Video

class oceanai.modules.lab.video.VideoMessages(lang: str = 'ru', color_simple: str = '#666', color_info: str = '#1776D2', color_err: str = '#FF0000', color_true: str = '#008001', bold_text: bool = True, text_runtime: str = '', num_to_df_display: int = 30)[source]

Bases: Download

Class for messages

Parameters:
class oceanai.modules.lab.video.Video(lang: str = 'ru', color_simple: str = '#666', color_info: str = '#1776D2', color_err: str = '#FF0000', color_true: str = '#008001', bold_text: bool = True, text_runtime: str = '', num_to_df_display: int = 30)[source]

Bases: VideoMessages

Video processing class

Parameters:
__calc_reshape_img_coef(shape: Tuple[int] | List[int], new_shape: int | Tuple[int] | List[int], out: bool = True) → float

Calculating the image resizing factor

Note

private method

Parameters:

  • shape (Union[Tuple[int], List[int]]) – Current image size (width, height)

  • new_shape (Union[int, Tuple[int], List[int]]) – Desired image size

  • out (bool) – Display

Returns:

Image resizing factor

Return type:

float

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4
5video._Video__calc_reshape_img_coef(
6    shape = (1280, 720),
7    new_shape = 224,
8    out = True
9)
[1]:
10.175

True – 2 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4
5video._Video__calc_reshape_img_coef(
6    shape = (1280, 720),
7    new_shape = (1920, 1080),
8    out = True
9)
[1]:
11.5

Error – 1 –

In [3]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video._Video__calc_reshape_img_coef(
6    shape = (1280, 720),
7    new_shape = '',
8    out = True
9)
[4]:
1[2022-10-29 13:24:27] Invalid argument types or values in "Video.__calc_reshape_img_coef" ...
2
3-1.0
__concat_pred(pred_hc: ndarray, pred_nn: ndarray, out: bool = True) → List[ndarray | None]

Concatenation of scores by hand-crafted and deep features

Note

private method

Parameters:

  • pred_hc (np.ndarray) – Scores on hand-crafted features

  • pred_nn (np.ndarray) – Scores on deep features

  • out (bool) – Display

Returns:

Concatenated scores by hand-crafted and deep features

Return type:

List[Optional[np.ndarray]]

Examples

True – 1 –

In [1]:
 1import numpy as np
 2from oceanai.modules.lab.video import Video
 3
 4video = Video()
 5
 6arr_hc = np.array([
 7    [0.64113516, 0.6217892, 0.54451424, 0.6144415, 0.59334993],
 8    [0.6652424, 0.63606125, 0.572305, 0.63169795, 0.612515]
 9])
10
11arr_nn = np.array([
12    [0.56030345, 0.7488746, 0.44648764, 0.59893465, 0.5701077],
13    [0.5900006, 0.7652722, 0.4795154, 0.6409055, 0.6088242]
14])
15
16video._Video__concat_pred(
17    pred_hc = arr_hc,
18    pred_nn = arr_nn,
19    out = True
20)
[1]:
 1[
 2    array([
 3            0.64113516, 0.6652424, 0.65318878, 0.65318878, 0.65318878,
 4            0.65318878, 0.65318878, 0.65318878, 0.65318878, 0.65318878,
 5            0.65318878, 0.65318878, 0.65318878, 0.65318878, 0.65318878,
 6            0.65318878, 0.56030345, 0.5900006, 0.57515202, 0.57515202,
 7            0.57515202, 0.57515202, 0.57515202, 0.57515202, 0.57515202,
 8            0.57515202, 0.57515202, 0.57515202, 0.57515202, 0.57515202,
 9            0.57515202, 0.57515202
10    ]),
11    array([
12            0.6217892, 0.63606125, 0.62892523, 0.62892523, 0.62892523,
13            0.62892523, 0.62892523, 0.62892523, 0.62892523, 0.62892523,
14            0.62892523, 0.62892523, 0.62892523, 0.62892523, 0.62892523,
15            0.62892523, 0.7488746, 0.7652722, 0.7570734, 0.7570734,
16            0.7570734, 0.7570734, 0.7570734, 0.7570734, 0.7570734,
17            0.7570734, 0.7570734, 0.7570734, 0.7570734, 0.7570734,
18            0.7570734, 0.7570734
19    ]),
20    array([
21            0.54451424, 0.572305, 0.55840962, 0.55840962, 0.55840962,
22            0.55840962, 0.55840962, 0.55840962, 0.55840962, 0.55840962,
23            0.55840962, 0.55840962, 0.55840962, 0.55840962, 0.55840962,
24            0.55840962, 0.44648764, 0.4795154, 0.46300152, 0.46300152,
25            0.46300152, 0.46300152, 0.46300152, 0.46300152, 0.46300152,
26            0.46300152, 0.46300152, 0.46300152, 0.46300152, 0.46300152,
27            0.46300152, 0.46300152
28    ]),
29    array([
30            0.6144415, 0.63169795, 0.62306972, 0.62306972, 0.62306972,
31            0.62306972, 0.62306972, 0.62306972, 0.62306972, 0.62306972,
32            0.62306972, 0.62306972, 0.62306972, 0.62306972, 0.62306972,
33            0.62306972, 0.59893465, 0.6409055, 0.61992008, 0.61992008,
34            0.61992008, 0.61992008, 0.61992008, 0.61992008, 0.61992008,
35            0.61992008, 0.61992008, 0.61992008, 0.61992008, 0.61992008,
36            0.61992008, 0.61992008
37    ]),
38    array([
39            0.59334993, 0.612515, 0.60293247, 0.60293247, 0.60293247,
40            0.60293247, 0.60293247, 0.60293247, 0.60293247, 0.60293247,
41            0.60293247, 0.60293247, 0.60293247, 0.60293247, 0.60293247,
42            0.60293247, 0.5701077, 0.6088242, 0.58946595, 0.58946595,
43            0.58946595, 0.58946595, 0.58946595, 0.58946595, 0.58946595,
44            0.58946595, 0.58946595, 0.58946595, 0.58946595, 0.58946595,
45            0.58946595, 0.58946595
46    ])
47]

Error – 1 –

In [2]:
 1import numpy as np
 2from oceanai.modules.lab.video import Video
 3
 4video = Video(lang = 'en')
 5
 6arr_hc = np.array([
 7    [0.64113516, 0.6217892, 0.54451424, 0.6144415],
 8    [0.6652424, 0.63606125, 0.572305, 0.63169795, 0.612515]
 9])
10
11arr_nn = np.array([
12    [0.56030345, 0.7488746, 0.44648764, 0.59893465, 0.5701077],
13    [0.5900006, 0.7652722, 0.4795154, 0.6409055, 0.6088242]
14])
15
16video._Video__concat_pred(
17    pred_hc = arr_hc,
18    pred_nn = arr_nn,
19    out = True
20)
[3]:
1[2024-10-09 11:34:39] Что-то пошло не так ... конкатенация оценок по экспертным и нейросетевым признакам не произведена (видео модальность) ...
2
3[]
__load_model_weights(url: str, force_reload: bool = True, info_text: str = '', out: bool = True, runtime: bool = True, run: bool = True) → bool

Downloading the weights of the neural network model

Note

private method

Parameters:

  • url (str) – Full path to the file with weights of the neural network model

  • force_reload (bool) – Forced download of a file with weights of a neural network model from the network

  • info_text (str) – Text for informational message

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the weights of the neural network model are downloaded, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8video._Video__load_model_weights(
 9    url = 'https://drive.usercontent.google.com/download?id=1QF7ReDQXpCciF7aWjbEt4Q-x06hwDrMZ&export=download&authuser=2&confirm=t&uuid=c2fd5a21-7af7-4b7f-8419-d7d628847768&at=AO7h07eilj-Bm5RIk0HwQBEr37ri:1727175670133',
10    force_reload = True,
11    info_text = 'Загрузка весов нейросетевой модели',
12    out = True, runtime = True, run = True
13)
[1]:
1[2024-10-09 11:42:21] Загрузка весов нейросетевой модели
2
3[2024-10-09 11:42:25] Загрузка файла "weights_2022-03-22_16-31-48.pth" 100.0% ...
4
5--- Время выполнения: 4.117 сек. ---
6
7True

– 2 –

In [2]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8video._Video__load_model_weights(
 9    url = './models/weights_2022-03-22_16-31-48.pth',
10    force_reload = True,
11    info_text = 'Загрузка весов нейросетевой модели',
12    out = True, runtime = True, run = True
13)
[2]:
1[2024-10-09 11:46:15] Загрузка весов нейросетевой модели
2
3--- Время выполнения: 0.005 сек. ---
4
5True

Error – 1 –

In [3]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8video._Video__load_model_weights(
 9    url = 'https://drive.usercontent.google.com/download?id=1QF7ReDQXpCciF7aWjbEt4Q-x06hwDrMZ&export=download&authuser=2&confirm=t&uuid=c2fd5a21-7af7-4b7f-8419-d7d628847768&at=AO7h07eilj-Bm5RIk0HwQBEr37ri:1727175670133',
10    force_reload = True, info_text = '',
11    out = True, runtime = True, run = True
12)
[3]:
1[2022-10-27 12:48:24] Invalid argument types or values in "Video.__load_model_weights" ...
2
3False
__load_video_model_b5(show_summary: bool = False, out: bool = True) → Module | None

Formation of the neural network architecture of the model to obtain the personality traits scores

Note

private method

Parameters:

  • show_summary (bool) – Displaying the formed neural network architecture of the model

  • out (bool) – Display

Returns:

None если неверные типы или значения аргументов, в обратном случае нейросетевая модель nn.Module для получения результата оценки персонального качества

Return type:

Optional[nn.Module]

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4
5video._Video__load_video_model_b5(
6    show_summary = True, out = True
7)
[1]:
1video_model_b5(
2    (fc): Linear(in_features=32, out_features=1, bias=True)
3    (sigmoid): Sigmoid()
4)
5video_model_b5(
6    (fc): Linear(in_features=32, out_features=1, bias=True)
7    (sigmoid): Sigmoid()
8)

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video._Video__load_video_model_b5(
6    show_summary = True, out = []
7)
[3]:
1[2022-10-17 10:53:03] Invalid argument types or values in "Video.__load_video_model_b5" ...
__norm_pred(pred_data: ndarray, len_nn: int = 16, out: bool = True) → ndarray

Normalization of scores by hand-crafted and deep features

Note

private method

Parameters:

  • pred_data (np.ndarray) – Scores

  • len_nn (int) – The maximum size of the scores vector

  • out (bool) – Display

Returns:

Normalized scores by hand-crafted and deep features

Return type:

np.ndarray

Examples

True – 1 –

In [1]:
 1import numpy as np
 2from oceanai.modules.lab.video import Video
 3
 4video = Video()
 5
 6arr = np.array([
 7    [0.64113516, 0.6217892, 0.54451424, 0.6144415, 0.59334993],
 8    [0.6652424, 0.63606125, 0.572305, 0.63169795, 0.612515]
 9])
10
11video._Video__norm_pred(
12    pred_data = arr,
13    len_nn = 4,
14    out = True
15)
[1]:
1array([
2    [0.64113516, 0.6217892 , 0.54451424, 0.6144415 , 0.59334993],
3    [0.6652424 , 0.63606125, 0.572305  , 0.63169795, 0.612515],
4    [0.65318878, 0.62892523, 0.55840962, 0.62306972, 0.60293247],
5    [0.65318878, 0.62892523, 0.55840962, 0.62306972, 0.60293247]
6])

Error – 1 –

In [2]:
 1import numpy as np
 2from oceanai.modules.lab.video import Video
 3
 4video = Video(lang = 'en')
 5
 6arr = np.array([])
 7
 8video._Video__norm_pred(
 9    pred_data = arr,
10    len_nn = 4,
11    out = True
12)
[3]:
1[2022-10-20 22:03:17] Invalid argument types or values in "Video.__norm_pred" ...
2
3array([], dtype=float64)
_get_visual_features(path: str, reduction_fps: int = 5, window: int = 10, step: int = 5, lang: str = 'ru', last: bool = False, out: bool = True, runtime: bool = True, run: bool = True) → Tuple[ndarray, ndarray, ndarray][source]

Extracting features from a visual signal (without clearing the message output history in a Jupyter cell)

Note

protected method

Parameters:

  • path (str) – Path to video file

  • reduction_fps (int) – Frame rate reduction

  • window (int) – Signal segment window size (in frames)

  • step (int) – Signal segment window shift step (frames)

  • lang (str) – Language

  • last (bool) – Replacing the last message

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

Tuple with two np.ndarray:

  1. np.ndarray with hand-crafted features

  2. np.ndarray with deep features

  3. np.ndarray с эмоциональными предсказаниями

Return type:

Tuple[np.ndarray, np.ndarray, np.ndarray]

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video(lang = 'en')
 4
 5res_load_model_deep_fe = video.load_video_model_deep_fe(
 6    show_summary = False,
 7    out = True,
 8    runtime = True,
 9    run = True
10)
[1]:
1[2022-11-03 16:37:12] Formation of neural network architecture for obtaining deep features (video modality) ...
2
3--- Runtime: 1.564 sec. ---
In [2]:
 1video.path_to_save_ = './models'
 2video.chunk_size_ = 2000000
 3
 4url = video.weights_for_big5_['video']['fi']['fe']['googledisk']
 5
 6res_load_video_model_weights_deep_fe = video.load_video_model_weights_deep_fe(
 7    url = url,
 8    force_reload = True, out = True,
 9    runtime = True, run = True
10)
[2]:
1[2024-10-09 12:19:15] Загрузка весов нейросетевой модели для получения нейросетевых признаков (видео модальность) ...
2
3[2024-10-09 12:19:20] Загрузка файла "weights_2022-11-01_12-27-07.pth" 100.0% ...
4
5--- Время выполнения: 5.445 сек. ---
In [3]:
1path = '/Users/dl/GitHub/oceanai/oceanai/dataset/test80_01/glgfB3vFewc.004.mp4'
2
3hc_features, nn_features, _ = video.get_visual_features(
4    path = path, reduction_fps = 5,
5    window = 10, step = 5,
6    out = True, runtime = True, run = True
7)
[3]:
 1[2024-10-09 12:20:39] Извлечение признаков (экспертных и нейросетевых) из визуального сигнала ...
 2
 3[2024-10-09 12:20:46] Статистика извлеченных признаков из визуального сигнала:
 4    Общее количество сегментов с:
 5    1. экспертными признаками: 12
 6    2. нейросетевыми признаками: 12
 7    Размерность матрицы экспертных признаков одного сегмента: 10 ✕ 109
 8    Размерность матрицы с нейросетевыми признаками одного сегмента: 10 ✕ 512
 9    Понижение кадровой частоты: с 30 до 5
10
11--- Время выполнения: 7.123 сек. ---

Error – 1 –

In [4]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5path = '/Users/dl/GitHub/oceanai/oceanai/dataset/test80_01/glgfB3vFewc.004.mp4'
 6
 7hc_features, nn_features, _ = video.get_visual_features(
 8    path = path, reduction_fps = 5,
 9    window = 10, step = 5,
10    out = True, runtime = True, run = True
11)
[4]:
1[2024-10-09 12:21:55] Извлечение признаков (экспертных и нейросетевых) из визуального сигнала ...
2
3[2024-10-09 12:21:57] Что-то пошло не так ... нейросетевая архитектура модели для получения нейросетевых признаков не сформирована (видео модальность) ...
4
5--- Время выполнения: 1.202 сек. ---
get_video_union_predictions(depth: int = 1, recursive: bool = False, reduction_fps: int = 5, window: int = 10, step: int = 5, lang: str = 'ru', accuracy=True, url_accuracy: str = '', logs: bool = True, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Get video scores

Parameters:

  • depth (int) – Hierarchy depth for getting data

  • recursive (bool) – Recursive data search

  • reduction_fps (int) – Frame rate reduction

  • window (int) – Signal segment window size (in frames)

  • step (int) – Signal segment window shift step (frames)

  • lang (str) – Language

  • accuracy (bool) – Accuracy calculation

  • url_accuracy (str) – Full path to the file with ground truth scores for calculating accuracy

  • logs (bool) – If necessary, generate a LOG file

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if scores are successfully received, otherwise False

Return type:

bool

Example

get_visual_features(path: str, reduction_fps: int = 5, window: int = 10, step: int = 5, lang: str = 'ru', out: bool = True, runtime: bool = True, run: bool = True) → Tuple[ndarray, ndarray, ndarray][source]

Extracting features from a visual signal

Parameters:

  • path (str) – Path to video file

  • reduction_fps (int) – Frame rate reduction

  • window (int) – Signal segment window size (in frames)

  • step (int) – Signal segment window shift step (frames)

  • lang (str) – Language

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

Кортеж с тремя np.ndarray:

  1. np.ndarray with hand-crafted features

  2. np.ndarray with deep features

  3. np.ndarray с эмоциональными предсказаниями

Return type:

Tuple[np.ndarray, np.ndarray, np.ndarray]

Example

load_video_model_deep_fe(show_summary: bool = False, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Formation of neural network architecture for obtaining neural network features

Parameters:

  • show_summary (bool) – Displaying the formed neural network architecture of the model

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the neural network architecture of the model is formed, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_model_deep_fe(
5    show_summary = True, out = True,
6    runtime = True, run = True
7)
[1]:
  1[2024-10-09 12:22:54] Формирование нейросетевой архитектуры для получения нейросетевых признаков (видео модальность) ...
  2
  3ResNet(
  4    (conv_layer_s2_same): Conv2dSame(3, 64, kernel_size=(7, 7), stride=(2, 2), bias=False)
  5    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
  6    (relu): ReLU()
  7    (max_pool): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
  8    (layer1): Sequential(
  9    (0): Bottleneck(
 10    (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 11    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 12    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 13    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 14    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 15    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 16    (i_downsample): Sequential(
 17        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 18        (1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 19    )
 20    (relu): ReLU()
 21    )
 22    (1): Bottleneck(
 23    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 24    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 25    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 26    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 27    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 28    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 29    (relu): ReLU()
 30    )
 31    (2): Bottleneck(
 32    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 33    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 34    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 35    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 36    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 37    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 38    (relu): ReLU()
 39    )
 40    )
 41    (layer2): Sequential(
 42    (0): Bottleneck(
 43    (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
 44    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 45    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 46    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 47    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 48    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 49    (i_downsample): Sequential(
 50        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
 51        (1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 52    )
 53    (relu): ReLU()
 54    )
 55    (1): Bottleneck(
 56    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 57    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 58    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 59    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 60    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 61    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 62    (relu): ReLU()
 63    )
 64    (2): Bottleneck(
 65    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 66    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 67    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 68    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 69    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 70    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 71    (relu): ReLU()
 72    )
 73    (3): Bottleneck(
 74    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 75    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 76    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 77    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 78    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 79    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 80    (relu): ReLU()
 81    )
 82    )
 83    (layer3): Sequential(
 84    (0): Bottleneck(
 85    (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
 86    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 87    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 88    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 89    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
 90    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 91    (i_downsample): Sequential(
 92        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
 93        (1): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 94    )
 95    (relu): ReLU()
 96    )
 97    (1): Bottleneck(
 98    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 99    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
100    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
101    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
102    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
103    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
104    (relu): ReLU()
105    )
106    (2): Bottleneck(
107    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
108    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
109    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
110    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
111    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
112    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
113    (relu): ReLU()
114    )
115    (3): Bottleneck(
116    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
117    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
118    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
119    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
120    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
121    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
122    (relu): ReLU()
123    )
124    (4): Bottleneck(
125    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
126    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
127    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
128    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
129    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
130    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
131    (relu): ReLU()
132    )
133    (5): Bottleneck(
134    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
135    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
136    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
137    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
138    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
139    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
140    (relu): ReLU()
141    )
142    )
143    (layer4): Sequential(
144    (0): Bottleneck(
145    (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
146    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
147    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
148    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
149    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
150    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
151    (i_downsample): Sequential(
152        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
153        (1): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
154    )
155    (relu): ReLU()
156    )
157    (1): Bottleneck(
158    (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
159    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
160    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
161    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
162    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
163    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
164    (relu): ReLU()
165    )
166    (2): Bottleneck(
167    (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
168    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
169    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
170    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
171    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
172    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
173    (relu): ReLU()
174    )
175    )
176    (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
177    (fc1): Linear(in_features=2048, out_features=512, bias=True)
178    (relu1): ReLU()
179    (fc2): Linear(in_features=512, out_features=7, bias=True)
180)
181--- Время выполнения: 0.222 сек. ---
182
183True

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_model_deep_fe(
5    show_summary = 1, out = True,
6    runtime = True, run = True
7)
[2]:
1[2022-11-01 12:21:23] Invalid argument types or values in "Video.load_video_model_deep_fe" ...
2
3False
load_video_model_hc(lang: str = 'ru', show_summary: bool = False, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Formation of the neural network architecture of the model for obtaining scores by hand-crafted features

Parameters:

  • lang (str) – Language

  • show_summary (bool) – Displaying the formed neural network architecture of the model

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the neural network architecture of the model is formed, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4video.load_video_model_hc(
5    lang = 'en',
6    show_summary = False, out = True,
7    runtime = True, run = True
8)
[1]:
1[2022-10-25 16:37:43] Formation of the neural network architecture of the model for obtaining scores by hand-crafted features (video modality) ...
2
3--- Runtime: 0.659 sec. ---
4
5True

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4video.load_video_model_hc(
5    lang = 'en',
6    show_summary = 1, out = True,
7    runtime = True, run = True
8)
[2]:
1[2022-10-26 12:27:41] Invalid argument types or values in "Video.load_video_model_hc" ...
2
3False
load_video_model_nn(show_summary: bool = False, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Formation of a neural network architecture for obtaining scores by deep features

Parameters:

  • show_summary (bool) – Displaying the formed neural network architecture of the model

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the neural network architecture of the model is formed, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_model_nn(
5    show_summary = True, out = True,
6    runtime = True, run = True
7)
[1]:
 1[2024-10-09 12:49:36] Формирование нейросетевой архитектуры для получения оценок по нейросетевым признакам (видео модальность) ...
 2
 3video_model_nn(
 4    (lstm1): LSTM(512, 1024, batch_first=True)
 5    (dropout1): Dropout(p=0.2, inplace=False)
 6    (fc): Linear(in_features=1024, out_features=5, bias=True)
 7)
 8--- Время выполнения: 0.052 сек. ---
 9
10True

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_model_nn(
5    show_summary = 1, out = True,
6    runtime = True, run = True
7)
[2]:
1[2022-10-27 14:47:22] Invalid argument types or values in "Video.load_video_model_nn" ...
2
3False
load_video_model_weights_deep_fe(url: str, force_reload: bool = True, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Downloading weights of a neural network model to obtain neural network features

Parameters:

  • url (str) – Full path to the file with weights of the neural network model

  • force_reload (bool) – Forced download of a file with weights of a neural network model from the network

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the weights of the neural network model are downloaded, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.load_video_model_deep_fe(
6    show_summary = False, out = True,
7    runtime = True, run = True
8)
[1]:
1[2022-11-01 12:41:59] Formation of neural network architecture for obtaining deep features (video modality) ...
2
3--- Runtime: 1.306 sec. ---
4
5True
In [2]:
 1video.path_to_save_ = './models'
 2video.chunk_size_ = 2000000
 3
 4url = video.weights_for_big5_['video']['fi']['fe']['googledisk']
 5
 6video.load_video_model_weights_deep_fe(
 7    url = url,
 8    force_reload = True,
 9    out = True,
10    runtime = True,
11    run = True
12)
[2]:
1[2024-10-09 13:00:35] Загрузка весов нейросетевой модели для получения нейросетевых признаков (видео модальность) ...
2
3[2024-10-09 13:00:41] Загрузка файла "weights_2022-11-01_12-27-07.pth" 100.0% ...
4
5--- Время выполнения: 5.557 сек. ---
6
7True

Error – 1 –

In [3]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8url = video.weights_for_big5_['video']['fi']['fe']['googledisk']
 9
10video.load_video_model_weights_deep_fe(
11    url = url,
12    force_reload = True,
13    out = True,
14    runtime = True,
15    run = True
16)
[3]:
1[2024-10-09 13:01:48] Загрузка весов нейросетевой модели для получения нейросетевых признаков (видео модальность) ...
2
3[2024-10-09 13:01:53] Загрузка файла "weights_2022-11-01_12-27-07.pth" 100.0% ...
4
5[2024-10-09 13:01:53] Что-то пошло не так ... нейросетевая архитектура модели для получения нейросетевых признаков не сформирована (видео модальность) ...
6
7--- Время выполнения: 4.712 сек. ---
8
9False
load_video_model_weights_hc(url: str, force_reload: bool = True, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Downloading the weights of the neural network model to obtain scores by hand-crafted features

Parameters:

  • url (str) – Full path to the file with weights of the neural network model

  • force_reload (bool) – Forced download of a file with weights of a neural network model from the network

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the weights of the neural network model are downloaded, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video()
4
5video.load_video_model_hc(
6    lang = 'en',
7    show_summary = False, out = True,
8    runtime = True, run = True
9)
[1]:
1[2022-10-27 12:55:31] Formation of the neural network architecture of the model for obtaining scores by hand-crafted features (video modality) ...
2
3--- Runtime: 0.606 sec. ---
4
5True
In [2]:
 1video.path_to_save_ = './models'
 2video.chunk_size_ = 2000000
 3
 4url = video.weights_for_big5_['video']['fi']['hc']['googledisk']
 5
 6video.load_video_model_weights_hc(
 7    url = url,
 8    force_reload = True,
 9    out = True,
10    runtime = True,
11    run = True
12)
[2]:
1[2024-10-09 13:06:56] Загрузка весов нейросетевой модели для получения оценок по экспертным признакам (видео модальность) ...
2
3[2024-10-09 13:06:58] Загрузка файла "weights_2022-08-27_18-53-35.pth" 100.0% ...
4
5--- Время выполнения: 2.49 сек. ---
6
7True

Error – 1 –

In [3]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8url = video.weights_for_big5_['video']['fi']['hc']['googledisk']
 9
10video.load_video_model_weights_hc(
11    url = url,
12    force_reload = True,
13    out = True,
14    runtime = True,
15    run = True
16)
[3]:
1[2024-10-09 13:07:56] Загрузка весов нейросетевой модели для получения оценок по экспертным признакам (видео модальность) ...
2
3[2024-10-09 13:07:59] Загрузка файла "weights_2022-08-27_18-53-35.pth" 100.0% ...
4
5[2024-10-09 13:07:59] Что-то пошло не так ... нейросетевая архитектура модели для получения оценок по экспертным признакам не сформирована (видео модальность) ...
6
7--- Время выполнения: 2.381 сек. ---
8
9False
load_video_model_weights_nn(url: str, force_reload: bool = True, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Downloading the weights of the neural network model to obtain scores for deep features

Parameters:

  • url (str) – Full path to the file with weights of the neural network model

  • force_reload (bool) – Forced download of a file with weights of a neural network model from the network

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the weights of the neural network model are downloaded, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.load_video_model_nn(
6    show_summary = False, out = True,
7    runtime = True, run = True
8)
[1]:
1[2022-10-27 15:17:13] Formation of a neural network architecture for obtaining scores by deep features (video modality) ...
2
3--- Runtime: 1.991 sec. ---
4
5True
In [2]:
 1video.path_to_save_ = './models'
 2video.chunk_size_ = 2000000
 3
 4url = video.weights_for_big5_['video']['fi']['nn']['googledisk']
 5
 6video.load_video_model_weights_nn(
 7    url = url,
 8    force_reload = True,
 9    out = True,
10    runtime = True,
11    run = True
12)
[2]:
1[2024-10-09 13:09:03] Загрузка весов нейросетевой модели для получения оценок по нейросетевым признакам (видео модальность) ...
2
3[2024-10-09 13:09:08] Загрузка файла "weights_2022-03-22_16-31-48.pth" 100.0% ...
4
5--- Время выполнения: 5.798 сек. ---
6
7True

Error – 1 –

In [3]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8url = video.weights_for_big5_['video']['fi']['nn']['googledisk']
 9
10video.load_video_model_weights_nn(
11    url = url,
12    force_reload = True,
13    out = True,
14    runtime = True,
15    run = True
16)
[3]:
1[2024-10-09 13:09:56] Загрузка весов нейросетевой модели для получения оценок по нейросетевым признакам (видео модальность) ...
2
3[2024-10-09 13:10:02] Загрузка файла "weights_2022-03-22_16-31-48.pth" 100.0% ...
4
5[2024-10-09 13:10:02] Что-то пошло не так ... нейросетевая архитектура модели для получения оценок по нейросетевым признакам не сформирована (видео модальность) ...
6
7--- Время выполнения: 5.9 сек. ---
8
9False
load_video_models_b5(show_summary: bool = False, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Formation of neural network architectures of models for obtaining the personality traits scores

Parameters:

  • show_summary (bool) – Displaying the last generated neural network architecture of models

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the neural network architectures of the model are formed, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_models_b5(
5    show_summary = True, out = True,
6    runtime = True, run = True
7)
[1]:
1[2024-10-09 13:12:19] Формирование нейросетевых архитектур моделей для получения результатов оценки персональных качеств (видео модальность) ...
2
3video_model_b5(
4(fc): Linear(in_features=32, out_features=1, bias=True)
5(sigmoid): Sigmoid()
6)
7--- Время выполнения: 0.009 сек. ---
8
9True

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4video.load_video_models_b5(
5    show_summary = 1, out = True,
6    runtime = True, run = True
7)
[2]:
1[2022-11-04 15:30:15] Invalid argument types or values in "Video.load_video_models_b5" ...
2
3False
load_video_models_weights_b5(url_openness: str, url_conscientiousness: str, url_extraversion: str, url_agreeableness: str, url_non_neuroticism: str, force_reload: bool = True, out: bool = True, runtime: bool = True, run: bool = True) → bool[source]

Downloading the weights of neural network models to obtain the personality traits scores

Parameters:

  • url_openness (str) – Full path to the file with the weights of the neural network model (openness)

  • url_conscientiousness (str) – Full path to the file with the weights of the neural network model (conscientiousness)

  • url_extraversion (str) – Full path to the file with the weights of the neural network model (extraversion)

  • url_agreeableness (str) – Full path to the file with the weights of the neural network model (agreeableness)

  • url_non_neuroticism (str) – Full path to the file with the weights of the neural network model (non-neuroticism)

  • force_reload (bool) – Forced download of files with weights of neural network models from the network

  • out (bool) – Display

  • runtime (bool) – Runtime count

  • run (bool) – Run blocking

Returns:

True if the weights of the neural network models are downloaded, otherwise False

Return type:

bool

Examples

True – 1 –

In [1]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.load_video_models_b5(
6    show_summary = False, out = True,
7    runtime = True, run = True
8)
[1]:
1[2022-11-04 18:56:41] Formation of neural network architectures of models for obtaining the personality traits scores (video modality) ...
2
3--- Runtime: 0.117 sec. ---
4
5True
In [2]:
 1video.path_to_save_ = './models'
 2video.chunk_size_ = 2000000
 3
 4url_openness = video.weights_for_big5_['video']['fi']['b5']['openness']['googledisk']
 5url_conscientiousness = video.weights_for_big5_['video']['fi']['b5']['conscientiousness']['googledisk']
 6url_extraversion = video.weights_for_big5_['video']['fi']['b5']['extraversion']['googledisk']
 7url_agreeableness = video.weights_for_big5_['video']['fi']['b5']['agreeableness']['googledisk']
 8url_non_neuroticism = video.weights_for_big5_['video']['fi']['b5']['non_neuroticism']['googledisk']
 9
10video.load_video_models_weights_b5(
11    url_openness = url_openness,
12    url_conscientiousness = url_conscientiousness,
13    url_extraversion = url_extraversion,
14    url_agreeableness = url_agreeableness,
15    url_non_neuroticism = url_non_neuroticism,
16    force_reload = True,
17    out = True,
18    runtime = True,
19    run = True
20)
[2]:
 1[2024-10-09 13:14:48] Загрузка весов нейросетевых моделей для получения результатов оценки персональных качеств (видео модальность) ...
 2
 3[2024-10-09 13:14:50] Загрузка файла "weights_2022-06-15_16-46-30.pth" 100.0% ... Открытость опыту
 4
 5[2024-10-09 13:14:52] Загрузка файла "weights_2022-06-15_16-48-50.pth" 100.0% ... Добросовестность
 6
 7[2024-10-09 13:14:55] Загрузка файла "weights_2022-06-15_16-54-06.pth" 100.0% ... Экстраверсия
 8
 9[2024-10-09 13:14:57] Загрузка файла "weights_2022-06-15_17-02-03.pth" 100.0% ... Доброжелательность
10
11[2024-10-09 13:15:00] Загрузка файла "weights_2022-06-15_17-06-15.pth" 100.0% ... Эмоциональная стабильность
12
13--- Время выполнения: 11.832 сек. ---
14
15True

Error – 1 –

In [3]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.path_to_save_ = './models'
 6video.chunk_size_ = 2000000
 7
 8url_openness = video.weights_for_big5_['video']['fi']['b5']['openness']['googledisk']
 9url_conscientiousness = video.weights_for_big5_['video']['fi']['b5']['conscientiousness']['googledisk']
10url_extraversion = video.weights_for_big5_['video']['fi']['b5']['extraversion']['googledisk']
11url_agreeableness = video.weights_for_big5_['video']['fi']['b5']['agreeableness']['googledisk']
12url_non_neuroticism = video.weights_for_big5_['video']['fi']['b5']['non_neuroticism']['googledisk']
13
14video.load_video_models_weights_b5(
15    url_openness = url_openness,
16    url_conscientiousness = url_conscientiousness,
17    url_extraversion = url_extraversion,
18    url_agreeableness = url_agreeableness,
19    url_non_neuroticism = url_non_neuroticism,
20    force_reload = True,
21    out = True,
22    runtime = True,
23    run = True
24)
[3]:
 1[2024-10-09 13:16:08] Загрузка весов нейросетевых моделей для получения результатов оценки персональных качеств (видео модальность) ...
 2
 3[2024-10-09 13:16:10] Загрузка файла "weights_2022-06-15_16-46-30.pth" 100.0% ...
 4
 5[2024-10-09 13:16:10] Что-то пошло не так ... не удалось загрузить веса нейросетевой модели ... Открытость опыту
 6
 7    Файл: /Users/dl/@DmitryRyumin/Python/envs/OCEANAI/lib/python3.9/site-packages/oceanai/modules/lab/video.py
 8    Линия: 3144
 9    Метод: load_video_models_weights_b5
10    Тип ошибки: AttributeError
11
12[2024-10-09 13:16:13] Загрузка файла "weights_2022-06-15_16-48-50.pth" 100.0% ...
13
14[2024-10-09 13:16:13] Что-то пошло не так ... не удалось загрузить веса нейросетевой модели ... Добросовестность
15
16    Файл: /Users/dl/@DmitryRyumin/Python/envs/OCEANAI/lib/python3.9/site-packages/oceanai/modules/lab/video.py
17    Линия: 3144
18    Метод: load_video_models_weights_b5
19    Тип ошибки: AttributeError
20
21[2024-10-09 13:16:16] Загрузка файла "weights_2022-06-15_16-54-06.pth" 100.0% ...
22
23[2024-10-09 13:16:16] Что-то пошло не так ... не удалось загрузить веса нейросетевой модели ... Экстраверсия
24
25    Файл: /Users/dl/@DmitryRyumin/Python/envs/OCEANAI/lib/python3.9/site-packages/oceanai/modules/lab/video.py
26    Линия: 3144
27    Метод: load_video_models_weights_b5
28    Тип ошибки: AttributeError
29
30[2024-10-09 13:16:19] Загрузка файла "weights_2022-06-15_17-02-03.pth" 100.0% ...
31
32[2024-10-09 13:16:19] Что-то пошло не так ... не удалось загрузить веса нейросетевой модели ... Доброжелательность
33
34    Файл: /Users/dl/@DmitryRyumin/Python/envs/OCEANAI/lib/python3.9/site-packages/oceanai/modules/lab/video.py
35    Линия: 3144
36    Метод: load_video_models_weights_b5
37    Тип ошибки: AttributeError
38
39[2024-10-09 13:16:21] Загрузка файла "weights_2022-06-15_17-06-15.pth" 100.0% ...
40
41[2024-10-09 13:16:21] Что-то пошло не так ... не удалось загрузить веса нейросетевой модели ... Эмоциональная стабильность
42
43    Файл: /Users/dl/@DmitryRyumin/Python/envs/OCEANAI/lib/python3.9/site-packages/oceanai/modules/lab/video.py
44    Линия: 3144
45    Метод: load_video_models_weights_b5
46    Тип ошибки: AttributeError
47
48--- Время выполнения: 13.055 сек. ---
49
50False
property video_model_deep_fe_: Module | None

Получение нейросетевой модели nn.Module для получения нейросетевых признаков

Returns:

Нейросетевая модель nn.Module или None

Return type:

Optional[nn.Module]

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video(lang = 'en')
 4
 5video.load_video_model_deep_fe(
 6    show_summary = False, out = True,
 7    runtime = True, run = True
 8)
 9
10video.video_model_deep_fe_
[1]:
  1[2024-10-09 13:17:09] Формирование нейросетевой архитектуры для получения нейросетевых признаков (видео модальность) ...
  2
  3--- Время выполнения: 0.228 сек. ---
  4
  5ResNet(
  6    (conv_layer_s2_same): Conv2dSame(3, 64, kernel_size=(7, 7), stride=(2, 2), bias=False)
  7    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
  8    (relu): ReLU()
  9    (max_pool): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
 10    (layer1): Sequential(
 11    (0): Bottleneck(
 12    (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 13    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 14    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 15    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 16    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 17    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 18    (i_downsample): Sequential(
 19        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 20        (1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 21    )
 22    (relu): ReLU()
 23    )
 24    (1): Bottleneck(
 25    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 26    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 27    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 28    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 29    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 30    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 31    (relu): ReLU()
 32    )
 33    (2): Bottleneck(
 34    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
 35    (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 36    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 37    (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 38    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
 39    (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 40    (relu): ReLU()
 41    )
 42    )
 43    (layer2): Sequential(
 44    (0): Bottleneck(
 45    (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
 46    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 47    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 48    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 49    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 50    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 51    (i_downsample): Sequential(
 52        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
 53        (1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 54    )
 55    (relu): ReLU()
 56    )
 57    (1): Bottleneck(
 58    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 59    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 60    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 61    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 62    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 63    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 64    (relu): ReLU()
 65    )
 66    (2): Bottleneck(
 67    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 68    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 69    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 70    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 71    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 72    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 73    (relu): ReLU()
 74    )
 75    (3): Bottleneck(
 76    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
 77    (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 78    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 79    (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 80    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
 81    (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 82    (relu): ReLU()
 83    )
 84    )
 85    (layer3): Sequential(
 86    (0): Bottleneck(
 87    (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
 88    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 89    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
 90    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 91    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
 92    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 93    (i_downsample): Sequential(
 94        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
 95        (1): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
 96    )
 97    (relu): ReLU()
 98    )
 99    (1): Bottleneck(
100    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
101    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
102    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
103    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
104    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
105    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
106    (relu): ReLU()
107    )
108    (2): Bottleneck(
109    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
110    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
111    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
112    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
113    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
114    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
115    (relu): ReLU()
116    )
117    (3): Bottleneck(
118    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
119    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
120    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
121    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
122    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
123    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
124    (relu): ReLU()
125    )
126    (4): Bottleneck(
127    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
128    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
129    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
130    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
131    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
132    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
133    (relu): ReLU()
134    )
135    (5): Bottleneck(
136    (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
137    (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
138    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
139    (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
140    (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
141    (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
142    (relu): ReLU()
143    )
144    )
145    (layer4): Sequential(
146    (0): Bottleneck(
147    (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
148    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
149    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
150    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
151    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
152    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
153    (i_downsample): Sequential(
154        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
155        (1): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
156    )
157    (relu): ReLU()
158    )
159    (1): Bottleneck(
160    (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
161    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
162    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
163    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
164    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
165    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
166    (relu): ReLU()
167    )
168    (2): Bottleneck(
169    (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
170    (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
171    (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)
172    (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
173    (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
174    (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)
175    (relu): ReLU()
176    )
177    )
178    (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
179    (fc1): Linear(in_features=2048, out_features=512, bias=True)
180    (relu1): ReLU()
181    (fc2): Linear(in_features=512, out_features=7, bias=True)
182)

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.video_model_deep_fe_
[2]:
1
property video_model_hc_: Module | None

Получение нейросетевой модели nn.Module для получения оценок по экспертным признакам

Returns:

Нейросетевая модель nn.Module или None

Return type:

Optional[nn.Module]

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.load_video_model_hc(
 6    lang = 'en',
 7    show_summary = False, out = True,
 8    runtime = True, run = True
 9)
10
11video.video_model_hc_
[1]:
 1[2024-10-09 13:19:24] Формирование нейросетевой архитектуры модели для получения оценок по экспертным признакам (видео модальность) ...
 2
 3--- Время выполнения: 0.005 сек. ---
 4
 5video_model_hc(
 6    (lstm1): LSTM(115, 64, batch_first=True)
 7    (dropout1): Dropout(p=0.2, inplace=False)
 8    (lstm2): LSTM(64, 128, batch_first=True)
 9    (dropout2): Dropout(p=0.2, inplace=False)
10    (fc): Linear(in_features=128, out_features=5, bias=True)
11)

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.video_model_hc_
[2]:
1
property video_model_nn_: Module | None

Получение нейросетевой модели nn.Module для получения оценок по нейросетевым признакам

Returns:

Нейросетевая модель nn.Module или None

Return type:

Optional[nn.Module]

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video(lang = 'en')
 4
 5video.load_video_model_nn(
 6    show_summary = False, out = True,
 7    runtime = True, run = True
 8)
 9
10video.video_model_nn_
[1]:
1[2024-10-09 13:20:47] Формирование нейросетевой архитектуры для получения оценок по нейросетевым признакам (видео модальность) ...
2
3--- Время выполнения: 0.055 сек. ---
4
5video_model_nn(
6    (lstm1): LSTM(512, 1024, batch_first=True)
7    (dropout1): Dropout(p=0.2, inplace=False)
8    (fc): Linear(in_features=1024, out_features=5, bias=True)
9)

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.video_model_nn_
[2]:
1
property video_models_b5_: Dict[str, Module | None]

Получение нейросетевых моделей nn.Module для получения результатов оценки персональных качеств

Returns:

Словарь с нейросетевыми моделями nn.Module

Return type:

Dict

Examples

True – 1 –

In [1]:
 1from oceanai.modules.lab.video import Video
 2
 3video = Video()
 4
 5video.load_video_models_b5(
 6    show_summary = False, out = True,
 7    runtime = True, run = True
 8)
 9
10video.video_models_b5_
[1]:
 1[2024-10-09 13:21:52] Формирование нейросетевых архитектур моделей для получения результатов оценки персональных качеств (видео модальность) ...
 2
 3--- Время выполнения: 0.004 сек. ---
 4
 5{'openness': video_model_b5(
 6    (fc): Linear(in_features=32, out_features=1, bias=True)
 7    (sigmoid): Sigmoid()
 8),
 9'conscientiousness': video_model_b5(
10    (fc): Linear(in_features=32, out_features=1, bias=True)
11    (sigmoid): Sigmoid()
12),
13'extraversion': video_model_b5(
14    (fc): Linear(in_features=32, out_features=1, bias=True)
15    (sigmoid): Sigmoid()
16),
17'agreeableness': video_model_b5(
18    (fc): Linear(in_features=32, out_features=1, bias=True)
19    (sigmoid): Sigmoid()
20),
21'non-neuroticism': video_model_b5(
22    (fc): Linear(in_features=32, out_features=1, bias=True)
23    (sigmoid): Sigmoid()
24)}

Error – 1 –

In [2]:
1from oceanai.modules.lab.video import Video
2
3video = Video(lang = 'en')
4
5video.video_models_b5_
[2]:
1{
2    'openness': None,
3    'conscientiousness': None,
4    'extraversion': None,
5    'agreeableness': None,
6    'non_neuroticism': None
7}