStripMenuItem item = Menu_AppendItem(menu, "Bake with Text", BakeGeometryFromMenu);
+            item.ToolTipText = "Bake not only geometry, but also result texts.";
+        }
+
+        private void BakeGeometryFromMenu(object sender, EventArgs e)
+        {
+            RhinoDoc doc = RhinoDoc.ActiveDoc;
+            var objIds = new List<Guid>();
+            BakeGeometry(doc, objIds);
+            doc.Views.Redraw();
+        }
+
         protected override Bitmap Icon => Resource.FishMarket;
         public override Guid ComponentGuid => new Guid("46E05DFF-8EFD-418A-AC5B-7C9F24559B2E");