English
资源说明：Useful utility functions for data analysis with Python.
# python-danutil
A collection of useful functions and classes for use in data analysis.
## Documentation
python-danutil provides functions and classes in these general areas:
* [General I/O](#general-i/o)
* [List Operations](#list-operations)
* [Utilities](#utilities)
* [Plotting Convenience](#plotting-convenience)
* [Curve Fitting](#curve-fitting)
* [Mathematical Functions for Curve Fitting](#mathematical-functions-for-curve-fitting)
* [Statistics](#statistics)
### General I/O
```
import_file(filename, datatype='', *args, **kwargs)
filename : string
File to import.
datatype : string
Use 'data' to import file as a data file.
Imports text and images files, automatically seperating text into arrays
based on the files extension, or optional type paramater.
```
```
print_table(input_list, headings=None, usetabs=False, fmt='%g')
input_list : array like
2D list to print out in table form.
headings : list
1D list of table headings.
usetabs : bool
Whether to seperate entries with tabs or spaces. Useful
if you want to copy the data into another program.
fmt : string
Format string to use when printing numbers.
```
```
savenpy(filename, X, delimiter=',')
filename : string
Location to save file.
X : Numpy Array
Array to be written to file
Saves a numpy array to file.
```
```
savecsv(fname, seq, headers=None)
fname : string
Location to save file.
seq : 2D array of objects
List of rows to be written.
headers : list of strings
Headers to be written at top of file.
Writes a csv file with the string representation of the data in seq.
Data does not have to be of the same data type.
```
### List Operations
```
all_indices(l, i)
l : list
list of items
i : object
item to look for in list
Returns a list of all indicies for item i.
```
```
delete_duplicates(seq)
seq : list
The sequence to delete duplicates from.
Returns a list with duplicates deleted from seq while maintaining ordering.
```
```
split(seq, key=None)
seq : list
The sequence to split.
key : callable
Split using value returned by applying key to item.
Splits list into runs of identical elements as specified by key.
Example:
This splits data based on first item in each list.
data = [['a', 1, 2], ['a', 3, 4], ['b', 1, 2]]
split(data, key=itemgetter(0))
outputs: [[['a', 1, 2], ['a', 3, 4]], [['b', 1, 2]]]
```
```
pack(*args)
*args : list
A series of lists.
pack(x, y) combines lists [x1, x2, x3], [y1, y2, y3]
into form [[x1, y1], [x2, y2], [x3, y3]]
It is essentially just the builtin zip.
```
```
unpack(seq, column)
seq : list
A packed list of form [[x1, y1], [x2, y2], [x3, y3]]
Extract column in packed list.
With input seq = [[x1, y1], [x2, y2], [x3, y3]]
unpack(seq, 0) returns [x1, x2, x3]
```
```
natural_sort(seq)
seq : list
A list of strings to sort.
Sorts the given sequence in place.
```
### Utilities
```
docs()
Opens a documentation browsers for numpy, scipy, matplotlib, and danutil.
```
```
isnumber(val)
val : object
The object to be tested
Returns True of val is an int, long, float, or complex number.
```
```
listcwd(startswith='', endswith='')
starts : string
Filter results such that they start with this string.
This can be a tuple of strings.
ends : string
Filter resutls such that they end with this string.
This can be a tuple of strings.
Returns a list of files in the current working directory that
start with 'starts' and end with 'ends'
```
### Plotting Convenience
```
date_plot(dates, data, *args, **kwargs)
dates : list
A 1D list of date strings.
data : list
A 1D list of data corresponding to the values in dates.
Plots dates vs. data.
```
```
plot_function(func, coef, xmin, xmax, num_vals=100, *args)
func : callable
The function to be plotted.
coef : list
List of coefficients for function.
xmin : float
The minimum x value to start plotting at.
xmax : float
The maximum x value to plot to.
num_vals : int
The number of discrete values to plot to.
Plots the function to the current matplotlib plot.
Example:
def _line_func(x, m, b):
return m * x + b
x = np.linspace(0, 10, 10)
y = [0.1, 1.3, 2.1, 3.2, 4.7, 5.6, 6.4, 7.5,9, 10.1]
fit = scipy.optimize.curve_fit(line_func, x, y)
plot_function(line_func, fit[0], 0, 10)
```
### Curve Fitting
```
fitline(x, y)
Fits the data x, y to a line.
Returns (slope, x_intercept, r_squared)
```
### Mathematical Functions for Curve Fitting
```
exp_func(x, a, b, c)
Returns a function of form: a*e^(b*x) + c
```
```
gaussian_func(x, a, b, c)
aReturns a function of form: *e^((x-b)^2/2c^2)
a = 1/(sigma*sqrt(2*pi))
b = mu (expected value)
c = sigma (standard deviation)
```
```
line_func(x, m, b)
Returns a function of form: m*x + b
```
```
sin_func(x, a, w, phi)
aReturns a function of form: *sin(w*x + phi)
```
### Statistics
```
std(seq, *args, **kwargs)
seq : array
The sequence to compute the standard deviation of.
Returns the sample standard deviation of seq.
```
```
ANOVA(object)
Calculates the one-way ANOVA of passed samples.
ANOVA(list1, list2, ...)
Attributes:
self.grand_mean Grand mean of all samples
self.sst Total sum of squares
self.ssb Between groups sum of squares
self.ssw Within group sum of squares
self.fstat F statistic
self.pvalue P value
Methods:
self.summary() Prints the ANOVA summary table.
```
本源码包内暂不包含可直接显示的源代码文件，请下载源码包。
