{
  "block": 82830466,
  "op": [
    "transfer",
    {
      "amount": "0.001 HIVE",
      "from": "terablock-hive",
      "memo": "Join 220+ Hive members in supporting our Hive CrossDEX proposal to expand Hive's DeFi ecosystem. Every vote brings us closer to a decentralised financial future. 🚀 Your support matters! Vote now: https://peakd.com/me/proposals/295",
      "to": "cryptopossible"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2024-02-15T13:36:00",
  "trx_id": "0ed98c58fe936e37c6eb26531eb8d858517dc92f",
  "trx_in_block": 27,
  "virtual_op": false
}

{
  "block": 79685179,
  "op": [
    "transfer",
    {
      "amount": "0.001 HIVE",
      "from": "hivesurvey",
      "memo": "By spending just 5-10 minutes of your time to answer an ONLINE SURVEY, you'll receive $1 worth of HIVE or STEEM (4.762 STEEM or 3.125 HIVE) as a token of our appreciation!!! Hello there! I'm Sichen DONG, a research postgraduate student at the University of Hong Kong. I'm currently organizing a paid survey as part of my research study. We kindly invite Steem/Hive members to participate in a survey that focuses on the social changes you've observed since the takeover of Steemit, Inc. by Tron on February 14, 2020. We're delving into the intriguing realm of decentralized autonomous organizations (DAOs) and exploring the impact of social norms on cooperation within these communities. Please note that the survey is conducted in English. Rest assured, your participation involves no more risk than your everyday activities. You retain the freedom to withdraw from the study at any point. Your support is invaluable to our research, and we're eagerly looking forward to your participation! Ready to dive in? Access the survey via this link: https://hivesurvey.vercel.app/",
      "to": "cryptopossible"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2023-10-29T05:24:42",
  "trx_id": "bf25da92628a2605c6031bb55227e98a7be9cd4e",
  "trx_in_block": 14,
  "virtual_op": false
}

{
  "block": 70614711,
  "op": [
    "expired_account_notification",
    {
      "account": "cryptopossible"
    }
  ],
  "op_in_trx": 3,
  "timestamp": "2022-12-17T21:24:15",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 46062933,
  "op": [
    "custom_json",
    {
      "id": "notify",
      "json": "[\"setLastRead\",{\"date\":\"2020-08-15T09:51:25\"}]",
      "required_auths": [],
      "required_posting_auths": [
        "cryptopossible"
      ]
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-08-15T09:53:33",
  "trx_id": "4c8b928f3f56f7076b4253d57ceb56d2f2c39916",
  "trx_in_block": 32,
  "virtual_op": false
}

{
  "block": 46062896,
  "op": [
    "transfer",
    {
      "amount": "996.592 HIVE",
      "from": "cryptopossible",
      "memo": "100925935",
      "to": "deepcrypto8"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-08-15T09:51:42",
  "trx_id": "9fcc2426bc8c5a1728ef09f4dbabfadc03556c33",
  "trx_in_block": 22,
  "virtual_op": false
}

{
  "block": 45486283,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "76.073 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402915 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-07-26T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 45285111,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "76.025 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-07-19T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 45084010,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.976 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-07-12T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44917890,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-07-06T13:28:15",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44882952,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.927 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-07-05T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44717233,
  "op": [
    "comment",
    {
      "author": "cryptopossible",
      "body": "@@ -1,69 +1,4 @@\n-Deep Learning Course Project: Japanese hostel price prediction. %0A\n %0A%0AHo\n",
      "json_metadata": "{\"app\":\"peakd/2020.06.2\",\"format\":\"markdown\",\"description\":\"Coding an algorithm from scratch!\",\"tags\":[\"ai\",\"ml\",\"deeplearning\",\"machinelearning\",\"zerotogans\",\"pytorch\",\"india\",\"learn\"],\"links\":[\"https://jovian.ml/oneworldcoder/05-02-japan-hostel\"],\"image\":[\"https://files.peakd.com/file/peakd-hive/cryptopossible/iAyEw5m0-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/vCHmayap-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/aAr1oesQ-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/7H4Epp6M-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/qa4CigZP-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/N0QTpHGR-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/esrHhtHT-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/UbqYMuwl-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/SZFcOdxN-image.png\"]}",
      "parent_author": "",
      "parent_permlink": "hive-175254",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "title": "Deep Learning Course Project: Japanese hostel price prediction. "
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:51:48",
  "trx_id": "df6650ebe0c8e31f72cb469210126d558c41530b",
  "trx_in_block": 1,
  "virtual_op": false
}

{
  "block": 44716868,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.000 HBD",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "rshares": 1173717929,
      "total_vote_weight": 1169,
      "voter": "putu300",
      "weight": 586
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-29T13:33:30",
  "trx_id": "73dac92a304e802bdbc9f69521e4200e953485c3",
  "trx_in_block": 22,
  "virtual_op": true
}

{
  "block": 44716868,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "voter": "putu300",
      "weight": 430
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:33:30",
  "trx_id": "73dac92a304e802bdbc9f69521e4200e953485c3",
  "trx_in_block": 22,
  "virtual_op": false
}

{
  "block": 44716847,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.000 HBD",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "rshares": 1130572104,
      "total_vote_weight": 583,
      "voter": "besheda",
      "weight": 468
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-29T13:32:27",
  "trx_id": "de6b1be32a53888fd45a2951d7572f6d2932998d",
  "trx_in_block": 2,
  "virtual_op": true
}

{
  "block": 44716847,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "voter": "besheda",
      "weight": 4300
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:32:27",
  "trx_id": "de6b1be32a53888fd45a2951d7572f6d2932998d",
  "trx_in_block": 2,
  "virtual_op": false
}

{
  "block": 44716786,
  "op": [
    "claim_reward_balance",
    {
      "account": "cryptopossible",
      "reward_hbd": "0.006 HBD",
      "reward_hive": "0.000 HIVE",
      "reward_vests": "50.551051 VESTS"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:29:24",
  "trx_id": "3abc5709efd796c1c6b39903327fa0f42affa4d7",
  "trx_in_block": 3,
  "virtual_op": false
}

{
  "block": 44716773,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.000 HBD",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "rshares": 37632149,
      "total_vote_weight": 18,
      "voter": "laissez-faire",
      "weight": 1
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-29T13:28:45",
  "trx_id": "1cc732ffaa2e2bcf6338b22beee7e0e5723b1298",
  "trx_in_block": 23,
  "virtual_op": true
}

{
  "block": 44716773,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "voter": "laissez-faire",
      "weight": 10000
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:28:45",
  "trx_id": "1cc732ffaa2e2bcf6338b22beee7e0e5723b1298",
  "trx_in_block": 23,
  "virtual_op": false
}

{
  "block": 44716768,
  "op": [
    "custom_json",
    {
      "id": "follow",
      "json": "[\"reblog\",{\"account\":\"cryptopossible\",\"author\":\"cryptopossible\",\"permlink\":\"deep-learning-course-project-japanese-hostel-price-prediction\"}]",
      "required_auths": [],
      "required_posting_auths": [
        "cryptopossible"
      ]
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:28:30",
  "trx_id": "9c784a3c54702ee4edff175e38061e3d69d6bf6b",
  "trx_in_block": 1,
  "virtual_op": false
}

{
  "block": 44716764,
  "op": [
    "comment",
    {
      "author": "cryptopossible",
      "body": "Deep Learning Course Project: Japanese hostel price prediction. \n\n\nHostels are very important for backpackers and teens because they provide an economical housing option and promote travel. When we travel, we meet new people, engage in meaningful conversation, and widen our horizon of knowledge. So, I decided to train a model to predict the hostel prices in Japan. \n\n```\ndef customize_dataset(dataframe_raw):\n    dataframe = dataframe_raw.copy(deep=True)\n    # drop some columns\n    dataframe = dataframe.drop(['Unnamed: 0', 'hostel.name', 'Distance', 'lon', 'lat', 'atmosphere', 'cleanliness',\n                                'facilities', 'location.y', 'security', 'staff'], axis=1)\n#     for col in ['City', 'rating.band']:\n#        # normalizing incoming data\n#        dataframe[col] = (dataframe[col] - min(dataframe[col])) / (max(dataframe[col]) - min(dataframe[col]))\n    \n    # dropping any row that contains at least on missing value\n    dataframe = dataframe.dropna(axis=0)  \n    \n    return dataframe\n\ndataframe = customize_dataset(dataframe)\n\ndataframe.head()\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/iAyEw5m0-image.png)\nMy main objective was to concentrate on four important factors which determine hostel prices. These factors were location, value for money, customer rating and the rating range. Once these factors were correlated, they could be used for predictions. Hence, finding the factors which could correlate with each other was very important.\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/vCHmayap-image.png)\n\n```\n# Convert from Pandas dataframe to numpy arrays\n\n\ndef dataframe_to_arrays(dataframe):\n    # Make a copy of the original dataframe\n    dataframe1 = dataframe.copy(deep=True)\n    # Convert non-numeric categorical columns to numbers\n    for col in categorical_cols:\n        dataframe1[col] = dataframe1[col].astype('category').cat.codes\n    \n    # Extract input & outupts as numpy arrays\n    inputs_array = dataframe1[input_cols].to_numpy()\n    targets_array = dataframe1[output_cols].to_numpy()\n    return inputs_array, targets_array\n\ninputs_array, targets_array = dataframe_to_arrays(dataframe)\ninputs_array, targets_array\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/aAr1oesQ-image.png)\n\n```\ninputs = torch.from_numpy(inputs_array).type(torch.float32)\ntargets = torch.from_numpy(targets_array).type(torch.float32)\n\ndataset = TensorDataset(inputs, targets)\n\nval_percent = 0.1 # between 0.1 and 0.2\nval_size = int(num_rows * val_percent)\ntrain_size = num_rows - val_size\n\n# Use the random_split function to split dataset into 2 parts of the desired length\ntrain_ds, val_ds = random_split(dataset, [train_size, val_size]) \n\ntrain_loader = DataLoader(train_ds, batch_size, shuffle=True)\nval_loader = DataLoader(val_ds, batch_size)\n\ninput_size = len(input_cols)\noutput_size = len(output_cols)\nprint(len(input_cols))\nprint(len(output_cols))\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/7H4Epp6M-image.png)\n\n\n\n\n```\nclass HousingModel(nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.linear1 = nn.Linear(input_size, 8)\n        self.linear2 = nn.Linear(8, 16)\n        self.linear3 = nn.Linear(16, output_size)\n        \n    def forward(self, xb):\n        out = self.linear1(xb)\n        out = F.relu(out)\n        out = self.linear2(out)\n        out = F.relu(out)\n        out = self.linear3(out)\n        return out\n    \n    def training_step(self, batch):\n        inputs, targets = batch \n        out = self(inputs)                 # Generate predictions\n        loss = F.l1_loss(out, targets)    # Calculate loss\n        return loss\n    \n    def validation_step(self, batch):\n        inputs, targets = batch \n        out = self(inputs)                 # Generate predictions\n        loss = F.l1_loss(out, targets)    # Calculate loss\n        return {'val_loss': loss.detach()}\n        \n    def validation_epoch_end(self, outputs):\n        batch_losses = [x['val_loss'] for x in outputs]\n        epoch_loss = torch.stack(batch_losses).mean()   # Combine losses\n        return {'val_loss': epoch_loss.item()}\n    \n    def epoch_end(self, epoch, result):\n        print(\"Epoch [{}], val_loss: {:.4f}\".format(epoch, result['val_loss']))\n    \nmodel = HousingModel()\ndef evaluate(model, val_loader):\n    outputs = [model.validation_step(batch) for batch in val_loader]\n    return model.validation_epoch_end(outputs)\n\ndef fit(epochs, max_lr, model, train_loader, val_loader, weight_decay=0, grad_clip=None, opt_func=torch.optim.SGD):\n    history = []\n    optimizer = opt_func(model.parameters(), max_lr, weight_decay=weight_decay)\n    sched = torch.optim.lr_scheduler.OneCycleLR(optimizer, max_lr, epochs=epochs, \n                                                steps_per_epoch=len(train_loader))\n    for epoch in range(epochs):\n        # Training Phase \n        for batch in train_loader:\n            loss = model.training_step(batch)\n            loss.backward()\n            if grad_clip: \n                nn.utils.clip_grad_value_(model.parameters(), grad_clip)\n            optimizer.step()\n            optimizer.zero_grad()\n        # Validation phase\n        result = evaluate(model, val_loader)\n        model.epoch_end(epoch, result)\n        history.append(result)\n    return history\n\nepochs = 40\nmax_lr = 1.5\ngrad_clip = 9\nweight_decay = 1e-8\nopt_func = torch.optim.Adam\n\n\nhistory= fit(epochs, max_lr, model, train_loader, val_loader, weight_decay=weight_decay, grad_clip=grad_clip, opt_func=opt_func)\n\nhistory\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/qa4CigZP-image.png)\n\n\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/N0QTpHGR-image.png)\n\n\nThis principle of correlation is commonly used by travel websites/apps to help customers find the right hostel by running price predicting algorithms.  \n```\nlosses = [r['val_loss'] for r in [result] + history]\nplt.plot(losses, '-x')\nplt.xlabel('epoch')\nplt.ylabel('val_loss')\nplt.title('val_loss vs. epochs');\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/esrHhtHT-image.png)\n\n\n```\ndef predict_single(x, model):\n    xb = x.unsqueeze(0)\n    return model(x).item()\n```\n```\nx, target = val_ds[4]\npred = predict_single(x, model)\nprint(\"Input: \", x)\nprint(\"Target: \", target.item())\nprint(\"Prediction:\", pred)\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/UbqYMuwl-image.png)\n\n\n```\nx, target = val_ds[1]\npred = predict_single(x, model)\nprint(\"Input: \", x)\nprint(\"Target: \", target.item())\nprint(\"Prediction:\", pred)\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/SZFcOdxN-image.png)\n\n\n\nAccording to my algorithm, the predictions were close to the actual price. Hence, to code a good algorithm our data should have an equal number of hostels from different price segments. \n \n\nTo view my Jupyter notebook [click here](https://jovian.ml/oneworldcoder/05-02-japan-hostel). Please feel free to give your feedback.\n",
      "json_metadata": "{\"app\":\"peakd/2020.06.2\",\"format\":\"markdown\",\"description\":\"Coding an algorithm from scratch!\",\"tags\":[\"ai\",\"ml\",\"deeplearning\",\"machinelearning\",\"zerotogans\",\"pytorch\",\"india\",\"learn\"],\"links\":[\"https://jovian.ml/oneworldcoder/05-02-japan-hostel\"],\"image\":[\"https://files.peakd.com/file/peakd-hive/cryptopossible/iAyEw5m0-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/vCHmayap-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/aAr1oesQ-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/7H4Epp6M-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/qa4CigZP-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/N0QTpHGR-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/esrHhtHT-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/UbqYMuwl-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/SZFcOdxN-image.png\"]}",
      "parent_author": "",
      "parent_permlink": "hive-175254",
      "permlink": "deep-learning-course-project-japanese-hostel-price-prediction",
      "title": "Deep Learning Course Project: Japanese hostel price prediction. "
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:28:18",
  "trx_id": "c666628a11adcf0fe62dddaae07bc17152f518d2",
  "trx_in_block": 26,
  "virtual_op": false
}

{
  "block": 44716255,
  "op": [
    "custom_json",
    {
      "id": "community",
      "json": "[\"subscribe\",{\"community\":\"hive-175254\"}]",
      "required_auths": [],
      "required_posting_auths": [
        "cryptopossible"
      ]
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-29T13:02:48",
  "trx_id": "8a0286c839b48607a8838ff3dac19ca5625ead93",
  "trx_in_block": 20,
  "virtual_op": false
}

{
  "block": 44681841,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.879 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 7,
  "timestamp": "2020-06-28T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44480686,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.830 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 3,
  "timestamp": "2020-06-21T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44429517,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10"
    }
  ],
  "op_in_trx": 6,
  "timestamp": "2020-06-19T13:33:45",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44429517,
  "op": [
    "comment_reward",
    {
      "author": "cryptopossible",
      "author_rewards": 52,
      "beneficiary_payout_value": "0.000 HBD",
      "curator_payout_value": "0.011 HBD",
      "payout": "0.024 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "total_payout_value": "0.012 HBD"
    }
  ],
  "op_in_trx": 5,
  "timestamp": "2020-06-19T13:33:45",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44429517,
  "op": [
    "author_reward",
    {
      "author": "cryptopossible",
      "curators_vesting_payout": "91.380746 VESTS",
      "hbd_payout": "0.006 HBD",
      "hive_payout": "0.000 HIVE",
      "payout_must_be_claimed": true,
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "vesting_payout": "50.551051 VESTS"
    }
  ],
  "op_in_trx": 4,
  "timestamp": "2020-06-19T13:33:45",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44393254,
  "op": [
    "claim_reward_balance",
    {
      "account": "cryptopossible",
      "reward_hbd": "0.072 HBD",
      "reward_hive": "0.000 HIVE",
      "reward_vests": "560.101856 VESTS"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-18T07:16:18",
  "trx_id": "ef44009260f6439f46fdce9330b89e2cc9f65dcd",
  "trx_in_block": 21,
  "virtual_op": false
}

{
  "block": 44343128,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california"
    }
  ],
  "op_in_trx": 9,
  "timestamp": "2020-06-16T13:24:33",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44343128,
  "op": [
    "comment_reward",
    {
      "author": "cryptopossible",
      "author_rewards": 576,
      "beneficiary_payout_value": "0.000 HBD",
      "curator_payout_value": "0.143 HBD",
      "payout": "0.288 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "total_payout_value": "0.144 HBD"
    }
  ],
  "op_in_trx": 8,
  "timestamp": "2020-06-16T13:24:33",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44343128,
  "op": [
    "author_reward",
    {
      "author": "cryptopossible",
      "curators_vesting_payout": "1108.534924 VESTS",
      "hbd_payout": "0.072 HBD",
      "hive_payout": "0.000 HIVE",
      "payout_must_be_claimed": true,
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "vesting_payout": "560.101856 VESTS"
    }
  ],
  "op_in_trx": 7,
  "timestamp": "2020-06-16T13:24:33",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44280547,
  "op": [
    "comment",
    {
      "author": "hivebuzz",
      "body": "<center>[![](https://images.hive.blog/175x175/http://hivebuzz.me/@cryptopossible/level.png?202006140849)](https://hivebuzz.me/@cryptopossible)\r\n<center>@cryptopossible, sorry to see you have less Hive Power.\r\nYour level lowered and you are now a **Red Fish**!</center>\n\n###### Support the HiveBuzz project. [Vote](https://hivesigner.com/sign/update_proposal_votes?proposal_ids=%5B%22109%22%5D&approve=true) for [our proposal](https://peakd.com/me/proposals/109)!",
      "json_metadata": "{\"image\":[\"http://hivebuzz.me/notify.t6.png\"]}",
      "parent_author": "cryptopossible",
      "parent_permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "permlink": "hivebuzz-notify-cryptopossible-20200614t090746000z",
      "title": ""
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-14T09:07:48",
  "trx_id": "7479eedecc9cdb7d9512fd781970ea0ca2090b0d",
  "trx_in_block": 15,
  "virtual_op": false
}

{
  "block": 44279547,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.783 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-14T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44228589,
  "op": [
    "claim_reward_balance",
    {
      "account": "cryptopossible",
      "reward_hbd": "0.125 HBD",
      "reward_hive": "0.000 HIVE",
      "reward_vests": "1008.401419 VESTS"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:43:27",
  "trx_id": "ec1dd81ab320d615d1950a7e1bfb463b4093d772",
  "trx_in_block": 6,
  "virtual_op": false
}

{
  "block": 44228496,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.026 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "rshares": 199261603,
      "total_vote_weight": 48524,
      "voter": "curbot",
      "weight": 97
    }
  ],
  "op_in_trx": 3,
  "timestamp": "2020-06-12T13:38:48",
  "trx_id": "975962eddbce0af00f82e0dcaf63bf16f3ba25ac",
  "trx_in_block": 25,
  "virtual_op": true
}

{
  "block": 44228496,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "voter": "curbot",
      "weight": 1000
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-12T13:38:48",
  "trx_id": "975962eddbce0af00f82e0dcaf63bf16f3ba25ac",
  "trx_in_block": 25,
  "virtual_op": false
}

{
  "block": 44228495,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.026 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "rshares": 81613550235,
      "total_vote_weight": 48427,
      "voter": "digital.mine",
      "weight": 39809
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-12T13:38:45",
  "trx_id": "245716b9acd6a1fe4cfb1aa6b9dd9b330faab078",
  "trx_in_block": 22,
  "virtual_op": true
}

{
  "block": 44228495,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "voter": "digital.mine",
      "weight": 40
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:38:45",
  "trx_id": "245716b9acd6a1fe4cfb1aa6b9dd9b330faab078",
  "trx_in_block": 22,
  "virtual_op": false
}

{
  "block": 44228492,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.004 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "rshares": 155605522,
      "total_vote_weight": 8215,
      "voter": "hf19",
      "weight": 73
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-12T13:38:36",
  "trx_id": "023c293f1384c1308e535c19050f5abc3aeebd78",
  "trx_in_block": 74,
  "virtual_op": true
}

{
  "block": 44228492,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "voter": "hf19",
      "weight": 10000
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:38:36",
  "trx_id": "023c293f1384c1308e535c19050f5abc3aeebd78",
  "trx_in_block": 74,
  "virtual_op": false
}

{
  "block": 44228478,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.004 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "rshares": 1276682799,
      "total_vote_weight": 8138,
      "voter": "besheda",
      "weight": 520
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-12T13:37:54",
  "trx_id": "488b8b832fa5e362a3838d66088bf115b49cf2f2",
  "trx_in_block": 48,
  "virtual_op": true
}

{
  "block": 44228478,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "voter": "besheda",
      "weight": 4300
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:37:54",
  "trx_id": "488b8b832fa5e362a3838d66088bf115b49cf2f2",
  "trx_in_block": 48,
  "virtual_op": false
}

{
  "block": 44228468,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.004 HBD",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "rshares": 15050281440,
      "total_vote_weight": 7503,
      "voter": "discernente",
      "weight": 5402
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-12T13:37:24",
  "trx_id": "58a1251fcd206508c54e3efea6dd7fa52966c9cf",
  "trx_in_block": 23,
  "virtual_op": true
}

{
  "block": 44228468,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "voter": "discernente",
      "weight": 200
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:37:24",
  "trx_id": "58a1251fcd206508c54e3efea6dd7fa52966c9cf",
  "trx_in_block": 23,
  "virtual_op": false
}

{
  "block": 44228399,
  "op": [
    "custom_json",
    {
      "id": "follow",
      "json": "[\"reblog\",{\"account\":\"cryptopossible\",\"author\":\"cryptopossible\",\"permlink\":\"deep-learning-neural-networks-on-images-from-cifar-10\"}]",
      "required_auths": [],
      "required_posting_auths": [
        "cryptopossible"
      ]
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:33:57",
  "trx_id": "003219346636cab9b32bb6dda79a55800dcb006d",
  "trx_in_block": 12,
  "virtual_op": false
}

{
  "block": 44228396,
  "op": [
    "comment",
    {
      "author": "cryptopossible",
      "body": "<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/9suSsTVhYuw\" frameborder=\"0\" allow=\"accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe> \n\nNeural Network is a tool which is used in Deep Learning. This tool uses a combination of linear and non-linear functions to make highly accurate predictions. For training our model, this combination of functions is used multiple times to cycle through the data.  It can be used for predictions on regression or classification problems.  \n\n<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/xYJEM2C0G5Q\" frameborder=\"0\" allow=\"accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe> \n\nHow we utilize the available data will depend on whether we are solving a regression or classification problem.\n\nHere our data consists of images and our task is to make a neural network to identify these images.\n\n```\nimport torch\nimport torchvision\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport torch.nn as nn\nimport torch.nn.functional as F\nfrom torchvision.datasets import CIFAR10\nfrom torchvision.transforms import ToTensor\nfrom torchvision.utils import make_grid\nfrom torch.utils.data.dataloader import DataLoader\nfrom torch.utils.data import random_split\n%matplotlib inline\n```\n```\n# Project name used for jovian.commit\nproject_name = '03-cifar10-feedforward'\n```\n```\ndataset = CIFAR10(root='data/', download=True, transform=ToTensor())\ntest_dataset = CIFAR10(root='data/', train=False, transform=ToTensor())\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/ipelgx0J-image.png)\n\n```\ndataset_size = len(dataset)\ndataset_size\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/nK4Gh2LE-image.png)\n\n```\ntest_dataset_size = len(test_dataset)\ntest_dataset_size\n```\n```\nclasses = dataset.classes\nclasses\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/yyzxAqcr-image.png)\n```\nimg, label = dataset[0]\nimg_shape = img.shape\nimg_shape\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/z51CPqg0-image.png)\n\n```\na = {}\n\n\nfor image,label in dataset:\n    num_label = classes[label]\n    if num_label not in a:\n        a[num_label] = 0\n    a[num_label] += 1\n\na\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/khLhkpvF-image.png)\n\n\n\nWe begin by importing the PyTorch packages and modules. Here we need to download the CIFAR-10 dataset. We loop through the data to determine the number of images for each object (class).\n```\ntorch.manual_seed(43)\nval_size = 5000\ntrain_size = len(dataset) - val_size\n```\n```\ntrain_ds, val_ds = random_split(dataset, [train_size, val_size])\nlen(train_ds), len(val_ds)\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/tf9hEsTC-image.png)\n\n```\nbatch_size=128\n```\n```\ntrain_loader = DataLoader(train_ds, batch_size, shuffle=True, num_workers=4, pin_memory=True)\nval_loader = DataLoader(val_ds, batch_size*2, num_workers=4, pin_memory=True)\ntest_loader = DataLoader(test_dataset, batch_size*2, num_workers=4, pin_memory=True)\n```\n\n\nIn this step, we split the data and then create DataLoader for batches of data. We have written a function to predict accuracy.\n\n\n```\ndef accuracy(outputs, labels):\n    _, preds = torch.max(outputs, dim=1)\n    return torch.tensor(torch.sum(preds == labels).item() / len(preds))\n```\n```\nclass ImageClassificationBase(nn.Module):\n    def training_step(self, batch):\n        images, labels = batch \n        out = self(images)                  # Generate predictions\n        loss = F.cross_entropy(out, labels) # Calculate loss\n        return loss\n    \n    def validation_step(self, batch):\n        images, labels = batch \n        out = self(images)                    # Generate predictions\n        loss = F.cross_entropy(out, labels)   # Calculate loss\n        acc = accuracy(out, labels)           # Calculate accuracy\n        return {'val_loss': loss.detach(), 'val_acc': acc}\n        \n    def validation_epoch_end(self, outputs):\n        batch_losses = [x['val_loss'] for x in outputs]\n        epoch_loss = torch.stack(batch_losses).mean()   # Combine losses\n        batch_accs = [x['val_acc'] for x in outputs]\n        epoch_acc = torch.stack(batch_accs).mean()      # Combine accuracies\n        return {'val_loss': epoch_loss.item(), 'val_acc': epoch_acc.item()}\n    \n    def epoch_end(self, epoch, result):\n        print(\"Epoch [{}], val_loss: {:.4f}, val_acc: {:.4f}\".format(epoch, result['val_loss'], result['val_acc']))\n```\n```\ndef evaluate(model, val_loader):\n    outputs = [model.validation_step(batch) for batch in val_loader]\n    return model.validation_epoch_end(outputs)\n\ndef fit(epochs, lr, model, train_loader, val_loader, opt_func=torch.optim.SGD):\n    history = []\n    optimizer = opt_func(model.parameters(), lr)\n    for epoch in range(epochs):\n        # Training Phase \n        for batch in train_loader:\n            loss = model.training_step(batch)\n            loss.backward()\n            optimizer.step()\n            optimizer.zero_grad()\n        # Validation phase\n        result = evaluate(model, val_loader)\n        model.epoch_end(epoch, result)\n        history.append(result)\n    return history\n```\n\n\nFor this step, we begin training  our model. Here we have written functions to make predictions and evaluate the accuracy.\n\n```\ntorch.cuda.is_available()\n```\n```\ndef get_default_device():\n    \"\"\"Pick GPU if available, else CPU\"\"\"\n    if torch.cuda.is_available():\n        return torch.device('cuda')\n    else:\n        return torch.device('cpu')\n```\n```\ndevice = get_default_device()\ndevice\n```\n```\ndef to_device(data, device):\n    \"\"\"Move tensor(s) to chosen device\"\"\"\n    if isinstance(data, (list,tuple)):\n        return [to_device(x, device) for x in data]\n    return data.to(device, non_blocking=True)\n\nclass DeviceDataLoader():\n    \"\"\"Wrap a dataloader to move data to a device\"\"\"\n    def __init__(self, dl, device):\n        self.dl = dl\n        self.device = device\n        \n    def __iter__(self):\n        \"\"\"Yield a batch of data after moving it to device\"\"\"\n        for b in self.dl: \n            yield to_device(b, self.device)\n\n    def __len__(self):\n        \"\"\"Number of batches\"\"\"\n        return len(self.dl)\n```\n```\ntrain_loader = DeviceDataLoader(train_loader, device)\nval_loader = DeviceDataLoader(val_loader, device)\ntest_loader = DeviceDataLoader(test_loader, device)\n```\n```\ninput_size = 3*32*32\noutput_size = 10\n```\n```\nclass CIFAR10Model(ImageClassificationBase):\n    def __init__(self):\n        super().__init__()\n        self.linear1 = nn.Linear(input_size, 1792)\n        self.linear2 = nn.Linear(1792, 896)\n        self.linear3 = nn.Linear(896, 448)\n        self.linear4 = nn.Linear(448, output_size)\n        \n        \n    def forward(self, xb):\n        # Flatten images into vectors\n        out = xb.view(xb.size(0), -1)\n        # Apply layers & activation functions\n        out = self.linear1(out)\n        out = F.relu(out)\n        out = self.linear2(out)\n        out = F.relu(out)\n        out = self.linear3(out)\n        out = F.relu(out)\n        out = self.linear4(out)\n        return out\n```\n\nNeural Network models need a Graphics Processing Unit (GPU) for making predictions.Therefore, we have written functions to use the GPU. We have also added different linear and non-linear (activation) functions to our model.\n\n```\nmodel = to_device(CIFAR10Model(), device)\n```\n```\nhistory += fit(45, 0.004, model, train_loader, val_loader)\n```\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/qjLGMdU3-image.png)\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/zCQel88o-image.png)\n\n\nThis step helped in predicting the accuracy of our model. This model achieved about 52% accuracy. To view my Jupyter notebook [click here](https://jovian.ml/oneworldcoder/03-cifar10-feedforward) Next week, we will learn how to increase accuracy!\n\n**References**\n[*Deep Learning with PyTorch:Zero to GAN's**](https://jovian.ml/forum/t/start-here-welcome-to-deep-learning-with-pytorch-zero-to-gans/1622?u=oneworldcoder) by [Aakash N S](https://jovian.ml/forum/u/aakashns/summary)\n[*But what is Neural Network?*](https://youtu.be/aircAruvnKk)  By [3Blue1Brown](https://www.3blue1brown.com/)\n[*CIFAR-10* Dataset](https://jovian.ml/outlink?url=https%3A%2F%2Fwww.cs.toronto.edu%2F~kriz%2Fcifar.html)\n",
      "json_metadata": "{\"app\":\"peakd/2020.05.5\",\"format\":\"markdown\",\"description\":\"Code in Deep Neural Networks for images from CIFAR-10.\",\"tags\":[\"ai\",\"code\",\"deeplearning\",\"pytorch\",\"python\",\"nn\",\"neuralnetworks\",\"beginner\"],\"links\":[\"https://jovian.ml/oneworldcoder/03-cifar10-feedforward\",\"https://jovian.ml/forum/t/start-here-welcome-to-deep-learning-with-pytorch-zero-to-gans/1622?u=oneworldcoder\",\"https://jovian.ml/forum/u/aakashns/summary\",\"https://youtu.be/aircAruvnKk\",\"https://www.3blue1brown.com/\",\"https://jovian.ml/outlink?url=https%3A%2F%2Fwww.cs.toronto.edu%2F~kriz%2Fcifar.html\"],\"image\":[\"https://files.peakd.com/file/peakd-hive/cryptopossible/ipelgx0J-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/nK4Gh2LE-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/yyzxAqcr-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/z51CPqg0-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/khLhkpvF-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/tf9hEsTC-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/qjLGMdU3-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/zCQel88o-image.png\"]}",
      "parent_author": "",
      "parent_permlink": "hive-174578",
      "permlink": "deep-learning-neural-networks-on-images-from-cifar-10",
      "title": "Deep Learning: Neural networks on images from CIFAR-10."
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-12T13:33:48",
  "trx_id": "4372a62621ad34b3687cbf38910caaa0777a05fb",
  "trx_in_block": 27,
  "virtual_op": false
}

{
  "block": 44142494,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.294 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 1554563223,
      "total_vote_weight": 434828,
      "voter": "new-steemit",
      "weight": 624
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:47:42",
  "trx_id": "28a5a40629e2387d04cab431f92202670e1fccf4",
  "trx_in_block": 19,
  "virtual_op": true
}

{
  "block": 44142494,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "new-steemit",
      "weight": 200
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:47:42",
  "trx_id": "28a5a40629e2387d04cab431f92202670e1fccf4",
  "trx_in_block": 19,
  "virtual_op": false
}

{
  "block": 44142493,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.293 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 754495326654,
      "total_vote_weight": 434204,
      "voter": "payroll",
      "weight": 329371
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:47:39",
  "trx_id": "2645acff000529162c783935530c9f8f532e6b1b",
  "trx_in_block": 8,
  "virtual_op": true
}

{
  "block": 44142493,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "payroll",
      "weight": 200
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:47:39",
  "trx_id": "2645acff000529162c783935530c9f8f532e6b1b",
  "trx_in_block": 8,
  "virtual_op": false
}

{
  "block": 44142275,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.060 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 199010042,
      "total_vote_weight": 104833,
      "voter": "curbot",
      "weight": 94
    }
  ],
  "op_in_trx": 3,
  "timestamp": "2020-06-09T13:36:42",
  "trx_id": "dcfbe45c97096588242c5c10eef46c61df5d0f0c",
  "trx_in_block": 10,
  "virtual_op": true
}

{
  "block": 44142275,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "curbot",
      "weight": 1000
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-09T13:36:42",
  "trx_id": "dcfbe45c97096588242c5c10eef46c61df5d0f0c",
  "trx_in_block": 10,
  "virtual_op": false
}

{
  "block": 44142165,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.060 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 2618022469,
      "total_vote_weight": 104739,
      "voter": "cizolf",
      "weight": 1243
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:31:12",
  "trx_id": "4109a6d0a84148f9b15c52b18b76ecf728a85732",
  "trx_in_block": 2,
  "virtual_op": true
}

{
  "block": 44142165,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "cizolf",
      "weight": 10000
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:31:12",
  "trx_id": "4109a6d0a84148f9b15c52b18b76ecf728a85732",
  "trx_in_block": 2,
  "virtual_op": false
}

{
  "block": 44142151,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.059 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 310775962,
      "total_vote_weight": 103496,
      "voter": "merlion",
      "weight": 148
    }
  ],
  "op_in_trx": 3,
  "timestamp": "2020-06-09T13:30:30",
  "trx_id": "92a6f9456798766a48a12c11aeb1a5cbed3d09cd",
  "trx_in_block": 19,
  "virtual_op": true
}

{
  "block": 44142151,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "merlion",
      "weight": 100
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-09T13:30:30",
  "trx_id": "92a6f9456798766a48a12c11aeb1a5cbed3d09cd",
  "trx_in_block": 19,
  "virtual_op": false
}

{
  "block": 44142151,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.059 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 21071982,
      "total_vote_weight": 103348,
      "voter": "hungrybear",
      "weight": 10
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:30:30",
  "trx_id": "92a6f9456798766a48a12c11aeb1a5cbed3d09cd",
  "trx_in_block": 19,
  "virtual_op": true
}

{
  "block": 44142151,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "hungrybear",
      "weight": 100
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:30:30",
  "trx_id": "92a6f9456798766a48a12c11aeb1a5cbed3d09cd",
  "trx_in_block": 19,
  "virtual_op": false
}

{
  "block": 44142142,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.059 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 2321725717,
      "total_vote_weight": 103338,
      "voter": "onepercentbetter",
      "weight": 1103
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:30:03",
  "trx_id": "ebe2655e71041a4df4cac59777ae987a7cbe1425",
  "trx_in_block": 23,
  "virtual_op": true
}

{
  "block": 44142142,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "onepercentbetter",
      "weight": 100
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:30:03",
  "trx_id": "ebe2655e71041a4df4cac59777ae987a7cbe1425",
  "trx_in_block": 23,
  "virtual_op": false
}

{
  "block": 44142134,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.058 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 98480965359,
      "total_vote_weight": 102235,
      "voter": "ctime",
      "weight": 47359
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:29:39",
  "trx_id": "80a2626e11e9ac34ad831c4764159195fdcb6272",
  "trx_in_block": 36,
  "virtual_op": true
}

{
  "block": 44142134,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "ctime",
      "weight": 100
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:29:39",
  "trx_id": "80a2626e11e9ac34ad831c4764159195fdcb6272",
  "trx_in_block": 36,
  "virtual_op": false
}

{
  "block": 44142134,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.030 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 110180997663,
      "total_vote_weight": 54876,
      "voter": "digital.mine",
      "weight": 54301
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:29:39",
  "trx_id": "942a0d59e970638cb29414c82a31c052d04c1aca",
  "trx_in_block": 33,
  "virtual_op": true
}

{
  "block": 44142134,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "digital.mine",
      "weight": 50
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:29:39",
  "trx_id": "942a0d59e970638cb29414c82a31c052d04c1aca",
  "trx_in_block": 33,
  "virtual_op": false
}

{
  "block": 44142115,
  "op": [
    "effective_comment_vote",
    {
      "author": "cryptopossible",
      "pending_payout": "0.000 HBD",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "rshares": 1152168268,
      "total_vote_weight": 575,
      "voter": "besheda",
      "weight": 471
    }
  ],
  "op_in_trx": 1,
  "timestamp": "2020-06-09T13:28:42",
  "trx_id": "fb5b8819613577f3907da9899c97734633e276b0",
  "trx_in_block": 13,
  "virtual_op": true
}

{
  "block": 44142115,
  "op": [
    "vote",
    {
      "author": "cryptopossible",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "voter": "besheda",
      "weight": 4300
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:28:42",
  "trx_id": "fb5b8819613577f3907da9899c97734633e276b0",
  "trx_in_block": 13,
  "virtual_op": false
}

{
  "block": 44142037,
  "op": [
    "custom_json",
    {
      "id": "follow",
      "json": "[\"reblog\",{\"account\":\"cryptopossible\",\"author\":\"cryptopossible\",\"permlink\":\"deep-learning-predicting-the-average-income-of-residents-in-california\"}]",
      "required_auths": [],
      "required_posting_auths": [
        "cryptopossible"
      ]
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:24:48",
  "trx_id": "d705bc9b81768500df8ea9dd1fd0a0bb90b5ba2c",
  "trx_in_block": 7,
  "virtual_op": false
}

{
  "block": 44142033,
  "op": [
    "comment",
    {
      "author": "cryptopossible",
      "body": "We can train the computer to predict the average income of people by considering a combination of different factors such as their residential neighbourhood, house size, house age, population of the city/town. I have written a code which can predict with about 98% accuracy  the average income of residents of California.\n \nLet's discuss the code here.\n \n```\nimport torch\nimport jovian\nimport torchvision\nimport torch.nn as nn\nimport pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport torch.nn.functional as F\nfrom torchvision.datasets.utils import download_url\nfrom torch.utils.data import DataLoader, TensorDataset, random_split\n\n```\n\nHere I am importing required PyTorch modules.\n\n```# Hyperparameters\nbatch_size=64\nlearning_rate=1e-5\n\n\n# Other constants\nDATASET_URL = \"/kaggle/input/housing.csv\"\nDATA_FILENAME = \"housing.csv\"\nTARGET_COLUMN = 'ocean_proximity'\ninput_size=9\noutput_size=1\n\n```\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/A2AZGXEm-image.png)\n\n \nFirst we need to gather as much data as possible about the residents of California. The dataset can be in .csv file. \n\n\n```\ndef customize_dataset(dataframe_raw):\n    dataframe = dataframe_raw.copy(deep=True)\n    # drop some columns\n    dataframe = dataframe.drop(['longitude', 'latitude'], axis=1)\n    #for col in ['housing_median_age', 'total_rooms', 'total_bedrooms', 'population', \n    #            'households', 'median_income' ,'median_house_value']:\n    #    # normalizing incoming data\n    #    dataframe[col] = (dataframe[col] - min(dataframe[col])) / (max(dataframe[col]) - min(dataframe[col]))\n    \n    # dropping any row that contains at least on missing value\n    # if you dont do that, loss function will be returning nan\n    dataframe = dataframe.dropna(axis=0)  \n    \n    return dataframe\n\n```\nBefore we begin training the model, we need to look at the data and sometimes we may need to customise the data to increase the accuracy of the result. In this case, I have excluded partially missing information in the dataset which has helped predict the accuracy.\n```\ndataframe = customize_dataset(dataframe)\n\ndataframe.head()\n```\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/vJnnoXlO-image.png)\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/zU6fDFSV-image.png)\n\n```\ninput_cols = list(dataframe.columns[0:4])+list(dataframe.columns[-1:])\n\ninput_cols\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/yQfGxh3y-image.png)\n\n```\noutput_cols = list(dataframe.columns[5:6])\nprint(len(output_cols))\nprint(output_cols)\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/7TkQOAJZ-image.png)\n\nAfter customising the data, we are ready to begin coding in the Jupyter notebook. Then, we segregate the data on which we will train our model.\n\n \n```\n# Convert from Pandas dataframe to numpy arrays\n\n\ndef dataframe_to_arrays(dataframe):\n    # Make a copy of the original dataframe\n    dataframe1 = dataframe.copy(deep=True)\n    # Convert non-numeric categorical columns to numbers\n    for col in categorical_cols:\n        dataframe1[col] = dataframe1[col].astype('category').cat.codes\n    # Extract input & outupts as numpy arrays\n    inputs_array = dataframe1[input_cols].to_numpy()\n    targets_array = dataframe1[output_cols].to_numpy()\n    return inputs_array, targets_array\n\ninputs_array, targets_array = dataframe_to_arrays(dataframe)\ninputs_array, targets_array\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/4MoomOlf-image.png)\nWe now convert the segregated data to Numpy arrays.\n\n\n```\n#To convert Numpy arrays to PyTorch Tensors\n\ninputs = torch.from_numpy(inputs_array).type(torch.float32)\ntargets = torch.from_numpy(targets_array).type(torch.float32)\n\n```\nWe will be using PyTorch to train our model so we need to convert the NumPy arrays into PyTorch tensors.\n \n```\ninputs.dtype, targets.dtype\ndataset = TensorDataset(inputs, targets)\nval_percent = 0.1 # between 0.1 and 0.2\nval_size = int(num_rows * val_percent)\ntrain_size = num_rows - val_size\n\n\ntrain_ds, val_ds = random_split(dataset, [train_size, val_size]) # Use the random_split function to split dataset into 2 parts of the desired length\n\ntrain_loader = DataLoader(train_ds, batch_size, shuffle=True)\nval_loader = DataLoader(val_ds, batch_size)\n\nfor xb, yb in train_loader:\n    print(\"inputs:\", xb)\n    print(\"targets:\", yb)\n    break\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/Pl5aorPs-image.png)\n\n```\ninput_size = len(input_cols)\noutput_size = len(output_cols)\nprint(len(input_cols))\nprint(len(output_cols))\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/LufsjzWe-image.png)\n\n\n \nWe can train our model with the data which is now in the form of PyTorch tensors.\n \n```\nclass HousingModel(nn.Module):\n    def __init__(self):\n        super().__init__()\n        self.linear = nn.Linear(input_size, output_size)\n        \n    def forward(self, xb):\n        out = self.linear(xb)\n        return out\n    \n    def training_step(self, batch):\n        inputs, targets = batch \n        out = self(inputs)                 # Generate predictions\n        loss = F.l1_loss(out, targets)    # Calculate loss\n        return loss\n    \n    def validation_step(self, batch):\n        inputs, targets = batch \n        out = self(inputs)                 # Generate predictions\n        loss = F.l1_loss(out, targets)    # Calculate loss\n        return {'val_loss': loss.detach()}\n        \n    def validation_epoch_end(self, outputs):\n        batch_losses = [x['val_loss'] for x in outputs]\n        epoch_loss = torch.stack(batch_losses).mean()   # Combine losses\n        return {'val_loss': epoch_loss.item()}\n    \n    def epoch_end(self, epoch, result):\n        print(\"Epoch [{}], val_loss: {:.4f}\".format(epoch, result['val_loss']))\n    \nmodel = HousingModel()\n\ndef evaluate(model, val_loader):\n    outputs = [model.validation_step(batch) for batch in val_loader]\n    return model.validation_epoch_end(outputs)\n\ndef fit(epochs, learning_rate, model, train_loader, val_loader, opt_func=torch.optim.SGD):\n    history = []\n    optimizer = opt_func(model.parameters(), learning_rate)\n    for epoch in range(epochs):\n        # Training Phase \n        for batch in train_loader:\n            loss = model.training_step(batch)\n            loss.backward()\n            optimizer.step()\n            optimizer.zero_grad()\n        # Validation phase\n        result = evaluate(model, val_loader)\n        model.epoch_end(epoch, result)\n        history.append(result)\n    return history\n\nresult = evaluate(model, val_loader)\nresult\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/pQJPJBAg-image.png)\n\n```\nlist(model.parameters())\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/Jl63uEwg-image.png)\n\n\n \nOnce our model is trained we can make predictions with it.\n\n```\nhistory2= fit(50, 1e-8, model, train_loader, val_loader, opt_func=torch.optim.SGD)\nhistory2\n\n```\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/nRe9v6cX-image.png)\n\n![image.png](https://files.peakd.com/file/peakd-hive/cryptopossible/45rEo3cU-image.png)\n\n\n \nTo view the Jupyter notebook [click here](https://jovian.ml/oneworldcoder/02-housing-optional). Feel free to give me any feedback or ask questions.\n \nIf you want to get into Deep Learning I highly recommend [\"Deep Learning with PyTorch: Zero to GAN’s”](https://jovian.ml/forum/t/start-here-welcome-to-deep-learning-with-pytorch-zero-to-gans/1622) taught by [Aakash N S](https://jovian.ml/forum/u/aakashns/summary) of [Jovian.ml](https://www.jovian.ml/). \n",
      "json_metadata": "{\"app\":\"peakd/2020.05.5\",\"format\":\"markdown\",\"description\":\"Zero to GAN's is an online certificate course that takes a hands-on coding-focused approach to deep learning.\",\"tags\":[\"zerotogans\",\"jovianml\",\"ai\",\"deeplearning\",\"machinelearning\",\"ml\",\"pytorch\"],\"links\":[\"https://jovian.ml/oneworldcoder/02-housing-optional\",\"https://jovian.ml/forum/t/start-here-welcome-to-deep-learning-with-pytorch-zero-to-gans/1622\",\"https://jovian.ml/forum/u/aakashns/summary\",\"https://www.jovian.ml/\"],\"image\":[\"https://files.peakd.com/file/peakd-hive/cryptopossible/A2AZGXEm-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/vJnnoXlO-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/zU6fDFSV-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/yQfGxh3y-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/7TkQOAJZ-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/4MoomOlf-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/Pl5aorPs-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/LufsjzWe-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/pQJPJBAg-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/Jl63uEwg-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/nRe9v6cX-image.png\",\"https://files.peakd.com/file/peakd-hive/cryptopossible/45rEo3cU-image.png\"]}",
      "parent_author": "",
      "parent_permlink": "hive-174578",
      "permlink": "deep-learning-predicting-the-average-income-of-residents-in-california",
      "title": "Deep Learning: Predicting the average income of residents in California."
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-09T13:24:36",
  "trx_id": "c1c6f0f2edb8c9037072ccc3a0b0b2dc064f48df",
  "trx_in_block": 2,
  "virtual_op": false
}

{
  "block": 44139747,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "qbaqm8"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-09T11:30:12",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44138117,
  "op": [
    "curation_reward",
    {
      "author": "appics",
      "curator": "cryptopossible",
      "payout_must_be_claimed": true,
      "permlink": "sneak-peek-new-appics-blockchain-with-cross-posting-to-hive",
      "reward": "25.298668 VESTS"
    }
  ],
  "op_in_trx": 76,
  "timestamp": "2020-06-09T10:08:27",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44078450,
  "op": [
    "fill_vesting_withdraw",
    {
      "deposited": "75.735 HIVE",
      "from_account": "cryptopossible",
      "to_account": "cryptopossible",
      "withdrawn": "147403.402918 VESTS"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-07T08:17:42",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44077385,
  "op": [
    "account_witness_vote",
    {
      "account": "cryptopossible",
      "approve": true,
      "witness": "threespeak"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-07T07:24:18",
  "trx_id": "6c15437093a0b3e937a95a3cd96523a293aa0b06",
  "trx_in_block": 20,
  "virtual_op": false
}

{
  "block": 44077380,
  "op": [
    "account_witness_vote",
    {
      "account": "cryptopossible",
      "approve": false,
      "witness": "arcange"
    }
  ],
  "op_in_trx": 0,
  "timestamp": "2020-06-07T07:24:03",
  "trx_id": "7785bbcfcadc01967a5a736026efe32cc2afbe15",
  "trx_in_block": 5,
  "virtual_op": false
}

{
  "block": 44026283,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "re-cryptopossible-qb3fbb"
    }
  ],
  "op_in_trx": 2,
  "timestamp": "2020-06-05T12:42:54",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}

{
  "block": 44023822,
  "op": [
    "comment_payout_update",
    {
      "author": "cryptopossible",
      "permlink": "pytorch-a-deep-learning-journey"
    }
  ],
  "op_in_trx": 14,
  "timestamp": "2020-06-05T10:39:27",
  "trx_id": "0000000000000000000000000000000000000000",
  "trx_in_block": 4294967295,
  "virtual_op": true
}