This tutorial shows how to use the NavData class, including how to initialize instances and perform basic operations.

[1]:

import numpy as np
import pandas as pd

import gnss_lib_py as glp

[2]:

# Get data path of example file
glp.make_dir("../data")
!wget https://raw.githubusercontent.com/Stanford-NavLab/gnss_lib_py/main/notebooks/tutorials/data/myreceiver.csv --quiet -nc -O "../data/myreceiver.csv"
data_path = "../data/myreceiver.csv"

NavData

Initializing

Create an empty NavData class instantiation

[3]:

empty_nav_data = glp.NavData()
print(empty_nav_data)

Empty NavData
Rows: []
Index: []

Create a NavData class from a csv file

[4]:

nav_data_csv = glp.NavData(csv_path=data_path)
print(nav_data_csv)

   myTimestamp  mySatId  myPseudorange
0           10       10      270000001
1           10       14      270000007
2           10        7      270000004
3           10        3      270000005
4           11       10      270000002
5           11       14      270000008
6           11        7      270000003
7           11        3      270000004

Create an empty NavData class from a pandas DataFrame

[5]:

pd_df = pd.read_csv(data_path)
nav_data_pd = glp.NavData(pandas_df=pd_df)
print(nav_data_pd)

   myTimestamp  mySatId  myPseudorange
0           10       10      270000001
1           10       14      270000007
2           10        7      270000004
3           10        3      270000005
4           11       10      270000002
5           11       14      270000008
6           11        7      270000003
7           11        3      270000004

Create an empty NavData class instantiation

[6]:

np_array = np.eye(4)
nav_data_np = glp.NavData(numpy_array=np_array)
print(nav_data_np)

     0    1    2    3
0  1.0  0.0  0.0  0.0
1  0.0  1.0  0.0  0.0
2  0.0  0.0  1.0  0.0
3  0.0  0.0  0.0  1.0

Accessing values

In this example, the csv file that we have loaded has the following contents:

myTimestamp	mySatId	myPseudorange
10	G10	270000001
10	G14	270000007
10	G07	270000004
10	G03	270000005
11	G10	270000002
11	G14	270000008
11	G07	270000003
11	G03	270000004

NavData stores the different types of measurements, like myTimestamp as rows. So, each row contains the variable at different times/measurements. In this case, the myPseudorange row contains values for different timestamps and satellite IDs

Values in rows can be accessed by using the name of the row or the number of the row

[7]:

nav_data_csv['myTimestamp']

[7]:

array([10, 10, 10, 10, 11, 11, 11, 11])

[8]:

nav_data_csv[0]

[8]:

array([10, 10, 10, 10, 11, 11, 11, 11])

By specifying a column number (or slice of columns) along with the row label/number, a sub part of the matrix can be accessed as well

[9]:

nav_data_csv['myPseudorange', :3]

[9]:

array([270000001, 270000007, 270000004])

To access multiple rows using labels, pass the labels as a list of strings or slice of numbers

[10]:

nav_data_csv[['myTimestamp', 'myPseudorange']]

[10]:

array([[1.00000000e+01, 1.00000000e+01, 1.00000000e+01, 1.00000000e+01,
        1.10000000e+01, 1.10000000e+01, 1.10000000e+01, 1.10000000e+01],
       [2.70000001e+08, 2.70000007e+08, 2.70000004e+08, 2.70000005e+08,
        2.70000002e+08, 2.70000008e+08, 2.70000003e+08, 2.70000004e+08]])

However, to ensure compatability with accessing operations, you cannot access string and numeric values jointly. This is to prevent unexpected behaviour that might result on performing mathematical operations with strings and numbers.

If you attempt to access a combination of strings and numbers, the method will return an AssertionError

[11]:

try:
    nav_data_csv[0:2]
except AssertionError as err:
    print(err)

Setting values

You can update values for existing rows by indexing those rows and assigning an array

[12]:

nav_data_csv['myPseudorange'] = 10*np.arange(8)
nav_data_csv['myPseudorange']

[12]:

array([ 0, 10, 20, 30, 40, 50, 60, 70])

You can also set individual values by accessing the specific rows and columns that you want to reassign

[13]:

nav_data_csv['myPseudorange', 2] = 111
nav_data_csv['myPseudorange']

[13]:

array([  0,  10, 111,  30,  40,  50,  60,  70])

Setting new row

To set numeric values for a new row, access the row with the label name and assign the value to that row.

[14]:

nav_data_csv['new_row'] = np.arange(8)
nav_data_csv['new_row']

[14]:

array([0, 1, 2, 3, 4, 5, 6, 7])

New string rows can also be created similarly.

Note that while creating rows with string values, you must set the dtype of the row to object

[15]:

nav_data_csv['new_string_row'] = np.array(['string1', 'string1', 'string1', 'string1',
                                            'string2', 'string2', 'string2', 'string2'], dtype=object)
nav_data_csv['new_string_row']

[15]:

array(['string1', 'string1', 'string1', 'string1', 'string2', 'string2',
       'string2', 'string2'], dtype=object)

Find rows, shape and length

You can see which rows are currently in NavData using NavData.rows. The number of columns can be viewed using len(NavData) and the shape of the array can be viewed using NavData.shape

[16]:

nav_data_csv.rows

[16]:

['myTimestamp', 'mySatId', 'myPseudorange', 'new_row', 'new_string_row']

[17]:

nav_data_csv.shape

[17]:

(5, 8)

[18]:

len(nav_data_csv)

[18]:

[19]:

nav_data_csv.num_cols

[19]:

Removing rows or columns

To remove rows, use the NavData.remove() method, specifying the rows and columns you want to remove.

[20]:

nav_data_removed = nav_data_csv.remove(rows=['new_row', 'new_string_row'])
print(nav_data_removed.rows)
print(nav_data_removed.shape)

['myTimestamp', 'mySatId', 'myPseudorange']
(3, 8)

[21]:

nav_data_removed = nav_data_csv.remove(cols=[0, 1])
print(nav_data_removed.shape)

(5, 6)

Copy

If you don’t specify any rows or columns, the entire NavData will be copied

[22]:

nav_data_np_copy = nav_data_np.copy()
print('Original NavData array')
print(nav_data_np[:, :])
print('Copied NavData array')
print(nav_data_np_copy[:, :])

Original NavData array
[[1. 0. 0. 0.]
 [0. 1. 0. 0.]
 [0. 0. 1. 0.]
 [0. 0. 0. 1.]]
Copied NavData array
[[1. 0. 0. 0.]
 [0. 1. 0. 0.]
 [0. 0. 1. 0.]
 [0. 0. 0. 1.]]

Specifying rows, columns or both copies specified rows and columns.

Note that currently, rows and columns must be specified as lists.

[23]:

nav_data_np_copy = nav_data_np.copy(rows=[0,1], cols=list(range(4)))
print('Original NavData array')
print(nav_data_np[:, :])
print('Copied NavData array')
print(nav_data_np_copy[:, :])

Original NavData array
[[1. 0. 0. 0.]
 [0. 1. 0. 0.]
 [0. 0. 1. 0.]
 [0. 0. 0. 1.]]
Copied NavData array
[[1. 0. 0. 0.]
 [0. 1. 0. 0.]]

Convert to `pd.DataFrame` and save to csv

You can use the NavData.to_csv() method to save the data as a csv file and NavData.pandas_df() to get the equivalent pd.DataFrame

[24]:

nav_data_pd.pandas_df()

[24]:

	myTimestamp	mySatId	myPseudorange
0	10	10	270000001
1	10	14	270000007
2	10	7	270000004
3	10	3	270000005
4	11	10	270000002
5	11	14	270000008
6	11	7	270000003
7	11	3	270000004

Looping over columns

You can loop over all columns using in-built looping functions. In this example, we terminate the loop after 3 runs for display clarity

[25]:

for col_idx, nav_data_col in enumerate(nav_data_pd):
    print('Current column number', col_idx)
    print('Current column')
    print(nav_data_col)
    if col_idx >= 3:
        break

Current column number 0
Current column
   myTimestamp  mySatId  myPseudorange
0           10       10      270000001
Current column number 1
Current column
   myTimestamp  mySatId  myPseudorange
0           10       14      270000007
Current column number 2
Current column
   myTimestamp  mySatId  myPseudorange
0           10        7      270000004
Current column number 3
Current column
   myTimestamp  mySatId  myPseudorange
0           10        3      270000005