![]() ![]() When you design a shirt, add elements that relate to your personality. Support your T-shirt design ideas with relevant and striking visual cues that resonate with you and to anyone who sees them. All you need is to get your creative juice flowing to bring out your personality on any design. Make your own T-shirt for any season or reason. ![]() All are within our design collection should you even plan something special for mom and dad or celebrate the holidays with your peers. Do you want to show your support for advocacy or a cause? Go between modern, minimalist, or typographical shirt ideas. Our custom T-shirt maker comes with free templates that you can browse for a bit of inspiration and personalize to highlight any theme or motif you’re going for.Īre you attending a creative event with your fellow artists? Find abstract, ornate, or whimsical layouts to feature your aesthetics. What you wear can be an extension of who you are. import matplotlib.pyplot as plt inertias = ks = range(1,11) for k in ks: inertias.append(KMeans(n_clusters=k).fit(data).inertia_) plt.plot(ks, inertias, '-o') plt.xticks(ks) plt.Style made simple with our shirt design maker One way to find the number of clusters in the data is with an inertia plot, and to look for the ‘elbow’. We want three, but we also want them to contain enough observations that it’s worthwhile creating a specific T-shirt size for them. The next step is to decide how many clusters are in the data. The first step was to split the data into subsets, remove rows with 0 values (59 of 1584 in this case), then converted to a numpy matrix for clustering: import numpy as np t_shirt_columns = young = df = 83)].loc old_1 = df >= 156)].loc old_2 = df >= 156)].loc drop_list = for i in range(len(middle.index)): if 0 in middle.ix.values: drop_list.append(i) middle = middle.drop(middle.index) data = middle.as_matrix() shoulder_data = df] shoulder_data = shoulder_data.apply( lambda x: x//12) y_1 = shoulder_data = 1].groupby('YEAR').mean() y_2 = shoulder_data = 2].groupby('YEAR').mean() x = list(shoulder_oupby('YEAR').mean().index) With this information, it is estimated that Sex 1 is Male, and Sex 2 is Female. ![]() A plot of the average sitting heights per age for both Sex 1 and Sex 2 suggests that there isn’t much difference between Males and Females until around age 15, when it would be assumed that Males grow slightly taller. There is also scope for overlap of Small, Medium and Large across these boundaries, that is, the Large Shirt for 7–12 year olds might be a similar size to the small for 13 to 20 year olds.Īlso of interest is the relationship between Sex and size. ![]() As such, there are three age intervals where shirt sizes need to be determined 0 to 6, 7 to 12, 13 to 20. 6–7 years old and 12–13 are these two age jumps where there is a visible increase growth rates. This shows a roughly linear growth rate, with two ages where there is a dramatic increase in height. demographic_attributes = df = df.drop(demographic_attributes, axis = 'columns') Demographic data was deemed to not be useful in this situation, as the aim was to design T-shirt shirts, so 22 demographic variables were also removed. import pandas as pd df = pd.read_csv('data.csv', index_col=0) remove_cols = for i in df.columns: if 3900 - df.loc.astype(bool).sum() > 2000: remove_cols.append(i) df = df.drop(remove_cols, axis = 'columns')ħ6 variables were removed using this criteria.Īs stated above, the dataset contains both anthropometric and demographic data. As such, any variables which contained more than 2,000 Null values (zero in this case) were removed. The first step was to examine the data, and it was seen that many objects contained empty values. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |