by Summary:
Ravel is a whole new way to analyse multidimensional data--which is anything that won't fit into the rows and columns of an Excel spreadsheet. This gives it much more power than Excel, but it also presents challenges when you import a CSV file into Ravel. The cause here is what is known as "the curse of dimensionality". A CSV file in Excel is just a 2-dimensional array of unrelated numbers, and its memory needs are the number of rows, times the number of columns, times how many bytes are used per cell, plus an overhead. So a 30 column by 100 row CSV file would take up about 30,000 bytes (with 10 bytes of data per cell). The same file in Ravel is a dimensioned object, where the number of dimensions is given by the number of axes, and the number of values on each dimension depends on how
Topics:
Steve Keen considers the following as important:
This could be interesting, too:
Ravel is a whole new way to analyse multidimensional data--which is anything that won't fit into the rows and columns of an Excel spreadsheet. This gives it much more power than Excel, but it also presents challenges when you import a CSV file into Ravel. The cause here is what is known as "the curse of dimensionality". A CSV file in Excel is just a 2-dimensional array of unrelated numbers, and its memory needs are the number of rows, times the number of columns, times how many bytes are used per cell, plus an overhead. So a 30 column by 100 row CSV file would take up about 30,000 bytes (with 10 bytes of data per cell). The same file in Ravel is a dimensioned object, where the number of dimensions is given by the number of axes, and the number of values on each dimension depends on how
Topics:
Steve Keen considers the following as important:
This could be interesting, too:
New Economics Foundation writes Building hope
New Economics Foundation writes Are oil and gas workers the coalminers of our generation?
Mike Norman writes Tariffs As A Fiscal Tool? — Brian Romanchuk
John Quiggin writes Trump’s dictatorship is a fait accompli
| Ravel is a whole new way to analyse multidimensional data--which is anything that won't fit into the rows and columns of an Excel spreadsheet. This gives it much more power than Excel, but it also presents challenges when you import a CSV file into Ravel. The cause here is what is known as "the curse of dimensionality". A CSV file in Excel is just a 2-dimensional array of unrelated numbers, and its memory needs are the number of rows, times the number of columns, times how many bytes are used per cell, plus an overhead. So a 30 column by 100 row CSV file would take up about 30,000 bytes (with 10 bytes of data per cell). The same file in Ravel is a dimensioned object, where the number of dimensions is given by the number of axes, and the number of values on each dimension depends on how many unique dimension values there are. With 30 dimensions and 100 unique entries per dimension, that is 30 to the power of 100 potential cells--and that's a substantially larger number than the number of atoms in the Universe. Ravel implements what's known as sparse data technology (see https://en.wikipedia.org/wiki/Sparse_matrix for an explanation if needed), which reduces the data needs drastically, but even so, it's easy to throw what looks like a small file at Ravel and generate an impossibly large data object. Therefore, you have to take care when you load data into Ravel, and this short video shows the basics. |