issue in combining python list and numpy array?

Question

I have two python lists and a numpy array. The numpy array looks like this:

[array([93495052.969556, 98555123.061462])]
[array([1000976814.605984, 998276347.359732])]
[array([6868127850.435482, 6903911250.620625])]
[array([775127467.947004, 802369832.938230])]

this numpy array is formed from following code:

array1 = []
company = []
state = []

def process_chunk(chuk):

    training_set_feature_list = []
    training_set_label_list = []
    test_set_feature_list = []
    test_set_label_list = []
    np.set_printoptions(suppress=True)

    array2 = []


    # to divide into training & test, I am putting line 10th and 11th in test set
    count = 0
    for line in chuk:
        # Converting strings to numpy arrays
        if count == 9:   
            test_set_feature_list.append(np.array(line[3:4],dtype = np.float))
            test_set_label_list.append(np.array(line[2],dtype = np.float))
            company.append(line[0])
            state.append(line[1])
        elif count == 10:
            test_set_feature_list.append(np.array(line[3:4],dtype = np.float))
            test_set_label_list.append(np.array(line[2],dtype = np.float))

        else:    
            training_set_feature_list.append(np.array(line[3:4],dtype = np.float))
            training_set_label_list.append(np.array(line[2],dtype = np.float))
        count += 1
    # Create linear regression object
    regr = linear_model.LinearRegression()
    # Train the model using the training sets
    regr.fit(training_set_feature_list, training_set_label_list)

    #print test_set_feature_list

    array2.append(np.array(regr.predict(test_set_feature_list),dtype = np.float))
    np.set_printoptions(formatter={'float_kind':'{:f}'.format})
    for items in array2:
        array1.append(items)

array1 is the numpy array that I want to join with two python lists

First python list is company which looks like this:

['OT', 'OT', 'OT', 'OT',....]

Second python list is state:

['Alabama', 'Alabama', 'Alabama', 'Alabama', ...]

Now what I am trying to do is form one list which has following structure:

('OT', 'Alabama', 729, 733)
('OT', 'Alabama', 124, 122)
('OT', 'Arizona', 122, 124)

I wrote this line of code - final_list = zip(company,state,array1) but this produces this output (with added array and [] around array elements):

('OT', 'Alabama', array([729, 733]))
('OT', 'Alabama', array([124, 122]))

How do I join these lists and array so as to form one list which does not have above issue?

What does array1 look like? Can you do print(array1) and update its result? — Anand S Kumar
– Anand S Kumar, Commented Sep 17, 2015 at 17:54
@AnandSKumar I have provided my exact code which forms array1 and array2. The function is run multiple times. — user2966197
– user2966197, Commented Sep 17, 2015 at 18:16

Anand S Kumar · Accepted Answer · 2015-09-17 18:22:51Z

2

If array1 looks something like -

array1 = np.array([np.array([729, 733]), np.array([124, 122]) ...])

Though given your code (and usage of array1.append() , it looks like array1 is a list) . You can first convert the array1 list into numpy.array as -

narray1 = np.array(array1)           #This step not necessary , if array1 is already numpy array , in that case use `array1` instead of `narray1` .

Then if there are only two values per elements, you can do -

final_list = zip(company,state,narray1[:,0], narray1[:,1])

Demo -

In [59]: array1 = [np.array([729, 733]), np.array([124, 122])]

In [60]: company = ['OT', 'OT']

In [61]: state = ['Alabama', 'Alabama']

In [62]: narray1 = np.array(array1)

In [63]: final_list = zip(company,state,narray1[:,0], narray1[:,1])

In [65]: final_list
Out[65]: [('OT', 'Alabama', 729, 733), ('OT', 'Alabama', 124, 122)]

edited Sep 17, 2015 at 18:22

answered Sep 17, 2015 at 17:57

Anand S Kumar

91.4k18 gold badges196 silver badges179 bronze badges

Sign up to request clarification or add additional context in comments.

10 Comments

user2966197 Over a year ago

I am getting some weird output structure. It might be because of the way my numpy array is formed. I have updated my post above to include code which form the numpy array

Anand S Kumar Over a year ago

What is the output you are getting after using * ?

Anand S Kumar Over a year ago

Instead of array1 , try using array2 , What is the result of regr.predict(test_set_feature_list) like ?

user2966197 Over a year ago

array2 is temporary array inside my code and array1 is the global array. I am not using array2 because with each iteration of the code it gets refreshed. So before it gets refreshed I add the contents of array2 to global array1

Anand S Kumar Over a year ago

Then you would need to show an exact print of array1 , the array you show does not look like a valid array (pointing to the array in the question) .

|

hpaulj · Accepted Answer · 2015-09-17 19:33:06Z

It looks like you are trying to create a structured array, or its list equivalent (a list of tuples)

For convenience, lets just creat a list of tuples with your data. I won't get into the details of how this can be created from the other pieces (for now):

In [19]: data = [('OT', 'Alabama', 729, 733),
('OT', 'Alabama', 124, 122),
('OT', 'Arizona', 122, 124)]

Define a compound dtype (one of several possible formats):

In [23]: dt = np.dtype('S2,S10,i,i')

and the structured array:

In [24]: A=np.array(data,dtype=dt)

In [25]: A
Out[25]: 
array([('OT', 'Alabama', 729, 733), ('OT', 'Alabama', 124, 122),
       ('OT', 'Arizona', 122, 124)], 
      dtype=[('f0', 'S2'), ('f1', 'S10'), ('f2', '<i4'), ('f3', '<i4')])

which can be accessed by field name:

In [26]: A['f1']
Out[26]: 
array(['Alabama', 'Alabama', 'Arizona'], 
      dtype='|S10')

In [27]: A['f2']
Out[27]: array([729, 124, 122])

convert back to a list of tuples:

In [28]: A.tolist()
Out[28]: 
[('OT', 'Alabama', 729, 733),
 ('OT', 'Alabama', 124, 122),
 ('OT', 'Arizona', 122, 124)]

You can also create an empty (or zero) array of the right size and dtype, and fill it, row by row, or field by field

A1 = np.zeros((3,),dtype=dt)
A1['f0']=['OT','OT','OT']
A1['f2']=np.array([729,124,122])
etc

if the numbers are in a (3,2) integer array, you could use:

A1['f2']=x[:,0]
A1['f3']=x[:,1]

Collectives™ on Stack Overflow

issue in combining python list and numpy array?

2 Answers 2

10 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

10 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related