Count from a JSON file into python

Question

[
{
"frame_id":2, 
 "filename":"sample_imgs/2.jpg", 
 "objects": [ 
  {"class_id":8, "name":"boat", "relative_coordinates":{"left_x":1547, "top_y":1126, "width": 298, "height": 149}, "confidence":0.348690}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 128, "top_y":1327, "width":  30, "height":  89}, "confidence":0.972903}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 179, "top_y":1332, "width":  26, "height":  72}, "confidence":0.689888}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x":  79, "top_y":1324, "width":  29, "height":  73}, "confidence":0.663020}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 520, "top_y":1374, "width":  40, "height":  79}, "confidence":0.657240}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 565, "top_y":1367, "width":  36, "height":  84}, "confidence":0.447690}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 521, "top_y":1355, "width":  34, "height":  41}, "confidence":0.412204}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 159, "top_y":1318, "width":  27, "height":  77}, "confidence":0.306808}
 ]

How can I parse the file and count the number of IDs present in the JSON file using Python? The code i used displayed all the data, but id like something of narrowing down eg

"person": 7, 
"boat": 1

Here's my code

import json
from collections import Counter

with open('/home/output/result.json') as json_data:
    data = json.load(json_data)
    c = Counter(k[4:] for d in data for k, v in d.items() if k.startswith('class_id'))
print(data)
print(json.dumps(c, indent=2, sort_keys=True))

The values you want are in the name element, why are you using k.startswith('class_id')? — Barmar
– Barmar, Commented Nov 14, 2020 at 14:58

buran · Accepted Answer · 2020-11-14 15:00:07Z

1

data = [
{
"frame_id":2, 
 "filename":"sample_imgs/2.jpg", 
 "objects": [ 
  {"class_id":8, "name":"boat", "relative_coordinates":{"left_x":1547, "top_y":1126, "width": 298, "height": 149}, "confidence":0.348690}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 128, "top_y":1327, "width":  30, "height":  89}, "confidence":0.972903}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 179, "top_y":1332, "width":  26, "height":  72}, "confidence":0.689888}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x":  79, "top_y":1324, "width":  29, "height":  73}, "confidence":0.663020}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 520, "top_y":1374, "width":  40, "height":  79}, "confidence":0.657240}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 565, "top_y":1367, "width":  36, "height":  84}, "confidence":0.447690}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 521, "top_y":1355, "width":  34, "height":  41}, "confidence":0.412204}, 
  {"class_id":0, "name":"person", "relative_coordinates":{"left_x": 159, "top_y":1318, "width":  27, "height":  77}, "confidence":0.306808}
 ]
}]

import json
from collections import Counter
counter = Counter(obj.get('name') for item in data for obj in item['objects'])
print(counter.most_common())
with open('data.json', 'w') as f:
    json.dump(counter, f, indent=4

)

output

[('person', 7), ('boat', 1)]

data.json

{
    "boat": 1,
    "person": 7
}

This assumes you may have more than one frame in the JSON file and handles all frames at once. Of course you can convert counter to dict

edited Nov 14, 2020 at 15:00

answered Nov 14, 2020 at 14:53

buran

14.4k13 gold badges45 silver badges76 bronze badges

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

5 Comments

Pan Over a year ago

Yes, there are multiple frame into this JSON. Will converting to dict would display me the { "boat": 1, "person": 7 }

buran Over a year ago

Yes, it will. But if you going to write to JSON file you don't need to. And Counter object has dict-like interface. Do it's really just a presentation issue

Pan Over a year ago

Thanks for the reply, how its possible to convert this and getting a HEX string instead of a python list using this solution?

buran Over a year ago

I would suggest you ask separate question with the sample input and expected output.

Pan Over a year ago

While a modified solution of the above code worked as expected i cant represent while i try to print values above 255 that later generated in my JSON. i get a byte must be in range(0, 256) error. Bear in mind i converted the above tuple representation to a bytes string needed for my application. "print(bytes(str))"

BhishanPoudel · Accepted Answer · 2020-11-14 15:10:16Z

0

Assuming you have data file a.json, you can use pandas and json to manipulate the data.

import json
import pandas as pd

with open('a.json') as fi:
    jdata = json.load(fi)


df = pd.DataFrame(data=jdata[0]['objects'])
output = df['name'].value_counts().to_dict()
print(output)

with open('output.json','w') as fo:
    json.dump(output,fo,indent=4)

output

{'boat': 1, 'person': 7}

edited Nov 14, 2020 at 15:10

answered Nov 14, 2020 at 14:57

BhishanPoudel

17.4k27 gold badges123 silver badges190 bronze badges

3 Comments

Pan Over a year ago

Code is executing, but i dont get the print.

BhishanPoudel Over a year ago

I was working in jupyter notebook. The edited code should work now.

Pan Over a year ago

I get the printout thank you, im having a module were i want to sent Lora packages and in my code i sent messages in the form of lora.sent("Hello"). When im pasting the lora.sent(json.dump(output,fo,indent=4)) i get the error of "Encoding without a string argument. Probably that python list needs to be converted to a string or a hex.

Collectives™ on Stack Overflow

Count from a JSON file into python

2 Answers 2

5 Comments

output

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

5 Comments

output

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related