Hi -- that's a Jupyter widget from the ipywidgets library (ipywidgets.readthedocs.io/en/stable/). I have been considering making a video about them, so hopefully that will be coming up eventually. But for now, you can find the code I used to make this standard deviation widget on my GitHub page: github.com/kimfetti/Videos/blob/master/Shorts/standard_deviation_widget.ipynb
Is it possible to combine in a multi column graph using standard deviations and a heatmap showing the areas further from the mean in gradations of color signifying distance from the mean? i.e values close to the mean may be white and those further away might be some other darker color?
Sure thing - I think you could do this. I would probably load the data into a pandas data frame, and then scale each column (or row) by subtracting the mean and dividing by the standard deviation of each particular column (or row). The values should then be centered about zero and scaled in terms of the standard deviation. Then you could just do a seaborn heatmap of the entire pandas dataframe. 👍You could do a diverging color palette so that values much smaller than the mean would be one color and values much larger than the mean would be a different color... or you could do the absolute value of everything and then just use a sequential color palette.
@@KimberlyFessel Very clever, I like it. ty. Is it then possible to use annot=true to show the original value instead of the weighted formula value in the weighted matrix?
@@philr3630 annot=True would just pull in the current values that you are plotting... BUT you can actually pass a dataframe or a numpy array as the value for annot (for example, annot=df_labels) as long as the annotations are the same size as the current values of the heatmap.
Great job
I must subscribe to your channel 😍😍😍😍😍 thanks
awesome
Thank you! 😄
You make a wonderful job. It is very difficult to find this type of quality in Russian youtube. I am talking about your channel
Excellent -- thank you very much!
How do you build such an interactive chart where you can adjust like in this case the standard deviation? Thanks
Hi -- that's a Jupyter widget from the ipywidgets library (ipywidgets.readthedocs.io/en/stable/). I have been considering making a video about them, so hopefully that will be coming up eventually. But for now, you can find the code I used to make this standard deviation widget on my GitHub page: github.com/kimfetti/Videos/blob/master/Shorts/standard_deviation_widget.ipynb
@@KimberlyFessel Thanks Kimberly, looking forward to that video
I am confused with one thing Why and How mean shifts
Please clear my Confusion
With Pratical Example
Is it possible to combine in a multi column graph using standard deviations and a heatmap showing the areas further from the mean in gradations of color signifying distance from the mean? i.e values close to the mean may be white and those further away might be some other darker color?
Sure thing - I think you could do this. I would probably load the data into a pandas data frame, and then scale each column (or row) by subtracting the mean and dividing by the standard deviation of each particular column (or row). The values should then be centered about zero and scaled in terms of the standard deviation. Then you could just do a seaborn heatmap of the entire pandas dataframe. 👍You could do a diverging color palette so that values much smaller than the mean would be one color and values much larger than the mean would be a different color... or you could do the absolute value of everything and then just use a sequential color palette.
@@KimberlyFessel Thank you very much. I will try it very soon. Great suggestion.
@@KimberlyFessel Very clever, I like it. ty. Is it then possible to use annot=true to show the original value instead of the weighted formula value in the weighted matrix?
@@philr3630 annot=True would just pull in the current values that you are plotting... BUT you can actually pass a dataframe or a numpy array as the value for annot (for example, annot=df_labels) as long as the annotations are the same size as the current values of the heatmap.