Video classification

curl --request POST \ --url https://api.synapsai.cloud/v1/classifications/video \ --header 'Content-Type: application/json' \ --data ' { "inputs": "<string>", "top_k": 5, "num_frames": 16, "frame_sampling_rate": 1, "function_to_apply": "softmax" } '

import requests url = "https://api.synapsai.cloud/v1/classifications/video" payload = { "inputs": "<string>", "top_k": 5, "num_frames": 16, "frame_sampling_rate": 1, "function_to_apply": "softmax" } headers = {"Content-Type": "application/json"} response = requests.post(url, json=payload, headers=headers) print(response.text)

Body

application/json

Request schema for video classification pipelines.

inputs

required

Video URL or local path (single or list). Required.

top_k

integer

default:5

Number of top labels to return.

num_frames

integer

default:16

Number of frames sampled from the video to run inference on.

frame_sampling_rate

integer

default:1

Sampling rate used to select frames from the video.

function_to_apply

enum<string>

default:softmax

Function to apply to raw model outputs to obtain scores.

Available options:

sigmoid,

softmax,

none

Response

200 - application/json

Video classification results

label

string

Predicted label name.

score

number

Confidence score (probability-like) for the label.

Overview

Endpoints

Body

Response