plot_sandbox.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import torch\n",
    "from pathlib import Path\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from models import *\n",
    "from data.data_classes import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Configs\n",
    "index = 42\n",
    "\n",
    "batch_size = 50\n",
    "num_ctx_frames = 5\n",
    "num_tgt_frames = 5\n",
    "\n",
    "input_channels=3\n",
    "num_hidden=[64, 64, 64]\n",
    "kernel_size=5\n",
    "stride=1\n",
    "learning_rate=1e-3\n",
    "\n",
    "ckpt_path = Path(\"./logs/PredRNN_RGB/version_0/checkpoints/epoch=10-step=693.ckpt\")\n",
    "model = PredRNN(input_channels=input_channels,\n",
    "                num_hidden=num_hidden,\n",
    "                num_ctx_frames=num_ctx_frames,\n",
    "                num_tgt_frames=num_tgt_frames,\n",
    "                kernel_size=kernel_size,\n",
    "                stride=stride,\n",
    "                learning_rate=1e-3)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "PredRNN.__init__() missing 6 required positional arguments: 'input_channels', 'num_hidden', 'num_ctx_frames', 'num_tgt_frames', 'kernel_size', and 'stride'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m/home/kelvinfung/Documents/bounce-digits/plot_sandbox.ipynb Cell 4\u001b[0m in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      <a href='vscode-notebook-cell://ssh-remote%2Bcompute.3294756187141273406/home/kelvinfung/Documents/bounce-digits/plot_sandbox.ipynb#W3sdnNjb2RlLXJlbW90ZQ%3D%3D?line=0'>1</a>\u001b[0m \u001b[39m# Prepare trained model\u001b[39;00m\n\u001b[0;32m----> <a href='vscode-notebook-cell://ssh-remote%2Bcompute.3294756187141273406/home/kelvinfung/Documents/bounce-digits/plot_sandbox.ipynb#W3sdnNjb2RlLXJlbW90ZQ%3D%3D?line=1'>2</a>\u001b[0m model \u001b[39m=\u001b[39m model\u001b[39m.\u001b[39;49mload_from_checkpoint(ckpt_path)\n",
      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.10/site-packages/pytorch_lightning/core/saving.py:161\u001b[0m, in \u001b[0;36mModelIO.load_from_checkpoint\u001b[0;34m(cls, checkpoint_path, map_location, hparams_file, strict, **kwargs)\u001b[0m\n\u001b[1;32m    158\u001b[0m \u001b[39m# override the hparams with values that were passed in\u001b[39;00m\n\u001b[1;32m    159\u001b[0m checkpoint[\u001b[39mcls\u001b[39m\u001b[39m.\u001b[39mCHECKPOINT_HYPER_PARAMS_KEY]\u001b[39m.\u001b[39mupdate(kwargs)\n\u001b[0;32m--> 161\u001b[0m model \u001b[39m=\u001b[39m \u001b[39mcls\u001b[39;49m\u001b[39m.\u001b[39;49m_load_model_state(checkpoint, strict\u001b[39m=\u001b[39;49mstrict, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m    162\u001b[0m \u001b[39mreturn\u001b[39;00m model\n",
      "File \u001b[0;32m~/miniconda3/envs/pytorch/lib/python3.10/site-packages/pytorch_lightning/core/saving.py:203\u001b[0m, in \u001b[0;36mModelIO._load_model_state\u001b[0;34m(cls, checkpoint, strict, **cls_kwargs_new)\u001b[0m\n\u001b[1;32m    199\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m cls_spec\u001b[39m.\u001b[39mvarkw:\n\u001b[1;32m    200\u001b[0m     \u001b[39m# filter kwargs according to class init unless it allows any argument via kwargs\u001b[39;00m\n\u001b[1;32m    201\u001b[0m     _cls_kwargs \u001b[39m=\u001b[39m {k: v \u001b[39mfor\u001b[39;00m k, v \u001b[39min\u001b[39;00m _cls_kwargs\u001b[39m.\u001b[39mitems() \u001b[39mif\u001b[39;00m k \u001b[39min\u001b[39;00m cls_init_args_name}\n\u001b[0;32m--> 203\u001b[0m model \u001b[39m=\u001b[39m \u001b[39mcls\u001b[39;49m(\u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49m_cls_kwargs)\n\u001b[1;32m    205\u001b[0m \u001b[39m# give model a chance to load something\u001b[39;00m\n\u001b[1;32m    206\u001b[0m model\u001b[39m.\u001b[39mon_load_checkpoint(checkpoint)\n",
      "\u001b[0;31mTypeError\u001b[0m: PredRNN.__init__() missing 6 required positional arguments: 'input_channels', 'num_hidden', 'num_ctx_frames', 'num_tgt_frames', 'kernel_size', and 'stride'"
     ]
    }
   ],
   "source": [
    "# Prepare trained model\n",
    "model = model.load_from_checkpoint(ckpt_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([50, 3, 5, 128, 128])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rgb_movingmnist = TwoColourMovingMNISTDataModule(batch_size, num_ctx_frames, num_tgt_frames)\n",
    "rgb_movingmnist.setup()\n",
    "\n",
    "train_dataloader = rgb_movingmnist.train_dataloader()\n",
    "for train_x_batch, train_y_batch in train_dataloader:\n",
    "    break\n",
    "\n",
    "train_x_batch.shape  # bs x C x F x H x W"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.eval()\n",
    "\n",
    "y_pred = model(train_x_batch[index:index+1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([239208.,    893.,    669.,    557.,    503.,    594.,    555.,\n",
       "           599.,    872.,   1310.]),\n",
       " array([0.        , 0.09960784, 0.19921568, 0.29882354, 0.39843136,\n",
       "        0.49803922, 0.5976471 , 0.6972549 , 0.7968627 , 0.8964706 ,\n",
       "        0.99607843], dtype=float32),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(train_x_batch[index:index+1].flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([239003.,    924.,    692.,    583.,    514.,    615.,    576.,\n",
       "           623.,    899.,   1331.]),\n",
       " array([0.        , 0.09960784, 0.19921568, 0.29882354, 0.39843136,\n",
       "        0.49803922, 0.5976471 , 0.6972549 , 0.7968627 , 0.8964706 ,\n",
       "        0.99607843], dtype=float32),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(train_y_batch[index:index+1].detach().numpy().flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([2.20000e+01, 1.80000e+02, 5.64000e+02, 1.33200e+03, 1.08360e+04,\n",
       "        8.27470e+04, 1.41904e+05, 7.16500e+03, 9.76000e+02, 3.40000e+01]),\n",
       " array([-3.5708227 , -2.9811969 , -2.3915708 , -1.8019449 , -1.2123189 ,\n",
       "        -0.62269294, -0.03306699,  0.55655897,  1.1461849 ,  1.7358109 ,\n",
       "         2.3254368 ], dtype=float32),\n",
       " <BarContainer object of 10 artists>)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_pred.detach().numpy().flatten())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 3, 5, 128, 128])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.5 ('pytorch')",
   "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": "3.10.5"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "da4c3a4107fc661dfc1ddc51b98664f856b9baf685ab1745d9fa2472938977d8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}