{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Формирование нейросетевой архитектуры модели и загрузка ее весов для получения нейросетевых признаков (видео модальность)\n",
"\n",
"
\n",
"\n",
"> - `_b5.video_model_deep_fe_` - Нейросетевая модель **nn.Module** для получения нейросетевых признаков"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"nbsphinx": "hidden",
"tags": []
},
"outputs": [],
"source": [
"import os # Взаимодействие с файловой системой\n",
"import sys # Доступ к некоторым переменным и функциям Python\n",
"\n",
"PATH_TO_SOURCE = os.path.abspath(os.path.dirname(globals()['_dh'][0]))\n",
"PATH_TO_ROOT = os.path.join(PATH_TO_SOURCE, '..', '..', '..')\n",
"\n",
"sys.path.insert(0, os.path.abspath(PATH_TO_ROOT))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Импорт необходимых инструментов"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"from oceanai.modules.lab.build import Run"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Сборка"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
"**[2024-10-08 16:26:05] OCEANAI - персональные качества личности человека:**
Авторы:
Рюмина Елена [ryumina_ev@mail.ru]
Рюмин Дмитрий [dl_03.03.1991@mail.ru]
Карпов Алексей [karpov@iias.spb.su]
Сопровождающие:
Рюмина Елена [ryumina_ev@mail.ru]
Рюмин Дмитрий [dl_03.03.1991@mail.ru]
Версия: 1.0.0a40
Лицензия: BSD License"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"_b5 = Run(\n",
" lang = 'ru', # Язык\n",
" color_simple = '#333', # Цвет обычного текста (шестнадцатеричный код)\n",
" color_info = '#1776D2', # Цвет текста содержащего информацию (шестнадцатеричный код)\n",
" color_err = '#FF0000', # Цвет текста содержащего ошибку (шестнадцатеричный код)\n",
" color_true = '#008001', # Цвет текста содержащего положительную информацию (шестнадцатеричный код)\n",
" bold_text = True, # Жирное начертание текста\n",
" num_to_df_display = 30, # Количество строк для отображения в таблицах\n",
" text_runtime = 'Время выполнения', # Текст времени выполнения\n",
" metadata = True # Отображение информации о библиотеке\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Формирование нейросетевой архитектуры модели (`FI V2`)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
"**[2024-10-08 16:26:05] Формирование нейросетевой архитектуры для получения нейросетевых признаков (видео модальность) ...** "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**--- Время выполнения: 0.101 сек. ---**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"res_load_video_model_deep_fe = _b5.load_video_model_deep_fe(\n",
" show_summary = False, # Отображение сформированной нейросетевой архитектуры модели\n",
" out = True, # Отображение\n",
" runtime = True, # Подсчет времени выполнения\n",
" run = True # Блокировка выполнения\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Загрузка весов нейросетевой модели"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/markdown": [
"**[2024-10-08 16:26:06] Загрузка весов нейросетевой модели для получения нейросетевых признаков (видео модальность) ...** "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**[2024-10-08 16:26:09] Загрузка файла \"weights_2022-11-01_12-27-07.pth\" 100.0% ...** "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/markdown": [
"**--- Время выполнения: 3.537 сек. ---**"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Настройки ядра\n",
"_b5.path_to_save_ = './models' # Директория для сохранения файла\n",
"_b5.chunk_size_ = 2000000 # Размер загрузки файла из сети за 1 шаг\n",
"\n",
"url = _b5.weights_for_big5_['video']['fi']['fe']['googledisk']\n",
"\n",
"res_load_video_model_weights_deep_fe = _b5.load_video_model_weights_deep_fe(\n",
" url = url, # Полный путь к файлу с весами нейросетевой модели\n",
" force_reload = True, # Принудительная загрузка файла с весами нейросетевой модели из сети\n",
" out = True, # Отображение\n",
" runtime = True, # Подсчет времени выполнения\n",
" run = True # Блокировка выполнения\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Отображение сформированной нейросетевой архитектуры модели"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"ResNet(\n",
" (conv_layer_s2_same): Conv2dSame(3, 64, kernel_size=(7, 7), stride=(2, 2), bias=False)\n",
" (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" (max_pool): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
" (layer1): Sequential(\n",
" (0): Bottleneck(\n",
" (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (i_downsample): Sequential(\n",
" (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" )\n",
" (relu): ReLU()\n",
" )\n",
" (1): Bottleneck(\n",
" (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (2): Bottleneck(\n",
" (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(64, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" )\n",
" (layer2): Sequential(\n",
" (0): Bottleneck(\n",
" (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (i_downsample): Sequential(\n",
" (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" )\n",
" (relu): ReLU()\n",
" )\n",
" (1): Bottleneck(\n",
" (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (2): Bottleneck(\n",
" (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (3): Bottleneck(\n",
" (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(128, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" )\n",
" (layer3): Sequential(\n",
" (0): Bottleneck(\n",
" (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (i_downsample): Sequential(\n",
" (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (1): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" )\n",
" (relu): ReLU()\n",
" )\n",
" (1): Bottleneck(\n",
" (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (2): Bottleneck(\n",
" (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (3): Bottleneck(\n",
" (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (4): Bottleneck(\n",
" (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (5): Bottleneck(\n",
" (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(256, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(1024, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" )\n",
" (layer4): Sequential(\n",
" (0): Bottleneck(\n",
" (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (i_downsample): Sequential(\n",
" (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
" (1): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" )\n",
" (relu): ReLU()\n",
" )\n",
" (1): Bottleneck(\n",
" (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" (2): Bottleneck(\n",
" (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm1): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=same, bias=False)\n",
" (batch_norm2): BatchNorm2d(512, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
" (batch_norm3): BatchNorm2d(2048, eps=0.001, momentum=0.99, affine=True, track_running_stats=True)\n",
" (relu): ReLU()\n",
" )\n",
" )\n",
" (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
" (fc1): Linear(in_features=2048, out_features=512, bias=True)\n",
" (relu1): ReLU()\n",
" (fc2): Linear(in_features=512, out_features=7, bias=True)\n",
")"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"_b5.video_model_deep_fe_"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.11"
}
},
"nbformat": 4,
"nbformat_minor": 4
}