Merge pull request #44 from peter-shoes/gui_functionality

Gui functionality

igel/gui.py

@@ -0,0 +1,297 @@
+import os
+import platform
+import tkinter as tk
+from tkinter.filedialog import askopenfilename
+
+# really shitty function to get the readme as stantard ascii
+def readme():
+    try:
+        import docutils
+    except ImportError:
+        try:
+            with open(os.path.join(os.path.dirname(__file__), '../README.rst')) as f:
+                return f.read()
+        except (IOError, OSError):
+            return ''
+    with open(os.path.join(os.path.dirname(__file__), 'README.rst')) as f:
+        document = docutils.core.publish_doctree(f.read())
+        nodes = list(document)
+        description = ''
+        for node in nodes:
+            if str(node).startswith('<topic classes="contents"'):
+                break
+            if type(node) is docutils.nodes.comment\
+            or type(node) is docutils.nodes.title:
+                continue
+            description += node.astext() + '\n'
+        return description.encode('ascii', 'ignore').strip()
+
+def entry_str(name,col,row,set_def=None):
+    tk.Label(root, text=name).grid(column=col,row=row)
+    var = tk.StringVar()
+    var.set(set_def)
+    tk.Entry(root,textvariable=var).grid(column=col+1,row=row)
+    return var
+
+def entry_int(name,col,row,set_def=None):
+    tk.Label(root, text=name).grid(column=col,row=row)
+    var = tk.IntVar()
+    var.set(set_def)
+    tk.Entry(root,textvariable=var).grid(column=col+1,row=row)
+    return var
+
+def optionmenu(name,col,row,options):
+    tk.Label(root, text=name).grid(column=col,row=row)
+    var = tk.StringVar(root)
+    var.set(options[0])
+    tk.OptionMenu(root,var,*options).grid(column=col+1,row=row)
+    return var
+
+def checkbox_bool(name,col,row):
+    tk.Label(root, text=name).grid(column=col,row=row)
+    var = tk.BooleanVar()
+    var.set(False)
+    tk.Checkbutton(root,text='True',variable=var).grid(column=col+1,row=row)
+    return var
+
+def opvars():
+    global opvars_rowcount
+    global opvars_complete
+    op = opvar_var.get()
+    tk.Label(root,text=op).grid(column=0,row=opvars_rowcount)
+    var = tk.StringVar()
+    tk.Entry(root,textvariable=var).grid(column=1,row=opvars_rowcount)
+    opvars_rowcount +=1
+    opvars_complete.append([op,var])
+    pass
+
+def param_grid_add_row():
+    global param_rowcount
+    global param_complete
+    global param_grid_no_cols
+    params_list = []
+    for col in range(param_grid_no_cols.get()):
+        var = tk.StringVar()
+        tk.Entry(root,textvariable=var).grid(column=5+col,row=param_rowcount)
+        params_list.append(var)
+    param_complete.append(params_list)
+    param_rowcount+=1
+    pass
+
+def get_model_type():
+    global textbase
+    global model_algorithm_menu
+    global model_algorithm_var
+    global model_type
+    model_algorithm_var.set('')
+    model_type_var = model_type.get()
+    algo_options_regression = []
+    algo_options_classification = []
+    algo_options_clustering = []
+    for line in textbase[107:134]:
+        temp_algo_list = []
+        for i in line.split('|'):
+            temp_algo_list.append(i)
+        temp_algo_list = [i.strip() for i in temp_algo_list[1:4]]
+        algo_options_regression.append(temp_algo_list[0])
+        algo_options_classification.append(temp_algo_list[1])
+        algo_options_clustering.append(temp_algo_list[2])
+
+    if model_type_var=='regression':
+        algo_options=algo_options_regression
+    elif model_type_var=='classification':
+        algo_options=algo_options_classification
+    elif model_type_var=='clustering':
+        algo_options=algo_options_clustering
+
+    model_algorithm_menu['menu'].delete(0,'end')
+    for option in algo_options:
+        model_algorithm_menu['menu'].add_command(label=option, command=tk._setit(model_algorithm_var,option))
+
+def main():
+    global root
+    root = tk.Tk()
+    root.title("Igel GUI")
+
+# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
+    # set root directory for files, check if windows and pull up filesystem
+    rootdir = '/'
+    if platform.system()=='Windows':
+        rootdir='C:\\'
+    filename = askopenfilename(initialdir=rootdir)
+    # filename = askopenfilename()
+    tk.Label(root, text='Chosen File: %s'%filename).grid(column=0,row=0)
+    # get filetype
+    filetype = filename.split('/')[-1].split('.')[-1]
+    tk.Label(root, text='File Type: %s'%filetype).grid(column=0,row=1)
+
+    # do the whole read_data_options nonsense
+    global textbase
+    textbase = readme().split('\n')
+    read_data_options = []
+    for line in textbase[276:314]:
+        read_data_options.append(line.split(':')[0].strip())
+    tk.Label(root, text='Read Data Options:').grid(column=0,row=2)
+
+# i'm sorry this is a huge fucking mess
+    global opvar_var
+    opvar_var = tk.StringVar(root)
+    opvar_var.set(read_data_options[0])
+    tk.OptionMenu(root,opvar_var,*read_data_options).grid(column=1,row=2)
+    global opvars_rowcount
+    opvars_rowcount = 3
+    tk.Button(root, text='Add Option',command=opvars).grid(column=2,row=2)
+    global opvars_complete
+    opvars_complete = []
+
+# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
+    tk.Label(root, text='Split Options').grid(column=4,row=0)
+
+    split_test_size = entry_int('Test Size',4,1,0.2)
+
+    split_shuffle = checkbox_bool('Shuffle',4,2)
+
+    split_stratify = entry_str('Stratify (Seperate by comma)',4,3)
+
+# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
+    tk.Label(root, text='Preprocess').grid(column=4,row=4)
+
+    preprocess_missing_values_list = ['mean','median','most_frequent','constant']
+    preprocess_missing_values = optionmenu('Missing Values',4,5,preprocess_missing_values_list)
+
+    preprocess_encoding_type_list = ['oneHotEncoding','labelEncoding']
+    preprocess_encoding_type = optionmenu('Encoding Type',4,6,preprocess_encoding_type_list)
+
+    preprocess_scale_method_list = ['standard','minmax']
+    preprocess_scale_method = optionmenu('Scaling Method',4,7,preprocess_scale_method_list)
+
+    preprocess_scale_target_list = ['inputs','outputs','all']
+    preprocess_scale_target = optionmenu('Scaling Target',4,8,preprocess_scale_target_list)
+
+# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
+    tk.Label(root, text='Model Definition').grid(column=4,row=9)
+
+    model_type_list = ['regression','classification','clustering']
+    global model_type
+    model_type = optionmenu('Model Type',4,10,model_type_list)
+
+    tk.Button(root, text='Confirm Type',command=get_model_type).grid(column=6,row=10)
+
+    global model_algorithm_menu
+    global model_algorithm_var
+    tk.Label(root, text='Algorithm').grid(column=4,row=11)
+    model_algorithm_var = tk.StringVar(root)
+    model_algorithm_var.set('select model type')
+    model_algorithm_menu = tk.OptionMenu(root,model_algorithm_var,'select model type')
+    model_algorithm_menu.grid(column=5,row=11)
+
+    model_arguments = entry_str('Arguments (Seperate by comma)',4,12)
+
+    model_use_cv_estimator = checkbox_bool('Use CV Estimator',4,13)
+
+# ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+
+    tk.Label(root, text='Cross Validate').grid(column=4,row=14)
+
+    model_cross_validate_cv = entry_int('CV',4,15,5)
+
+    model_cross_validate_n_jobs = entry_str('N Jobs',4,16)
+
+    model_cross_validate_verbose = entry_int('Verbose',4,17,0)
+
+    tk.Label(root,text='Hyperparameter Search').grid(column=4,row=18)
+
+    model_hyperparameter_search_method_list = ['grid_search','random_search']
+    model_hyperparameter_search_method = optionmenu('Method',4,19,model_hyperparameter_search_method_list)
+
+    tk.Label(root, text='Parameter Grid').grid(column=4,row=20)
+    global param_grid_no_cols
+    param_grid_no_cols = entry_int('Number of Columns',4,21,2)
+    global param_rowcount
+    global param_complete
+    param_complete = []
+    param_rowcount = 22
+    tk.Button(root, text='Add Grid Row',command=param_grid_add_row).grid(column=6,row=21)
+
+    # model_hyperparameter_search_arguments = entry_str('Arguments (Seperate by comma)',6,0)
+
+    model_hyperparameter_search_arguments_cv = entry_int('CV',6,1,5)
+
+    model_hyperparameter_search_arguments_refit = checkbox_bool('Refit',6,2)
+
+    model_hyperparameter_search_arguments_return_train_score = checkbox_bool('Return Train Score',6,3)
+
+    model_hyperparameter_search_arguments_verbose = entry_int('Verbose',6,4,0)
+
+    target = entry_str('Target (Seperate by comma)',6,5)
+
+    root.mainloop()
+
+    opvars_dict = {}
+    for x in opvars_complete:
+        opvars_dict[x[0]] = x[1].get()
+
+    parameter_grid_dict = {}
+    param_num = 1
+    for x in param_complete:
+        param_getter_list = []
+        for i in x:
+            param_getter_list.append(i.get())
+        param_num_str = 'param%d'%param_num
+        parameter_grid_dict[param_num_str] = param_getter_list
+        param_num+=1
+
+    yaml_dict = {
+        'dataset':[filename, {
+            'type':filetype,
+            'read_data_options':opvars_dict,
+            'split': {
+                'test_size':split_test_size.get(),
+                'shuffle':split_shuffle.get(),
+                'stratify':split_stratify.get().split(',')
+            },
+            'preprocess': {
+                'missing_values': preprocess_missing_values.get(),
+                'encoding': {
+                    'type':preprocess_encoding_type.get()
+                },
+                'scale': {
+                    'method':preprocess_scale_method.get(),
+                    'target':preprocess_scale_target.get()
+                }
+            }
+        }],
+        'model': {
+            'type': model_type.get(),
+            'algorithm':model_algorithm_var.get(),
+            'arguments':model_arguments.get().split(','),
+            'use_cv_estimator':model_use_cv_estimator.get(),
+            'cross_validate': {
+                'cv':model_cross_validate_cv.get(),
+                'n_jobs':model_cross_validate_n_jobs.get(),
+                'verbose':model_cross_validate_verbose.get()
+            },
+            'hyperparameter_search': {
+                'method':model_hyperparameter_search_method.get(),
+                'parameter_grid':parameter_grid_dict,
+                'arguments':{
+                    'cv':model_hyperparameter_search_arguments_cv.get(),
+                    'refit':model_hyperparameter_search_arguments_refit.get(),
+                    'return_train_score':model_hyperparameter_search_arguments_return_train_score.get(),
+                    'verbose':model_hyperparameter_search_arguments_verbose.get()
+                }
+            }
+
+        },
+        'target': target.get().split(',')
+    }
+
+    return(yaml_dict)
+
+
+if __name__ == '__main__':
+    main()