{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "56f79218-026b-403d-8caa-d5aae41bb3e0",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:37.162054Z",
     "start_time": "2024-03-02T07:57:31.733202900Z"
    }
   },
   "outputs": [],
   "source": [
    "from torch import nn\n",
    "from torch import optim\n",
    "from torchvision import transforms\n",
    "from torch.optim import lr_scheduler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a64dd1a6-0197-424b-b109-f88787b18164",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.404732Z",
     "start_time": "2024-03-02T07:57:37.165358400Z"
    }
   },
   "outputs": [],
   "source": [
    "from generate_c3d_model import generate_model\n",
    "from train import train_epoch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "33b89569-a272-4d8a-8ece-e0fc3054e9bb",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.431727100Z",
     "start_time": "2024-03-02T07:57:38.406924200Z"
    }
   },
   "outputs": [],
   "source": [
    "from datasets.nv import NV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "41220539-449a-478f-954e-ecf9982388e5",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.446055400Z",
     "start_time": "2024-03-02T07:57:38.426055300Z"
    }
   },
   "outputs": [],
   "source": [
    "from utils import *\n",
    "from target_transforms import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1e969200-a07e-445f-b638-a5d84b6892d8",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.459855800Z",
     "start_time": "2024-03-02T07:57:38.440573600Z"
    }
   },
   "outputs": [],
   "source": [
    "from logger.logger import get_logger\n",
    "logger = get_logger(__name__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b7d5fa6a-adae-47cc-a5c2-605f3773ed1e",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.491833Z",
     "start_time": "2024-03-02T07:57:38.454971500Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<torch._C.Generator at 0x27d81b14e50>"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.manual_seed(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bedb9441-e776-4f4f-b14e-60e99e78118b",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:57:38.492929700Z",
     "start_time": "2024-03-02T07:57:38.473197300Z"
    }
   },
   "outputs": [],
   "source": [
    "arch = '{}'.format('c3d')\n",
    "n_epochs = 35\n",
    "n_classes = 26\n",
    "sample_size = 112\n",
    "sample_duration = 10\n",
    "ft_portion = \"last_layer\"\n",
    "downsample = 2\n",
    "scale_step = 0.84089641525\n",
    "scales = [1.0]\n",
    "for i in range(1, 5):\n",
    "    scales.append(scales[-1] * scale_step)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fc3d13b8-6f90-42bf-aebc-ebbbf3a2e7e8",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:58:00.830367500Z",
     "start_time": "2024-03-02T07:58:00.069619100Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "generate_c3d_model 2024-03-02 10:58:00,066 INFO Torch version: 2.2.1\n",
      "generate_c3d_model 2024-03-02 10:58:00,068 INFO Is CUDA enabled? True\n",
      "generate_c3d_model 2024-03-02 10:58:00,565 INFO Total number of trainable parameters: 31913114\n",
      "generate_c3d_model 2024-03-02 10:58:00,567 INFO  Converting the pretrained model to RGB+D init model\n",
      "generate_c3d_model 2024-03-02 10:58:00,810 INFO  Done. RGB-D model ready.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "last_layer\n"
     ]
    }
   ],
   "source": [
    "model, parameters = generate_model(n_classes, sample_size, ft_portion)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f547dfcb-bded-41a1-b7c5-314c51cee32c",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:58:04.335008400Z",
     "start_time": "2024-03-02T07:58:04.312769200Z"
    }
   },
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss()\n",
    "criterion = criterion.cuda()\n",
    "spatial_transform = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0, 0, 0], [1, 1, 1])\n",
    "])\n",
    "temporal_transform = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0, 0, 0], [1, 1, 1])])\n",
    "target_transform = ClassLabel()\n",
    "optimizer = optim.SGD(\n",
    "    parameters,\n",
    "    lr=0.1,\n",
    "    momentum=0.9,\n",
    "    dampening=0.9,\n",
    "    weight_decay=1e-3,\n",
    "    nesterov=False)\n",
    "\n",
    "scheduler = lr_scheduler.ReduceLROnPlateau(\n",
    "    optimizer, 'min', patience=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f024f129-7d3f-42b3-af89-36612b5f2c43",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:58:09.870821600Z",
     "start_time": "2024-03-02T07:58:09.730071200Z"
    }
   },
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: './annotation_nvGesture_v1/nvall_but_None.json'",
     "output_type": "error",
     "traceback": [
      "\u001B[1;31m---------------------------------------------------------------------------\u001B[0m",
      "\u001B[1;31mFileNotFoundError\u001B[0m                         Traceback (most recent call last)",
      "Cell \u001B[1;32mIn[12], line 1\u001B[0m\n\u001B[1;32m----> 1\u001B[0m training_data \u001B[38;5;241m=\u001B[39m \u001B[43mNV\u001B[49m\u001B[43m(\u001B[49m\n\u001B[0;32m      2\u001B[0m \u001B[43m    \u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43m./nvGesture_v1.1/nvGesture_v1\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m,\u001B[49m\n\u001B[0;32m      3\u001B[0m \u001B[43m    \u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43m./annotation_nvGesture_v1/nvall_but_None.json\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m,\u001B[49m\n\u001B[0;32m      4\u001B[0m \u001B[43m    \u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43mtraining\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m,\u001B[49m\n\u001B[0;32m      5\u001B[0m \u001B[43m    \u001B[49m\u001B[43mspatial_transform\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mspatial_transform\u001B[49m\u001B[43m,\u001B[49m\n\u001B[0;32m      6\u001B[0m \u001B[43m    \u001B[49m\u001B[43mtemporal_transform\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mtemporal_transform\u001B[49m\u001B[43m,\u001B[49m\n\u001B[0;32m      7\u001B[0m \u001B[43m    \u001B[49m\u001B[43mtarget_transform\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mtarget_transform\u001B[49m\u001B[43m,\u001B[49m\n\u001B[0;32m      8\u001B[0m \u001B[43m    \u001B[49m\u001B[43msample_duration\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43msample_duration\u001B[49m\u001B[43m,\u001B[49m\n\u001B[0;32m      9\u001B[0m \u001B[43m    \u001B[49m\u001B[43mmodality\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[38;5;124;43mRGB-D\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[43m)\u001B[49m\n",
      "File \u001B[1;32mD:\\current\\gesture\\new\\datasets\\nv.py:192\u001B[0m, in \u001B[0;36mNV.__init__\u001B[1;34m(self, root_path, annotation_path, subset, n_samples_for_each_video, spatial_transform, temporal_transform, target_transform, sample_duration, modality, get_loader)\u001B[0m\n\u001B[0;32m    181\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21m__init__\u001B[39m(\u001B[38;5;28mself\u001B[39m,\n\u001B[0;32m    182\u001B[0m              root_path,\n\u001B[0;32m    183\u001B[0m              annotation_path,\n\u001B[1;32m   (...)\u001B[0m\n\u001B[0;32m    190\u001B[0m              modality\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mRGB\u001B[39m\u001B[38;5;124m'\u001B[39m,\n\u001B[0;32m    191\u001B[0m              get_loader\u001B[38;5;241m=\u001B[39mget_default_video_loader):\n\u001B[1;32m--> 192\u001B[0m     \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mdata, \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mclass_names \u001B[38;5;241m=\u001B[39m \u001B[43mmake_dataset\u001B[49m\u001B[43m(\u001B[49m\n\u001B[0;32m    193\u001B[0m \u001B[43m        \u001B[49m\u001B[43mroot_path\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mannotation_path\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43msubset\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mn_samples_for_each_video\u001B[49m\u001B[43m,\u001B[49m\n\u001B[0;32m    194\u001B[0m \u001B[43m        \u001B[49m\u001B[43msample_duration\u001B[49m\u001B[43m)\u001B[49m\n\u001B[0;32m    196\u001B[0m     \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mspatial_transform \u001B[38;5;241m=\u001B[39m spatial_transform\n\u001B[0;32m    197\u001B[0m     \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39mtemporal_transform \u001B[38;5;241m=\u001B[39m temporal_transform\n",
      "File \u001B[1;32mD:\\current\\gesture\\new\\datasets\\nv.py:116\u001B[0m, in \u001B[0;36mmake_dataset\u001B[1;34m(root_path, annotation_path, subset, n_samples_for_each_video, sample_duration)\u001B[0m\n\u001B[0;32m    115\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21mmake_dataset\u001B[39m(root_path, annotation_path, subset, n_samples_for_each_video, sample_duration):\n\u001B[1;32m--> 116\u001B[0m     data \u001B[38;5;241m=\u001B[39m \u001B[43mload_annotation_data\u001B[49m\u001B[43m(\u001B[49m\u001B[43mannotation_path\u001B[49m\u001B[43m)\u001B[49m\n\u001B[0;32m    117\u001B[0m     video_names, annotations \u001B[38;5;241m=\u001B[39m get_video_names_and_annotations(data, subset)\n\u001B[0;32m    118\u001B[0m     class_to_idx \u001B[38;5;241m=\u001B[39m get_class_labels(data)\n",
      "File \u001B[1;32mD:\\current\\gesture\\new\\datasets\\nv.py:88\u001B[0m, in \u001B[0;36mload_annotation_data\u001B[1;34m(data_file_path)\u001B[0m\n\u001B[0;32m     87\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21mload_annotation_data\u001B[39m(data_file_path):\n\u001B[1;32m---> 88\u001B[0m     \u001B[38;5;28;01mwith\u001B[39;00m \u001B[38;5;28;43mopen\u001B[39;49m\u001B[43m(\u001B[49m\u001B[43mdata_file_path\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43mr\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m)\u001B[49m \u001B[38;5;28;01mas\u001B[39;00m data_file:\n\u001B[0;32m     89\u001B[0m         \u001B[38;5;28;01mreturn\u001B[39;00m json\u001B[38;5;241m.\u001B[39mload(data_file)\n",
      "\u001B[1;31mFileNotFoundError\u001B[0m: [Errno 2] No such file or directory: './annotation_nvGesture_v1/nvall_but_None.json'"
     ]
    }
   ],
   "source": [
    "training_data = NV(\n",
    "    './nvGesture_v1.1/nvGesture_v1',\n",
    "    './annotation_nvGesture_v1/nvall_but_None.json',\n",
    "    'training',\n",
    "    spatial_transform=spatial_transform,\n",
    "    temporal_transform=temporal_transform,\n",
    "    target_transform=target_transform,\n",
    "    sample_duration=sample_duration,\n",
    "    modality=\"RGB-D\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5ed0e8a9-5fae-4eda-acd9-f1f27d442826",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "end_time": "2024-03-02T07:46:53.578865700Z",
     "start_time": "2024-03-02T07:46:53.568462300Z"
    }
   },
   "outputs": [],
   "source": [
    "train_loader = torch.utils.data.DataLoader(\n",
    "    training_data,\n",
    "    batch_size=80,\n",
    "    shuffle=True,\n",
    "    num_workers=12,\n",
    "    pin_memory=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d8e9ff8c-d19b-4b0a-aac4-ff49feb4440c",
   "metadata": {
    "tags": [],
    "ExecuteTime": {
     "start_time": "2024-03-02T07:46:53.572952800Z"
    }
   },
   "outputs": [],
   "source": [
    "# logger.info(f\"run\")\n",
    "# best_prec1 = 0\n",
    "# for i in range(1, n_epochs + 1):\n",
    "#     # for i in range(opt.begin_epoch, opt.begin_epoch + 10):\n",
    "#     torch.cuda.empty_cache()\n",
    "#     adjust_learning_rate(optimizer, i)\n",
    "#     train_epoch(i, train_loader, model, criterion, optimizer)\n",
    "#     state = {\n",
    "#         'epoch': i,\n",
    "#         'arch': arch,\n",
    "#         'state_dict': model.state_dict(),\n",
    "#         'optimizer': optimizer.state_dict(),\n",
    "#         'best_prec1': best_prec1\n",
    "#     }\n",
    "#     save_checkpoint(state, False) \n",
    "#     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0364f529-f663-417b-ad0e-db46d443d147",
   "metadata": {
    "ExecuteTime": {
     "start_time": "2024-03-02T07:46:53.577765700Z"
    }
   },
   "outputs": [],
   "source": [
    "if __name__ == '__main__':\n",
    "    logger.info(f\"run\")\n",
    "    best_prec1 = 0\n",
    "    for i in range(1, n_epochs + 1):\n",
    "        # for i in range(opt.begin_epoch, opt.begin_epoch + 10):\n",
    "        torch.cuda.empty_cache()\n",
    "        adjust_learning_rate(optimizer, i)\n",
    "        train_epoch(i, train_loader, model, criterion, optimizer)\n",
    "        state = {\n",
    "            'epoch': i,\n",
    "            'arch': arch,\n",
    "            'state_dict': model.state_dict(),\n",
    "            'optimizer': optimizer.state_dict(),\n",
    "            'best_prec1': best_prec1\n",
    "        }\n",
    "        save_checkpoint(state, False) \n",
    "        "
   ]
  }
 ],
 "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}