examples/notebook/fedscale_demo_client.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "466d23de",
   "metadata": {},
   "source": [
    "# Federated Learning for Image Classification using Fedscale"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17da3fe2",
   "metadata": {},
   "source": [
    "## Client Side"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "40dc07a4",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "(02-27) 12:24:33 INFO     [fllibs.py:97] Initializing the model ...\n",
      "(02-27) 12:24:33 INFO     [executor.py:75] (EXECUTOR:1) is setting up environ ...\n",
      "0.0%"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to /Users/fan/cifar10/cifar-10-python.tar.gz\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100.0%\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting /Users/fan/cifar10/cifar-10-python.tar.gz to /Users/fan/cifar10/\n",
      "Files already downloaded and verified\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "(02-27) 12:24:53 INFO     [executor.py:123] Data partitioner starts ...\n",
      "(02-27) 12:24:53 INFO     [divide_data.py:105] Randomly partitioning data, 50000 samples...\n",
      "(02-27) 12:24:53 INFO     [divide_data.py:105] Randomly partitioning data, 10000 samples...\n",
      "(02-27) 12:24:53 INFO     [executor.py:134] Data partitioner completes ...\n",
      "(02-27) 12:24:53 INFO     [channel_context.py:20] %%%%%%%%%% Opening grpc connection to 127.0.0.1 %%%%%%%%%%\n",
      "(02-27) 12:24:53 INFO     [executor.py:372] Start monitoring events ...\n",
      "(02-27) 12:26:30 INFO     [model_test_module.py:307] Rank 1: Test set: Average loss: 2.3027, Top-1 Accuracy: 1000.0/10000 (0.1), Top-5 Accuracy: 0.5\n",
      "(02-27) 12:26:30 INFO     [torch_client.py:264] Test results: Eval_time 94.6244, test_loss 2.3027, test_accuracy 10.00%, test_5_accuracy 50.00% \n",
      "\n",
      "(02-27) 12:26:31 INFO     [torch_client.py:49] Start to train (CLIENT: 1) ...\n",
      "(02-27) 12:27:33 INFO     [torch_client.py:84] Training of (CLIENT: 1) completes, {'client_id': 1, 'moving_loss': 6.559604595482319, 'trained_size': 600, 'success': True}\n",
      "(02-27) 12:27:33 INFO     [torch_client.py:49] Start to train (CLIENT: 2) ...\n",
      "(02-27) 12:28:36 INFO     [torch_client.py:84] Training of (CLIENT: 2) completes, {'client_id': 2, 'moving_loss': 5.693940820821615, 'trained_size': 600, 'success': True}\n",
      "(02-27) 12:28:36 INFO     [torch_client.py:49] Start to train (CLIENT: 3) ...\n",
      "(02-27) 12:29:37 INFO     [torch_client.py:84] Training of (CLIENT: 3) completes, {'client_id': 3, 'moving_loss': 5.515216224300269, 'trained_size': 600, 'success': True}\n",
      "(02-27) 12:29:38 INFO     [torch_client.py:49] Start to train (CLIENT: 4) ...\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[1], line 15\u001b[0m\n\u001b[1;32m     13\u001b[0m parser\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39muse_cuda \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFalse\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     14\u001b[0m Demo_Executor \u001b[38;5;241m=\u001b[39m Executor(parser\u001b[38;5;241m.\u001b[39margs)\n\u001b[0;32m---> 15\u001b[0m \u001b[43mDemo_Executor\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/executor.py:144\u001b[0m, in \u001b[0;36mExecutor.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    142\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining_sets, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtesting_sets \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minit_data()\n\u001b[1;32m    143\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msetup_communication()\n\u001b[0;32m--> 144\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mevent_monitor\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/executor.py:385\u001b[0m, in \u001b[0;36mExecutor.event_monitor\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    383\u001b[0m train_config[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m=\u001b[39m train_model\n\u001b[1;32m    384\u001b[0m train_config[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mclient_id\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mint\u001b[39m(train_config[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mclient_id\u001b[39m\u001b[38;5;124m'\u001b[39m])\n\u001b[0;32m--> 385\u001b[0m client_id, train_res \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mTrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtrain_config\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    387\u001b[0m \u001b[38;5;66;03m# Upload model updates\u001b[39;00m\n\u001b[1;32m    388\u001b[0m future_call \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maggregator_communicator\u001b[38;5;241m.\u001b[39mstub\u001b[38;5;241m.\u001b[39mCLIENT_EXECUTE_COMPLETION\u001b[38;5;241m.\u001b[39mfuture(\n\u001b[1;32m    389\u001b[0m     job_api_pb2\u001b[38;5;241m.\u001b[39mCompleteRequest(client_id\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mstr\u001b[39m(client_id), executor_id\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexecutor_id,\n\u001b[1;32m    390\u001b[0m                                 event\u001b[38;5;241m=\u001b[39mcommons\u001b[38;5;241m.\u001b[39mUPLOAD_MODEL, status\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m, msg\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[1;32m    391\u001b[0m                                 meta_result\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, data_result\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mserialize_response(train_res)\n\u001b[1;32m    392\u001b[0m                                 ))\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/executor.py:204\u001b[0m, in \u001b[0;36mExecutor.Train\u001b[0;34m(self, config)\u001b[0m\n\u001b[1;32m    202\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m object must be a non-null value in the training config.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    203\u001b[0m client_conf \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39moverride_conf(train_config)\n\u001b[0;32m--> 204\u001b[0m train_res \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtraining_handler\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    205\u001b[0m \u001b[43m    \u001b[49m\u001b[43mclient_id\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mclient_id\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconf\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mclient_conf\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mconfig\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mmodel\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    207\u001b[0m \u001b[38;5;66;03m# Report execution completion meta information\u001b[39;00m\n\u001b[1;32m    208\u001b[0m response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maggregator_communicator\u001b[38;5;241m.\u001b[39mstub\u001b[38;5;241m.\u001b[39mCLIENT_EXECUTE_COMPLETION(\n\u001b[1;32m    209\u001b[0m     job_api_pb2\u001b[38;5;241m.\u001b[39mCompleteRequest(\n\u001b[1;32m    210\u001b[0m         client_id\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mstr\u001b[39m(client_id), executor_id\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexecutor_id,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    213\u001b[0m     )\n\u001b[1;32m    214\u001b[0m )\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/executor.py:309\u001b[0m, in \u001b[0;36mExecutor.training_handler\u001b[0;34m(self, client_id, conf, model)\u001b[0m\n\u001b[1;32m    303\u001b[0m client_data \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining_sets \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mtask \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrl\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m \\\n\u001b[1;32m    304\u001b[0m     select_dataset(client_id, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining_sets,\n\u001b[1;32m    305\u001b[0m                    batch_size\u001b[38;5;241m=\u001b[39mconf\u001b[38;5;241m.\u001b[39mbatch_size, args\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs,\n\u001b[1;32m    306\u001b[0m                    collate_fn\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcollate_fn\n\u001b[1;32m    307\u001b[0m                    )\n\u001b[1;32m    308\u001b[0m client \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mget_client_trainer(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs)\n\u001b[0;32m--> 309\u001b[0m train_res \u001b[38;5;241m=\u001b[39m \u001b[43mclient\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    310\u001b[0m \u001b[43m    \u001b[49m\u001b[43mclient_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mclient_data\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodel_adapter\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_model\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconf\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mconf\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    312\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m train_res\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/torch_client.py:71\u001b[0m, in \u001b[0;36mTorchClient.train\u001b[0;34m(self, client_data, model, conf)\u001b[0m\n\u001b[1;32m     69\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcompleted_steps \u001b[38;5;241m<\u001b[39m conf\u001b[38;5;241m.\u001b[39mlocal_steps:\n\u001b[1;32m     70\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 71\u001b[0m         \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain_step\u001b[49m\u001b[43m(\u001b[49m\u001b[43mclient_data\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mconf\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moptimizer\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcriterion\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     72\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m ex:\n\u001b[1;32m     73\u001b[0m         error_type \u001b[38;5;241m=\u001b[39m ex\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/fedscale/cloud/execution/torch_client.py:235\u001b[0m, in \u001b[0;36mTorchClient.train_step\u001b[0;34m(self, client_data, conf, model, optimizer, criterion)\u001b[0m\n\u001b[1;32m    233\u001b[0m \u001b[38;5;66;03m# ========= Define the backward loss ==============\u001b[39;00m\n\u001b[1;32m    234\u001b[0m optimizer\u001b[38;5;241m.\u001b[39mzero_grad()\n\u001b[0;32m--> 235\u001b[0m \u001b[43mloss\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    236\u001b[0m optimizer\u001b[38;5;241m.\u001b[39mstep()\n\u001b[1;32m    238\u001b[0m \u001b[38;5;66;03m# ========= Weight handler ========================\u001b[39;00m\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/torch/_tensor.py:488\u001b[0m, in \u001b[0;36mTensor.backward\u001b[0;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[1;32m    478\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m has_torch_function_unary(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m    479\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m handle_torch_function(\n\u001b[1;32m    480\u001b[0m         Tensor\u001b[38;5;241m.\u001b[39mbackward,\n\u001b[1;32m    481\u001b[0m         (\u001b[38;5;28mself\u001b[39m,),\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    486\u001b[0m         inputs\u001b[38;5;241m=\u001b[39minputs,\n\u001b[1;32m    487\u001b[0m     )\n\u001b[0;32m--> 488\u001b[0m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mautograd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    489\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgradient\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minputs\u001b[49m\n\u001b[1;32m    490\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/anaconda3/envs/fedscale/lib/python3.10/site-packages/torch/autograd/__init__.py:197\u001b[0m, in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[1;32m    192\u001b[0m     retain_graph \u001b[38;5;241m=\u001b[39m create_graph\n\u001b[1;32m    194\u001b[0m \u001b[38;5;66;03m# The reason we repeat same the comment below is that\u001b[39;00m\n\u001b[1;32m    195\u001b[0m \u001b[38;5;66;03m# some Python versions print out the first line of a multi-line function\u001b[39;00m\n\u001b[1;32m    196\u001b[0m \u001b[38;5;66;03m# calls in the traceback and some print out the last line\u001b[39;00m\n\u001b[0;32m--> 197\u001b[0m \u001b[43mVariable\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_execution_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_backward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Calls into the C++ engine to run the backward pass\u001b[39;49;00m\n\u001b[1;32m    198\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtensors\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgrad_tensors_\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    199\u001b[0m \u001b[43m    \u001b[49m\u001b[43mallow_unreachable\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maccumulate_grad\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import logging\n",
    "import math\n",
    "from torch.autograd import Variable\n",
    "import numpy as np\n",
    "\n",
    "import sys, os\n",
    "\n",
    "import fedscale.cloud.config_parser as parser\n",
    "from fedscale.cloud.execution.torch_client import TorchClient\n",
    "from fedscale.cloud.execution.executor import Executor\n",
    "### On CPU\n",
    "parser.args.use_cuda = \"False\"\n",
    "Demo_Executor = Executor(parser.args)\n",
    "Demo_Executor.run()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "227bece8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.10.9"
  },
  "vscode": {
   "interpreter": {
    "hash": "c61f092c149070809c0e629644d57c0873f394886eddd38a7b619aa7d2ef8d75"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}