{
  "trx_id": "f393184a6e5b26b904528af8a92e7f71f3ba3e8f",
  "block": 43192257,
  "trx_in_block": 8,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2020-05-08T08:35:39",
  "op": [
    "delegate_vesting_shares",
    {
      "delegator": "steem",
      "delegatee": "djlemonskull",
      "vesting_shares": "0.000000 VESTS"
    }
  ]
}

{
  "trx_id": "58ece06eb9dd719ba47b464106338d14f87404cf",
  "block": 38176211,
  "trx_in_block": 43,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2019-11-14T19:49:18",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "dhumphrey",
      "permlink": "gpg-pgp-command-line-basic-tutorial",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "fe8b2e7af397c661ac9c4231baa05a998f5f2271",
  "block": 37214281,
  "trx_in_block": 26,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2019-10-12T08:37:36",
  "op": [
    "delegate_vesting_shares",
    {
      "delegator": "steem",
      "delegatee": "djlemonskull",
      "vesting_shares": "1976.846111 VESTS"
    }
  ]
}

{
  "trx_id": "7308f894d20748785c44eceb2478e2574146350f",
  "block": 36641987,
  "trx_in_block": 11,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2019-09-22T10:43:06",
  "op": [
    "comment",
    {
      "parent_author": "djlemonskull",
      "parent_permlink": "why-btc-price-is-stable-around-6-5k-usd-2-hypothesis",
      "author": "steemitboard",
      "permlink": "steemitboard-notify-djlemonskull-20190922t104306000z",
      "title": "",
      "body": "Congratulations @djlemonskull! You received a personal award!\n\n<table><tr><td>https://steemitimages.com/70x70/http://steemitboard.com/@djlemonskull/birthday1.png</td><td>Happy Birthday! - You are on the Steem blockchain for 1 year!</td></tr></table>\n\n<sub>_You can view [your badges on your Steem Board](https://steemitboard.com/@djlemonskull) and compare to others on the [Steem Ranking](https://steemitboard.com/ranking/index.php?name=djlemonskull)_</sub>\n\n\n**Do not miss the last post from @steemitboard:**\n<table><tr><td><a href=\"https://steemit.com/steemfest/@steemitboard/steemitboard-supports-the-steemfest-travel-reimbursement-fund\"><img src=\"https://steemitimages.com/64x128/https://cdn.steemitimages.com/DQmXDHs9xfx8ZZ3DESFUqHRUQAcQT5kUWobArsRoJg2Yz1F/image.png\"></a></td><td><a href=\"https://steemit.com/steemfest/@steemitboard/steemitboard-supports-the-steemfest-travel-reimbursement-fund\">SteemitBoard supports the SteemFest⁴ Travel Reimbursement Fund.</a></td></tr></table>\n\n###### [Vote for @Steemitboard as a witness](https://v2.steemconnect.com/sign/account-witness-vote?witness=steemitboard&approve=1) to get one more award and increased upvotes!",
      "json_metadata": "{\"image\":[\"https://steemitboard.com/img/notify.png\"]}"
    }
  ]
}

{
  "trx_id": "6a250654dd32f0eec017a6f9f9b761c50218eeb0",
  "block": 36187275,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2019-09-06T14:42:15",
  "op": [
    "delegate_vesting_shares",
    {
      "delegator": "steem",
      "delegatee": "djlemonskull",
      "vesting_shares": "15485.685570 VESTS"
    }
  ]
}

{
  "trx_id": "27824713a749360d6d7b11d1abaee9a65a0ad797",
  "block": 34619909,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2019-07-13T07:21:51",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "radix42",
      "permlink": "pirate-s-chest-nodes-now-available-for-order",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "2e158ea4c3c5639aa1ffe552c3baf09a7508dee4",
  "block": 28164121,
  "trx_in_block": 2,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-30T21:41:15",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "giffordde2000",
      "permlink": "upgrade-guide-a-summary-of-new-features-coming-on-december-15-november-27-2018-ben-blockchain-cryptocurrency-komodo",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "b7494673e719779a0cd214ef9ade01b991576ec9",
  "block": 28075232,
  "trx_in_block": 1,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-27T19:35:21",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "giffordde2000",
      "permlink": "pirate-independent-secure-anonymous-an-interview-with-the-piratechain-team",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "0000000000000000000000000000000000000000",
  "block": 27855031,
  "trx_in_block": 4294967295,
  "op_in_trx": 0,
  "virtual_op": 5,
  "timestamp": "2018-11-20T03:59:33",
  "op": [
    "curation_reward",
    {
      "curator": "djlemonskull",
      "reward": "4.029486 VESTS",
      "comment_author": "piratechain",
      "comment_permlink": "welcome-aboard-pirates-pirate-the-anonymous-and-private-cryptocurrency"
    }
  ]
}

{
  "trx_id": "f3982d6c2167deb61c2e571fcf84bc5c9057789b",
  "block": 27689015,
  "trx_in_block": 10,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-14T09:33:21",
  "op": [
    "vote",
    {
      "voter": "pabisovsky",
      "author": "djlemonskull",
      "permlink": "why-btc-price-is-stable-around-6-5k-usd-2-hypothesis",
      "weight": 500
    }
  ]
}

{
  "trx_id": "e9a0a617edda060bf9e320afc2026f1dd248b608",
  "block": 27688922,
  "trx_in_block": 24,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-14T09:28:42",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "btc",
      "author": "djlemonskull",
      "permlink": "why-btc-price-is-stable-around-6-5k-usd-2-hypothesis",
      "title": "Why BTC price is stable around 6.5k $ - 2 hypothesis",
      "body": "![](https://cdn.steemitimages.com/DQmfSGX5V6Nv7fCSi1hFYJ5ac6jKCVSuocoNdn7LATtuGfU/image.png)\n\nMany BTC price observers are amazed by the price stability recent months. Me too. Since I watch crypto market didn't see such for 3 years. Does not look natural...\n\nWhat could be reasons behind it? I see two possibilities:\n\n1. [a conspiracy theory] Some big players make BTC price stable with their buy/sell orders to get ETF (which requires price stability). \n\nIf this scenario would be true it also means someone is able to control price to a degree where trading is not even gambling - someone has control of it, so he is the winner.\n\n2. Same effect I imagine could be created by big players abandoning market for some time. Big whales make big waves. Small fishes - small waves. It's possible currently on BTC market there are relatively small fishes - they don't make big waves and price stays stable just above the mining costs.\n\nOther theories?",
      "json_metadata": "{\"tags\":[\"btc\",\"price\"],\"image\":[\"https://cdn.steemitimages.com/DQmfSGX5V6Nv7fCSi1hFYJ5ac6jKCVSuocoNdn7LATtuGfU/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "8647002684866b7d406d8c889e74b098a05d96ca",
  "block": 27672476,
  "trx_in_block": 27,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-13T19:45:51",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "piratechain",
      "permlink": "an-interview-with-the-pirate-chain-development-team-by-lordcatoshi",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "edba92b3c10dacedda91eee5d1cc19da70a33b2d",
  "block": 27672456,
  "trx_in_block": 32,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-13T19:44:51",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "giffordde2000",
      "permlink": "usdarrr-pirate-the-new-privacy-coin-this-is-my-message-to-you-please-read-and-share-thank-you",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "7a347de58ca857d9c36a699589131ef65ceb67d2",
  "block": 27660756,
  "trx_in_block": 30,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-13T09:59:33",
  "op": [
    "comment",
    {
      "parent_author": "djlemonskull",
      "parent_permlink": "getting-familiar-with-neural-networks",
      "author": "steemitboard",
      "permlink": "steemitboard-notify-djlemonskull-20181113t095934000z",
      "title": "",
      "body": "Congratulations @djlemonskull! You have completed the following achievement on the Steem blockchain and have been rewarded with new badge(s) :\n\n<table><tr><td>https://steemitimages.com/60x70/http://steemitboard.com/@djlemonskull/votes.png?201811130824</td><td>You made more than 10 upvotes. Your next target is to reach 50 upvotes.</td></tr>\n</table>\n\n<sub>_[Click here to view your Board of Honor](https://steemitboard.com/@djlemonskull)_</sub>\n<sub>_If you no longer want to receive notifications, reply to this comment with the word_ `STOP`</sub>\n\n\n\n**Do not miss the last post from @steemitboard:**\n<table><tr><td><a href=\"https://steemit.com/steemfest/@steemitboard/steemfest3-and-steemitboard-meet-the-steemians-contest\"><img src=\"https://steemitimages.com/64x128/https://cdn.steemitimages.com/DQmeLukvNFRsa7RURqsFpiLGEZZD49MiU52JtWmjS5S2wtW/image.png\"></a></td><td><a href=\"https://steemit.com/steemfest/@steemitboard/steemfest3-and-steemitboard-meet-the-steemians-contest\">SteemFest3 and SteemitBoard - Meet the Steemians Contest</a></td></tr></table>\n\n> Support [SteemitBoard's project](https://steemit.com/@steemitboard)! **[Vote for its witness](https://v2.steemconnect.com/sign/account-witness-vote?witness=steemitboard&approve=1)** and **get one more award**!",
      "json_metadata": "{\"image\":[\"https://steemitboard.com/img/notify.png\"]}"
    }
  ]
}

{
  "trx_id": "9a739e2aeec26240cc67d33b5b20b72d51121a41",
  "block": 27657888,
  "trx_in_block": 16,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-13T07:36:06",
  "op": [
    "custom_json",
    {
      "required_auths": [],
      "required_posting_auths": [
        "djlemonskull"
      ],
      "id": "follow",
      "json": "[\"follow\",{\"follower\":\"djlemonskull\",\"following\":\"piratechain\",\"what\":[\"blog\"]}]"
    }
  ]
}

{
  "trx_id": "1c91a3f1d3e8247c134695429fb62f907a995021",
  "block": 27657882,
  "trx_in_block": 6,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-13T07:35:48",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "piratechain",
      "permlink": "welcome-aboard-pirates-pirate-the-anonymous-and-private-cryptocurrency",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "e88a1f3cc18c0eb7ab1c646cef3a9ef91366709f",
  "block": 27584010,
  "trx_in_block": 15,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-10T17:59:36",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -97,16 +97,109 @@\n matters.\n+ After you read it you will understand that 2 myths regarding anonymity are clearly not true.\n %3C/b%3E%3Cbr%3E\n@@ -2191,16 +2191,760 @@\n br%3E%3Cbr%3E%0A\n+%0A%3Cb%3EMyths debunked%3C/b%3E%0A%3Cp%3EMyth number 1 says that anonymity is for criminals mostly. But as you understand now - it's for majority of regular people to protect them from criminals in the first place.%3C/p%3E%0A%3Cp%3EMyth number 2 says that anonymity is for rich people. Also false. Most people suffering from thief's or robbers are regular people.  It's very hard to steal multi-millionaires' life savings, but it's easy to do so to normal people, which will suffer much more in that case. %3C/p%3E%0ABoth myths basically claim that anonymity is for small minority and anyone who spreads them actually works against the majority of people who needs protection (=anonymity) from criminals at least on the same level as regular bank account give them.%3Cbr%3E%3Cbr%3E%0A%0A\n %3Cb%3EPS%3C/b\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\",\"links\":[\"https://twitter.com/PirateChain/\",\"https://pirate.black/\"]}"
    }
  ]
}

{
  "trx_id": "462a04e5d50f2a890bec418f03bf53e96afdc0f0",
  "block": 27473587,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T21:55:21",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "democracy",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "title": "Open Democracy and extinction of politicians",
      "body": "@@ -775,16 +775,19 @@\n ocracy. \n+%3Cb%3E\n Politici\n@@ -840,16 +840,20 @@\n ad ideas\n+%3C/b%3E\n , so if \n@@ -4347,16 +4347,57 @@\n  engine \n++ online profile for personal trust rank \n are need\n@@ -4710,17 +4710,16 @@\n titution\n-s\n  and the\n",
      "json_metadata": "{\"tags\":[\"democracy\",\"politics\",\"massive\",\"knowledge\",\"processing\"],\"image\":[\"https://assets.fnlondon.com/2016/02/IMG009915_full4x3.jpg\",\"https://cdn.steemitimages.com/DQmU74n1JUU9yAdu2hrq9DZbDAaU6Fe1BrSWvin9XzZ4NLE/dinosaur.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "8a1ebd8678aac60a8e7295ddf327151793519c7d",
  "block": 27473481,
  "trx_in_block": 13,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T21:50:03",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "democracy",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "title": "Open Democracy and extinction of politicians",
      "body": "@@ -345,19 +345,25 @@\n r past, \n-lik\n+there wer\n e Greek \n@@ -367,14 +367,16 @@\n ek c\n-ountri\n+ity-stat\n es, \n",
      "json_metadata": "{\"tags\":[\"democracy\",\"politics\",\"massive\",\"knowledge\",\"processing\"],\"image\":[\"https://assets.fnlondon.com/2016/02/IMG009915_full4x3.jpg\",\"https://cdn.steemitimages.com/DQmU74n1JUU9yAdu2hrq9DZbDAaU6Fe1BrSWvin9XzZ4NLE/dinosaur.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "202e876d8a9960cc04f13c641da6b4536e137f8a",
  "block": 27472624,
  "trx_in_block": 28,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T21:07:06",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "getting-familiar-with-neural-networks",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "03f46cbb97264931199aa8016901ee8d18db2c63",
  "block": 27472531,
  "trx_in_block": 15,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T21:02:27",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "neuralnetworks",
      "author": "djlemonskull",
      "permlink": "getting-familiar-with-neural-networks",
      "title": "Getting familiar with neural networks",
      "body": "@@ -2661,16 +2661,54 @@\n  I used \n+%3Ca href=%22https://www.tensorflow.org/%22%3E\n Tensorfl\n@@ -2709,16 +2709,20 @@\n nsorflow\n+%3C/a%3E\n  python \n",
      "json_metadata": "{\"tags\":[\"neuralnetworks\",\"deeplearning\",\"machinelearning\"],\"image\":[\"https://cdn.steemitimages.com/DQmNQphXs2CsbDQK9RDTkxtX1p3xbg1FqRoUYinSuiviw7S/ai2.PNG\"],\"links\":[\"http://yann.lecun.com/exdb/mnist/\",\"https://en.wikipedia.org/wiki/Multilayer_perceptron\",\"http://scikit-learn.org/\",\"https://www.tensorflow.org/\",\"https://keras.io/applications/\",\"https://towardsdatascience.com/build-your-own-convolution-neural-network-in-5-mins-4217c2cf964f\",\"https://towardsdatascience.com/intuitively-understanding-convolutions-for-deep-learning-1f6f42faee1\",\"https://medium.com/comet-app/review-of-deep-learning-algorithms-for-image-classification-5fdbca4a05e2\",\"https://towardsdatascience.com/a-guide-to-an-efficient-way-to-build-neural-network-architectures-part-ii-hyper-parameter-42efca01e5d7\",\"https://medium.com/@jon.froiland/python-deep-learning-part-3-9d9e4cf9035c\",\"https://www.programcreek.com/python/example/100068/keras.applications.resnet50.ResNet50\",\"https://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf\",\"https://hackernoon.com/a-comprehensive-design-guide-for-image-classification-cnns-46091260fb92\",\"https://towardsdatascience.com/how-does-a-face-detection-program-work-using-neural-networks-17896df8e6ff\",\"https://blog.athelas.com/a-brief-history-of-cnns-in-image-segmentation-from-r-cnn-to-mask-r-cnn-34ea83205de4\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "5dd80d770ea75341b989ee36735fb28ce9752b87",
  "block": 27472444,
  "trx_in_block": 11,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T20:58:06",
  "op": [
    "vote",
    {
      "voter": "coinboxstore",
      "author": "djlemonskull",
      "permlink": "getting-familiar-with-neural-networks",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "2899e73c5012a6046b5d4a4eafe2db2d026b80cb",
  "block": 27472441,
  "trx_in_block": 7,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-11-06T20:57:57",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "neuralnetworks",
      "author": "djlemonskull",
      "permlink": "getting-familiar-with-neural-networks",
      "title": "Getting familiar with neural networks",
      "body": "<div class=\"pull-right\"><img src=\"https://cdn.steemitimages.com/DQmNQphXs2CsbDQK9RDTkxtX1p3xbg1FqRoUYinSuiviw7S/ai2.PNG\"></div>My next step in approaching BTC price prediction using AI is to get familiar with neural networks and confirm I can use it correctly.<br><br>\n\nTo check my understanding of this tool I decided to test myself - if I can achieve good results on some publicly known data samples. Found that one of most documented databases with results is  <a href=\"http://yann.lecun.com/exdb/mnist/\"> MNIST</a>  - images 28x28 of handwritten digits 0-9 tagged with correct value assigned. Data set is split into 2 parts: 60k training examples and 10k test examples.\n\n\n<b>Test 1 - Multilayer Perceptron</b>\n\nFirst I wanted to see how far I can go with “most basic” neural network - a model called <a href=\"https://en.wikipedia.org/wiki/Multilayer_perceptron\">Multilayer Perceptron</a> (MLP). Why worth trying? This kind of neural network even with 1 hidden layer is said to be able to recreate any nonlinear input-output relation (also I tried this - and works fine for Gauss function or sinus!).\n\nTo try MLP I used <a href=\"http://scikit-learn.org/\">scikit learn</a>, well documented python library. I played around different parameters ( nr of layers, amount of hidden layer neurons, activation function, solver, some others...) to see how neural network works. The significant parameters I found were solver and hidden layer size. \n\nWith following config <b>MLPClassifier( hidden_layer_sizes=(38*5, ), solver='lbfgs', max_iter=500) </b>I could get as low as 2.2% errors on test set, which is quite good and in range of 1.53-3.05% results on MNIST page for 3 layers neural network. If my results were significantly better or significantly worse - I would know most probably I did something wrong. Seems not the case, so I can move forward.\n\nNote - here my reference are MNIST results for similar NN model. If I was doing prediction of unknown data with unknown results it would require much more work and more detailed analysis.\n\nMost important things to note:\n- To normalize your data correctly (for this purpose I scaled each image separately to have values from Gaussian distribution).\n- To set proper size of hidden layer - which is said to be somewhere between input and output layers size. Input in this case was 28x28 and output 10 (number of classes)\n\n<b>Test2 - CNN - Convolutional Neural Network</b>\n\nCNNs are a bit more sophisticated NN, which applies internal filters, helping to extract patterns. CNNs are used for image classifications a lot, like face recognition. Should be better for detecting handwritten digits, right? To check that I used Tensorflow python library for 2 reasons - it utilizes GPU processor and it allows for easy CNN design using Keras module. Also there is a lot of documentation and interesting articles containing examples of such CNNs in Tensorflow/Keras.\n\nI found following model getting good results (down to 0.8% error rate on testing set when lucky):\n\nmodel = Sequential() # model type\nmodel.add(Conv2D(32, kernel_size=(7,7), strides=(1, 1), activation='relu', input_shape=(28,28,1)))\nmodel.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2))) \nmodel.add(Conv2D(64, (3, 3), activation='relu'))\nmodel.add(MaxPooling2D(pool_size=(2, 2)))\nmodel.add(Flatten())\nmodel.add(Dense(512, activation='relu'))\nmodel.add(Dense(10, activation='softmax'))\nmodel.compile(loss=tf.keras.losses.categorical_crossentropy, optimizer=tf.keras.optimizers.Adam(), metrics=['accuracy'])\nmodel.fit(X, y_train,          batch_size=150,          epochs=4,           shuffle=True)\n\nAs you can see CNN consists of several layers - in this example they are 2D convolutional layers which create a 2D filters over the image, split by MaxPooling layers which resample the data taking most significant inputs.\n\nI’m not going any deeper here (to understand why it works this way, not the other) then simply checking few articles to see it’s how CNNs are designed for best results and practiced by many.\n\nI’m also quite satisfied with the results - achieving almost 2x-3x less errors (0.8% vs 2.2%) when comparing CNN to MLP and also achieving similar results to relevant articles on MNIST page (0.8%-09% test error rate).\n\nI was surprised using this module batch size have big impact on results and changing it to 180 or 90 showed worse results, even when adjusted epochs. Need to keep it in mind when working on unknown data.\n\n<b>Test 3 - modern deep learning for pattern/image recognition</b>\n\nAfter having results for MLP and CNN I tried to find if something better exist… and the search was very promising - every year there are contests for CNNs’ designers for a better CNN model. These are very sophisticated models designed to learn recognize many real life objects. These models usually have much more than 10 layers (for example InceptionResNetV2 has 572 layers!).\n\nI was happy to find these models were included in <a href=\"https://keras.io/applications/\">Tensorflow/Keras</a> and needed very little adjustments to run on MNIST dataset. I tried few of them: Xception, InceptionResNetV2, ResNet50 and InceptionV3. This time calculations took about 15-40 min depending on epoch number and model.\n\nI was deeply disappointed. Calculations took about 10x-100x more time then using the “normal” CNN described above and results were on the level of Multilayer Perceptron and sometimes much worse… I tried to play around changing epochs to avoid over-learning, but couldn’t get a decent results. So I think these super-fancy neural network give good results only on the data set they were designed to work on (maybe colors and image size is very important?). Still I believe these could give better results, but probably need more time tuning it to learn different data set. It is also possible the results was not good due to small batch numbers, but I could not afford higher batches due to lack of memory (used 32-64). The best results I achieved for InceptionV3 , where it worked with batch size 150. After 3 epochs I got 1.8% errors on test set.\n\n<b>Summary</b>\n\nIt seems best I can use right now without going deeply into creating my own neural network designs is to use CNN similar to the one described above. It gives good results and calculates very fast on GPU (about 1 minute total time).\n\n<b>Good read</b>\n\nCNN explained\nhttps://towardsdatascience.com/build-your-own-convolution-neural-network-in-5-mins-4217c2cf964f\nhttps://towardsdatascience.com/intuitively-understanding-convolutions-for-deep-learning-1f6f42faee1\n\nHistory of recent year's Deep Learning evolution\nhttps://medium.com/comet-app/review-of-deep-learning-algorithms-for-image-classification-5fdbca4a05e2\n\nBuilding CNN guide\nhttps://towardsdatascience.com/a-guide-to-an-efficient-way-to-build-neural-network-architectures-part-ii-hyper-parameter-42efca01e5d7\n\n\nPractical Keras Example\nhttps://medium.com/@jon.froiland/python-deep-learning-part-3-9d9e4cf9035c\n\nMany code sippets\nhttps://www.programcreek.com/python/example/100068/keras.applications.resnet50.ResNet50\n\n\n1D CNN usage\nhttps://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf\n\nDesign guide for image classification CNN\nhttps://hackernoon.com/a-comprehensive-design-guide-for-image-classification-cnns-46091260fb92\n\n\nFace detection\nhttps://towardsdatascience.com/how-does-a-face-detection-program-work-using-neural-networks-17896df8e6ff\n\nImage elements detection\nhttps://blog.athelas.com/a-brief-history-of-cnns-in-image-segmentation-from-r-cnn-to-mask-r-cnn-34ea83205de4",
      "json_metadata": "{\"tags\":[\"neuralnetworks\",\"deeplearning\",\"machinelearning\"],\"image\":[\"https://cdn.steemitimages.com/DQmNQphXs2CsbDQK9RDTkxtX1p3xbg1FqRoUYinSuiviw7S/ai2.PNG\"],\"links\":[\"http://yann.lecun.com/exdb/mnist/\",\"https://en.wikipedia.org/wiki/Multilayer_perceptron\",\"http://scikit-learn.org/\",\"https://keras.io/applications/\",\"https://towardsdatascience.com/build-your-own-convolution-neural-network-in-5-mins-4217c2cf964f\",\"https://towardsdatascience.com/intuitively-understanding-convolutions-for-deep-learning-1f6f42faee1\",\"https://medium.com/comet-app/review-of-deep-learning-algorithms-for-image-classification-5fdbca4a05e2\",\"https://towardsdatascience.com/a-guide-to-an-efficient-way-to-build-neural-network-architectures-part-ii-hyper-parameter-42efca01e5d7\",\"https://medium.com/@jon.froiland/python-deep-learning-part-3-9d9e4cf9035c\",\"https://www.programcreek.com/python/example/100068/keras.applications.resnet50.ResNet50\",\"https://blog.goodaudience.com/introduction-to-1d-convolutional-neural-networks-in-keras-for-time-sequences-3a7ff801a2cf\",\"https://hackernoon.com/a-comprehensive-design-guide-for-image-classification-cnns-46091260fb92\",\"https://towardsdatascience.com/how-does-a-face-detection-program-work-using-neural-networks-17896df8e6ff\",\"https://blog.athelas.com/a-brief-history-of-cnns-in-image-segmentation-from-r-cnn-to-mask-r-cnn-34ea83205de4\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "a390ea86fd8268f1b48610ff493418cb90a8b655",
  "block": 27283644,
  "trx_in_block": 20,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-31T07:30:12",
  "op": [
    "vote",
    {
      "voter": "jmcoins",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "7d31accec0ff980f91a8ed185ee08be3a65a7366",
  "block": 27065679,
  "trx_in_block": 10,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-23T17:43:33",
  "op": [
    "comment",
    {
      "parent_author": "anonymint",
      "parent_permlink": "re-djlemonskull-re-anonymint-why-social-media-software-sucks-20180927t190707766z",
      "author": "djlemonskull",
      "permlink": "re-anonymint-re-djlemonskull-re-anonymint-why-social-media-software-sucks-20181023t174332557z",
      "title": "",
      "body": "OK agree Reddit is not bad social media. But I'm thinking what we need more is a kind of <b>massive knowledge processing engine</b>, slightly described <a href=\"https://steemit.com/democracy/@djlemonskull/open-democracy-and-extinction-of-politicians \">here</a>. In short it would serve to apply kind of software production process to e.g. law creation where social network (very big community) could be involved.  It could revolutionize how people work together in my view. Ticket (problem to solve or sth to improve)->Categorization->Team (or few teams) assignment-> Cost estimation->Whitepapers->Voting for best solution->Implementation",
      "json_metadata": "{\"tags\":[\"politics\"],\"links\":[\"https://steemit.com/democracy/@djlemonskull/open-democracy-and-extinction-of-politicians \"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "61ac33bfd436759998b637486cc19317e1d5a5bd",
  "block": 27063811,
  "trx_in_block": 4,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-23T16:09:57",
  "op": [
    "comment",
    {
      "parent_author": "mattwaters.com",
      "parent_permlink": "hello-steem-friends-i-m-running-for-us-senate-pro-constitution-pro-liberty-pro-peace",
      "author": "djlemonskull",
      "permlink": "re-mattwaterscom-hello-steem-friends-i-m-running-for-us-senate-pro-constitution-pro-liberty-pro-peace-20181023t160956285z",
      "title": "",
      "body": "Hi Matt, welcome to steemit! What you think about the future of democracy? Any of these predictions may come true thanks to tools like steemit? https://steemit.com/democracy/@djlemonskull/open-democracy-and-extinction-of-politicians",
      "json_metadata": "{\"tags\":[\"virginia\"],\"links\":[\"https://steemit.com/democracy/@djlemonskull/open-democracy-and-extinction-of-politicians\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "0997f88ddc69f6d498a6ad85326006a5df8d7b5a",
  "block": 27009467,
  "trx_in_block": 16,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-21T18:50:51",
  "op": [
    "vote",
    {
      "voter": "miqyk",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "weight": 7500
    }
  ]
}

{
  "trx_id": "28d5407c1c3095fb86c5154d7380a4478c196422",
  "block": 27009421,
  "trx_in_block": 7,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-21T18:48:33",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "democracy",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "title": "Open Democracy and extinction of politicians",
      "body": "@@ -386,17 +386,16 @@\n nities w\n-h\n ere rela\n",
      "json_metadata": "{\"tags\":[\"democracy\",\"politics\",\"massive\",\"knowledge\",\"processing\"],\"image\":[\"https://assets.fnlondon.com/2016/02/IMG009915_full4x3.jpg\",\"https://cdn.steemitimages.com/DQmU74n1JUU9yAdu2hrq9DZbDAaU6Fe1BrSWvin9XzZ4NLE/dinosaur.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "c7d83caaa14bbb9e410d4f91f0941bed4ccbd449",
  "block": 27000410,
  "trx_in_block": 12,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-21T11:17:33",
  "op": [
    "vote",
    {
      "voter": "achaxyz",
      "author": "djlemonskull",
      "permlink": "does-purpose-contradicts-freedom",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "a2fd28fa6c04ee90dc9a5e13e20da2d6679086cd",
  "block": 26995322,
  "trx_in_block": 13,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-21T07:02:54",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "life",
      "author": "djlemonskull",
      "permlink": "does-purpose-contradicts-freedom",
      "title": "Does purpose contradicts freedom?",
      "body": "@@ -1414,17 +1414,15 @@\n  be \n-that it i\n+it seem\n s bi\n@@ -1537,20 +1537,16 @@\n %3C/b%3E%3Cbr%3E\n-%3Cbr%3E\n %0A%0AWhere \n@@ -1613,17 +1613,16 @@\n s to be \n-%E2%80%9C\n logic tr\n@@ -1630,17 +1630,16 @@\n nsferred\n-%E2%80%9D\n  from %E2%80%9Co\n@@ -1675,20 +1675,23 @@\n fe%E2%80%9D \n-is\n+are\n  all \n+%22\n small\n+%22\n  thi\n",
      "json_metadata": "{\"tags\":[\"life\",\"purpose\",\"freedom\",\"philosophy\"],\"image\":[\"http://timhillpsychotherapy.com/wp-content/uploads/2016/09/What-Would-You-Do-With-Your-Freedom.jpg\",\"https://i.stack.imgur.com/SSDlE.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "7bde99e4164efa0ddaace7444b43e24f7af541fe",
  "block": 26994948,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-21T06:44:12",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "life",
      "author": "djlemonskull",
      "permlink": "does-purpose-contradicts-freedom",
      "title": "Does purpose contradicts freedom?",
      "body": "<div class=\"pull-right\"><img src=\"http://timhillpsychotherapy.com/wp-content/uploads/2016/09/What-Would-You-Do-With-Your-Freedom.jpg\" width=\"500\"></div>\n\nWhen I ask myself what does it mean to have a purpose (big or small life mission) - it means to me to have a kind of feeling that I need to do something to get somewhere in the future. Do something to satisfy the purpose so that also I can feel satisfaction.<br><br>\nWhen I ask myself what does it mean to feel freedom - it is something I would call to feel complete.<br>\n\n<b>That is why I find freedom and purpose contradictory</b> <br>\n\nI need to do something when I'm not complete, so I am not totally free. And especially if there could exist any big ultimate purpose of life it would mean there is no real freedom - because until I satisfy it I cannot feel complete.<br>\nOn the opposite if we assume life has no purpose, you do not need to do anything to be complete or to satisfy some external purpose - it means you are absolutely free - because nothing really make sense, and you don’t need to satisfy anything or anybody.<br>\nIt’s sounds like a bitter freedom though since I’m used to have motivation for “my life actions” from believing in purpose. That’s why freedom is so hard actually. When people or nation get more freedom many times they get into troubles. Maybe we are not prepared for the absolute freedom? Freedom without purpose? But might be that it is bitter only because of our conditioning and what we get used to in life.<br>\n\n<b>The source of needing purpose</b><br><br>\n\nWhere does this feeling to need a purpose come from? To me is seems to be “logic transferred” from “operating life”. “Operating life” is all small things we do to support living conditions, like getting food or shelter or support in life. To get such you need to recognize a chain of actions and elements composed in a logic order. If there will be no logic order it will not get you to whatever you need to support your life. And this logic order means that every element, every action got its purpose - to prepare for the next action or fit the next element. The logic is = action is good (make sense, serves purpose) when it fits the next element. If something satisfied purpose - it is good. If it does not - it’s no good. This seems to be the source of purpose and source of what is ultimately good or bad definitions at the same time. And what we do is we get use to this so much that actually believe it’s The Truth and we transfer this logic of “life operations”, of what is good and bad, to our existence. This means we believe our life will be good if it got a purpose, because without a purpose it cannot be good.<br>\n\n<b>So what you pick? The “bitter” freedom or the life of purpose?</b><br>\nhttps://i.stack.imgur.com/SSDlE.jpg",
      "json_metadata": "{\"tags\":[\"life\",\"purpose\",\"freedom\",\"philosophy\"],\"image\":[\"http://timhillpsychotherapy.com/wp-content/uploads/2016/09/What-Would-You-Do-With-Your-Freedom.jpg\",\"https://i.stack.imgur.com/SSDlE.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "33f984e5db49216e7e2324a82c258bbfe00090e3",
  "block": 26923269,
  "trx_in_block": 43,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-18T18:58:39",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "8743eac35336ae85ed2c4ce90ed564dcc7570f7e",
  "block": 26922339,
  "trx_in_block": 14,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-18T18:12:06",
  "op": [
    "vote",
    {
      "voter": "council",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "weight": 1000
    }
  ]
}

{
  "trx_id": "fc62c56fbd0cf16f8a4b26880c817c19ff3441bf",
  "block": 26921938,
  "trx_in_block": 22,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-18T17:52:03",
  "op": [
    "vote",
    {
      "voter": "interfecto",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "weight": 400
    }
  ]
}

{
  "trx_id": "7621323eb3304e47b522570a698432357836b762",
  "block": 26921673,
  "trx_in_block": 4,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-18T17:38:48",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "democracy",
      "author": "djlemonskull",
      "permlink": "open-democracy-and-extinction-of-politicians",
      "title": "Open Democracy and extinction of politicians",
      "body": "<b>An essay about massive knowledge processing engine</b>\n<div class=\"pull-left\"><img src=\"https://assets.fnlondon.com/2016/02/IMG009915_full4x3.jpg\" width=\"500\"></div>\nThe fundamentals of democracy must change so that society moves to a more stable and more fair existence as current system does not fit the scale of society.<br><br>\nIn the far past, like Greek countries, where communities where relatively small, those communities had a big impact and very good contact with it’s politicians. In such small groups (thousands of people) democracy can work well - everybody can participate. <br><br>But it starts to be more chaotic when scale comes to the game and a nation is electing politicians they do not have contact with and cannot participate directly in democracy. Politician is someone who can have 1 good idea and 9 bad ideas, so if society has little contact with politician and cannot have direct impact, it works against society.\n<br><br>\n<b>Can democracy scale?</b>\n<br><br>\nMany things has changed since ancient times. Knowledge, science and technology upgraded a lot. The thing is they are not used to enhance the core of democracy.\n<br>\nA country is like a huge company. A company can be best described (like maybe everything in nature) as group of logic processes and elements participating in it and managing it. If processes and operators are logic and close to optimal all works good.<br><br>\nObserving nature shows that it is optimal when process is fluent, where there is not many interruptions, chaos and wrongly matched elements.<br>\n- Interruptions are when you need to change elements or operators (politicians) and break processes. For example new election comes, some processes will be cut or abandoned.<br>\n- Chaos is when bad law or bad decisions are created.<br>\n- Wrongly matched elements are politicians (people) on wrong positions.<br><br>\n\n<b>The extinctions of politicians</b>\n<div class=\"pull-left\"><img src=\"https://cdn.steemitimages.com/DQmU74n1JUU9yAdu2hrq9DZbDAaU6Fe1BrSWvin9XzZ4NLE/dinosaur.jpg\"></div>\nWe do not need politicians, nice faces with little knowledge. \n<br>\n<b>What we need is:<br>\n- Country managers, officers executing law, who can be easily changed on monthly basis if appears corrupted<br>\n- Open Democracy: ongoing law creation by experts from each branch with active masses of society</b><br>\n\nThat requires a big change, constitutional change. Hard to imagine it could happen. But possible to imagine an outline how it could work.<br>\n\n\n<b>Removing the uneducated law creators</b>\n\nWe can imagine law creation could look like something between Wikipedia creation and open source software creation.\n<br>\n\nOpen source software is designed by experts. If not-experts are making it - it will be a crap or will not work. If someone has better idea how to change it - he can improve it and create new version. This way law should be created. Many version, ongoing production, people voting to pick best solution for now. \n<br>\n\nSociety creates experts in educational process (schools, universities) -> experts create law propositions with “pros and cons”, then society votes (only those parts of society vote that the law impacts!)\n<br>\n\nIt should also have something from Wikipedia. It needs hierarchy, it requires people to have some profile, trust rank and proven knowledge to avoid spam.\n<br>\n\n<b>Tools missing</b>\n\nTo move law creation online special tools are needed for <b>massive knowledge processing</b>.<br> It’s not exactly the same as Wikipedia. It’s more like software production, where  <br>\n1. you start with a ticket (problem to solve), then <br>\n2. you categorize it with importance and branch of knowledge (e.g. medicine), impact on the system and estimate cost of such change. <br>\n3. Then it is assigned to a specific team (or many teams in parallel) to work it out (e.g. doctors and professors of medicine). <br>\n4. Then few propositions may come with different ups and downs. In the end society or part of it can vote to select a solution. <br>And all of this can be done without politician and elections breaking the whole process. It can be faster (done without elections breaking it), cheaper (less chaotic, no uneducated politicians deciding on medicine)  and more optimal (done by experts).<br><br>\n\nA decent forum engine + ticket engine + voting engine are needed. Secure and decentralized. This is still a challenge to have. And totality of it will be a social media platform, where new articles are created through <b>massive knowledge processing engine.</b><br><br>\n\n<b>The hardest part</b>\n\nHowever tools are not yet there is seems hardest would be to change constitutions and the overall country workflow.\n<br>\n\nPeople creating law would be limited to those well educated in the part they are changing (some may call it unfair, but this is same unfair like expecting person with math education to teach math at school, not geography or language).\n<br>\n\nPeople voting could be limited to the part of society that it will only impact (because it only make sense - but still someone wanting more power will call it unfair?).<br>\n\nIt’s a matter of question if voting should have some weights, like for example if I pay more taxes if my vote should be more important? OR maybe everyone voting for new country programs (actions or operations) should vote with the paid tax as weight to support his or her favorite programs?<br>\n\n<b>All this sci-fi for today... </b>",
      "json_metadata": "{\"tags\":[\"democracy\",\"politics\",\"massive\",\"knowledge\",\"processing\"],\"image\":[\"https://assets.fnlondon.com/2016/02/IMG009915_full4x3.jpg\",\"https://cdn.steemitimages.com/DQmU74n1JUU9yAdu2hrq9DZbDAaU6Fe1BrSWvin9XzZ4NLE/dinosaur.jpg\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "e8dbe35a199c5e1fe014209fd7b3ed14a0589ca1",
  "block": 26804789,
  "trx_in_block": 23,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-14T16:08:45",
  "op": [
    "vote",
    {
      "voter": "lilszi",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "7b786ee9547c3ccefe9547dc4045bc2c744d2232",
  "block": 26632270,
  "trx_in_block": 59,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-08T16:17:18",
  "op": [
    "delegate_vesting_shares",
    {
      "delegator": "steem",
      "delegatee": "djlemonskull",
      "vesting_shares": "2019.336318 VESTS"
    }
  ]
}

{
  "trx_id": "bcfda32ddb46d8833cd905a453a88b17720d3753",
  "block": 26596043,
  "trx_in_block": 16,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-07T10:05:06",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -1,8 +1,124 @@\n+%3Cb%3EIn this post I will explain why your bank account is more private then BTC and why anonymity matters.%3C/b%3E%3Cbr%3E%3Cbr%3E\n Money or\n@@ -190,19 +190,19 @@\n sferred \n-and\n+BUT\n  also ea\n@@ -203,16 +203,22 @@\n lso easy\n+ lost,\n  stolen \n@@ -363,23 +363,36 @@\n iv%3E%0A\n-%3Cbr%3E%0AWith\n+Using bank account and\n  fiat\n- \n+/\n cash\n@@ -441,16 +441,82 @@\n  society\n+ - nobody knows how much you have until you share such information\n . Some o\n@@ -668,16 +668,25 @@\n y he can\n+ (try to)\n  bribe t\n@@ -694,17 +694,35 @@\n ose guys\n-,\n+ (risking jail), to\n  know ho\n@@ -783,33 +783,21 @@\n Now \n-we have crypto asset like\n+lets consider\n  BTC\n@@ -801,20 +801,19 @@\n BTC. Is \n-that\n+BTC\n  anonymo\n@@ -821,18 +821,50 @@\n s? When \n-to\n+doing transaction on internet with\n  people \n@@ -887,14 +887,15 @@\n  is \n-pseudo\n+kind of\n  ano\n@@ -902,16 +902,63 @@\n nymous. \n+But what about doing transaction facet to face?\n %0A%3Cbr%3E%0A%3Cb\n@@ -967,18 +967,25 @@\n %0A%3Cb%3EWhy \n-is\n+anonymity\n  matters\n@@ -1171,16 +1171,30 @@\n you have\n+ left (0.9995)\n . For so\n@@ -1294,19 +1294,16 @@\n example \n-of \n why anon\n@@ -1392,24 +1392,131 @@\n here is!%3C/b%3E\n+ Notice also using credit card or fiat is anonymous in such case (merchant does not know how much you own).\n %0A%3Cbr%3E%3Cbr%3E%0A%3Cb\n@@ -1693,20 +1693,16 @@\n  allows \n-for \n privacy \n@@ -1929,16 +1929,175 @@\n %3Esafety.\n+ %3Cb%3EAs a bonus you get even stronger privacy then using online banking - even officers may not know how much you have if you do not share this information.%3C/b%3E\n %0A%3Cbr%3E%3Cbr\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\",\"links\":[\"https://twitter.com/PirateChain/\",\"https://pirate.black/\"]}"
    }
  ]
}

{
  "trx_id": "30fea6a1c7a7a3122afb3414d87b162f3399edf5",
  "block": 26567268,
  "trx_in_block": 9,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T10:05:15",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -1509,12 +1509,205 @@\n acy-%3Esafety.\n+%0A%3Cbr%3E%3Cbr%3E%0A%3Cb%3EPS%3C/b%3E %3Cbr%3E If you want to try Pirate (Arrr token) there is an official twitter account https://twitter.com/PirateChain/ %3Cbr%3E and website for more information https://pirate.black/\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\",\"links\":[\"https://twitter.com/PirateChain/\",\"https://pirate.black/\"]}"
    }
  ]
}

{
  "trx_id": "3772d6fe7f2c2fb71d7a2c007bfbd41f45726927",
  "block": 26567158,
  "trx_in_block": 2,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:59:45",
  "op": [
    "vote",
    {
      "voter": "inwang77",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "ded7bde71cbf1c4e9372bc317eeab37e237438be",
  "block": 26567002,
  "trx_in_block": 4,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:51:57",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "b6c09a47c93c68a1a002af5ac316524f663bf897",
  "block": 26566962,
  "trx_in_block": 50,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:49:57",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -954,16 +954,20 @@\n That is \n+the \n most sim\n@@ -978,16 +978,19 @@\n example \n+of \n why anon\n@@ -1032,19 +1032,19 @@\n d assets\n-. A\n+, a\n nd crypt\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "49479a3ec0ddace0a8391edda03a3a5154f32670",
  "block": 26566905,
  "trx_in_block": 15,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:47:06",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -658,11 +658,47 @@\n us. \n-But\n+%0A%3Cbr%3E%0A%3Cbr%3E%0A%3Cb%3EWhy is matters%3C/b%3E%3Cbr%3ENow\n  ima\n@@ -728,17 +728,26 @@\n ve 1 BTC\n-,\n+ and while\n  traveli\n@@ -777,13 +777,18 @@\n fee \n-using\n+for 0.0005\n  BTC\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "a21b21d2bfbe11a4efd7d5476586c18e0e26bdd3",
  "block": 26566860,
  "trx_in_block": 4,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:44:48",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "@@ -1036,16 +1036,38 @@\n br%3E%3Cbr%3E%0A\n+%3Cb%3ESolution?%3C/b%3E%0A%3Cbr%3E%0A\n Fortunat\n",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "f4b35ff821be72af2b6a7b7efdd7480d59e8a6bb",
  "block": 26566455,
  "trx_in_block": 6,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:24:33",
  "op": [
    "vote",
    {
      "voter": "ghisote",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "764598f2e15a0ba9b0c2aad85b5715206dc74631",
  "block": 26566388,
  "trx_in_block": 6,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:21:12",
  "op": [
    "vote",
    {
      "voter": "blackjok3r",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "a24bbd597ef389e4ef80ff7e6553ea168e6850d9",
  "block": 26566336,
  "trx_in_block": 6,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-06T09:18:33",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "bitcoin",
      "author": "djlemonskull",
      "permlink": "why-anonymity-in-money-either-fiat-or-crypto-asset-is-important",
      "title": "Why anonymity in money (either fiat or crypto asset) is important",
      "body": "Money or cash is a very liquid thing. It means it can be easy moved, transferred and also easy stolen or robbed.\n<div class='pull-left'>\n<img src='https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png'>\n</div>\n<br>\nWith fiat cash you are more or less anonymous to most of society. Some officers in tax department or your bankers might know your account holdings, but for most society it is hidden. Of course if someone wants strongly he can bribe those guys, know how much you have and abuse that information!\n<br>\n<br>\nNow we have crypto asset like BTC. Is that anonymous? When to people you never met it is pseudo anonymous. But imagine a scenario: you have 1 BTC, traveling you buy a cop of coffee using BTC. An evil merchant could track your transfers and possibly infer how much you have. For some people 1 BTC might be enough to threaten your life or health. <b> That is most simple example why anonymity matters in money / very liquid assets. And crypto is the most liquid asset there is!</b>\n<br><br>\nFortunately we have Zcash and other tokens from that family (KMD and assets, Hush and Pirate recently to name a few). Owning these tokens on z-address allows for privacy (and your life and health security if you understand above example!) - but only if you use the z-address. The most private of those is Pirate by far - the highest rate of z-transactions on that network. For your privacy->safety.",
      "json_metadata": "{\"tags\":[\"bitcoin\",\"cryptocurrency\",\"anonymity\",\"security\",\"safety\"],\"image\":[\"https://cdn.steemitimages.com/DQmf9mw5GXrXAitq3c1zSBEMCC6mEE9HJqoSXNn4NPTne2v/image.png\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "bba0d333dbdd87b22fd895198acbfaa346519754",
  "block": 26543477,
  "trx_in_block": 16,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-05T14:14:39",
  "op": [
    "comment",
    {
      "parent_author": "muratkbesiroglu",
      "parent_permlink": "can-the-future-be-predicted-english-turkish",
      "author": "djlemonskull",
      "permlink": "re-muratkbesiroglu-can-the-future-be-predicted-english-turkish-20181005t141440244z",
      "title": "",
      "body": "You can find patterns, relations (formulas) and trends, which also have their limitations, but if understood well and used according to correct assumptions may predict some future events. But it's not \"predicting the future\", more like giving probability of an event in the future.",
      "json_metadata": "{\"tags\":[\"life\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "7b25b41157ca91c2c74d11a07678f3c47ec9ec07",
  "block": 26476994,
  "trx_in_block": 19,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-03T06:48:12",
  "op": [
    "vote",
    {
      "voter": "anonymint",
      "author": "djlemonskull",
      "permlink": "re-anonymint-re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20181002t182312379z",
      "weight": 1000
    }
  ]
}

{
  "trx_id": "76b655aa3841ecd7209deae1cc7d360baa24d15a",
  "block": 26462254,
  "trx_in_block": 31,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-02T18:31:03",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "steemit",
      "author": "djlemonskull",
      "permlink": "true-activity-of-steem-network-someone-turned-off-bots",
      "title": "True activity of steem network (someone turned off bots?)",
      "body": "It used to be above 1 million, now it is about 100 000:\n\n![](https://cdn.steemitimages.com/DQmcJv8EC1eC5ZAMyq31whWr13Mp7cLexGe4GuKNnFRyV3S/image.png)\n\n\nSnapshot comes from http://blocktivity.info/ taken at October 2nd 8pm UTC.",
      "json_metadata": "{\"tags\":[\"steemit\",\"steem\"],\"image\":[\"https://cdn.steemitimages.com/DQmcJv8EC1eC5ZAMyq31whWr13Mp7cLexGe4GuKNnFRyV3S/image.png\"],\"links\":[\"http://blocktivity.info/\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "64372944ab6f054a893aa5cc292d6a0bdd71f68e",
  "block": 26462098,
  "trx_in_block": 2,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-02T18:23:15",
  "op": [
    "comment",
    {
      "parent_author": "anonymint",
      "parent_permlink": "re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20180420t060218449z",
      "author": "djlemonskull",
      "permlink": "re-anonymint-re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20181002t182312379z",
      "title": "",
      "body": "Seems someone turned off steem bots today :) https://imgur.com/a/hf4c9BV Naked truth about activity?",
      "json_metadata": "{\"tags\":[\"dlive\"],\"links\":[\"https://imgur.com/a/hf4c9BV\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "50bebd6cffda8e955bfb03f9e1cc9bc634b0a5ae",
  "block": 26425466,
  "trx_in_block": 21,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-10-01T11:50:54",
  "op": [
    "vote",
    {
      "voter": "anonymint",
      "author": "djlemonskull",
      "permlink": "re-anonymint-re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20180929t201627327z",
      "weight": 1000
    }
  ]
}

{
  "trx_id": "c8967c157f547f59221d1a508d12b42d4acdb973",
  "block": 26401032,
  "trx_in_block": 31,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-30T15:28:18",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "btc",
      "author": "djlemonskull",
      "permlink": "kys-before-teaching-your-ai",
      "title": "KYS before teaching your AI",
      "body": "@@ -4458,36 +4458,51 @@\n le style=%22width:\n-100%25\n+70%25; font-size:10px\n %22%3E%0A%3Ctr%3E%0A%3Cth%3EPerc\n",
      "json_metadata": "{\"tags\":[\"btc\",\"cryptocurrency\",\"bigdata\",\"analysis\",\"bitcoin\"],\"image\":[\"https://cdn.steemitimages.com/DQmR1odNnLaG2H3P8sikxyToKBxxic6RaxkQcJUyZ5151bN/ai.PNG\",\"https://cdn.steemitimages.com/DQmPeQ3Czb4dmBA1hEzDhVQpYTGKz4jnvViAGto5cVU7nff/overlaps.PNG\",\"https://cdn.steemitimages.com/DQmbR4aVhfNC8RZcyate5UbiTRHTjZzKzUYid8k9ma8rRwU/percentiles_chartPNG.PNG\"],\"links\":[\"https://en.wikipedia.org/wiki/Tay_\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "10d73ab46500ba001f4b334db488c5fd0e0e0562",
  "block": 26378015,
  "trx_in_block": 10,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-29T20:16:27",
  "op": [
    "comment",
    {
      "parent_author": "anonymint",
      "parent_permlink": "re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20180420t060218449z",
      "author": "djlemonskull",
      "permlink": "re-anonymint-re-jerrybanfield-3ffc3520-426a-11e8-836a-b9befc1a6029-20180929t201627327z",
      "title": "",
      "body": "Even today we can see few things:<br>\n1.  100k twitter followers for steem which is probably a closer estimation of real usage.<br>\n2. Surprisingly small activity on steem reddit<br>\n3. Recently after Hard Fork hitting users really hard (low users activity) the blocktivity was showing still very big steem usage. A clue it is fake...?",
      "json_metadata": "{\"tags\":[\"dlive\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "a8e551d52419b25579fe9dc82c20bb1531fb61a6",
  "block": 26346638,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-28T18:05:15",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "kys-before-teaching-your-ai",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "72ffc329395a43e89575ad3ce7c8fef1a41ca4e8",
  "block": 26339395,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-28T12:02:21",
  "op": [
    "vote",
    {
      "voter": "stretcher83",
      "author": "djlemonskull",
      "permlink": "kys-before-teaching-your-ai",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "1045e013796a7ae650ff5eadec37925cb86de2a5",
  "block": 26338754,
  "trx_in_block": 0,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-28T11:30:12",
  "op": [
    "custom_json",
    {
      "required_auths": [],
      "required_posting_auths": [
        "djlemonskull"
      ],
      "id": "follow",
      "json": "[\"follow\",{\"follower\":\"djlemonskull\",\"following\":\"cryptocurator\",\"what\":[\"blog\"]}]"
    }
  ]
}

{
  "trx_id": "b93eab44d075c3b7396d374e0ca8bfe2e620b747",
  "block": 26338567,
  "trx_in_block": 2,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-28T11:20:51",
  "op": [
    "vote",
    {
      "voter": "actionslacker",
      "author": "djlemonskull",
      "permlink": "kys-before-teaching-your-ai",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "2c2b1b7efc0d1d38e3c7978fa5b52ee215de971c",
  "block": 26338501,
  "trx_in_block": 1,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-28T11:17:33",
  "op": [
    "comment",
    {
      "parent_author": "",
      "parent_permlink": "btc",
      "author": "djlemonskull",
      "permlink": "kys-before-teaching-your-ai",
      "title": "KYS before teaching your AI",
      "body": "<b>In this article you will learn how many opportunities there were in the past 12 months to make profit on BTC and how many opportunities to avoid loss :)</b>\n<div class=\"pull-left\"><img src=\"https://cdn.steemitimages.com/DQmR1odNnLaG2H3P8sikxyToKBxxic6RaxkQcJUyZ5151bN/ai.PNG\" /></div>\nKnow Your Signal before teaching your Neural Network. Your are the teacher for AI, so you need to know the subject you are teaching, right?<br>\nAI is the student here. What is AI? It tends to be shown in a context where AI = Super Intelligence. Well - it’s not. It’s what the name says - it’s Artificial Intelligence, it is mechanical, a learning engine. Engines are always amazing tools. It can be super when applied correctly in some context, but only if someone will use the tool (=teach) correctly. You might have heard about Microsoft racist AI? https://en.wikipedia.org/wiki/Tay_(bot)<br>\n<br>\nSo before I start learning neural network, I need to do my own lesson and learn a bit about my time series also TO UNDERSTAND WHAT I’M REALLY SEARCHING AND PLANNING TO DO.<br>\n<br>\nSignal is a time series - a parameter that changes in time. My goal here is to find if it can be predicted and used to generate profit. There are numbers of hidden assumptions in the previous statement:<br>\n<br>1. <b>Given time series value can be predicted - I assume it’s true, but is it?</b> Here we can go even deeper and ask what is true? In science often there is no 100% truth, just probability of events. Here it is good enough. If the time series can be predicted in 60% of cases this could be good enough to generate profit.<br>\n<br>2. <b>What is profit?</b> Time series changes all the time, up and down. By profit we could say 4% growth or 10% growth. But it’s not precise at all until you set the time frame. Is 4% in a year good enough? For sure not for me. So maybe 4% in a week? But why in a week not in 5 days? Why 4% , not 9%? - A VALID QUESTIONS TO ANSWER.<br>\n<br>\nSo I have found that I do not know something basic - what kind of profit is important for me and what kind of profit I could get.<br>\nName it short - what is the dynamic of significant changes in the signal. How often does it go up and down during period of 1 year. Dynamics is not only how often it changes significantly, but also how rapid is the change.<br>\n<br>\nMy next step is to find how often price changes significantly and how rapid changes are. Since I need to define significant growth or decrease I will take some nice numbers like 3, 5, 8, 13, 21.<br>\n<br>\nAfter I get the answer I will not only know what I should search next, but also how to mark my signal to train the neural network later. There is many ways to teach, not many good ways though.<br>\n<br>\nSo I did a quick script to answer my question and found one more important thing to precise - it’s better to search for unique significant change, as these changes may overlap, as you can see on the picture below: \n<div class=\"pull-left\"><img src=\"https://cdn.steemitimages.com/DQmPeQ3Czb4dmBA1hEzDhVQpYTGKz4jnvViAGto5cVU7nff/overlaps.PNG\" /></div>\n<br><br>To have unique changes I decided to take the shorter of overlapping two while moving forward.\n<br>\nModifying script to make it efficiently took me few days. I’m new to python; initial script was taking a very long time (like a week for only 3%  :) to find unique changes using my simple algorithm. Trying to optimize it and improve my python skills took few days, after which it went down to 6 hours for 3% growth (still not amazing but good enough for me) and overall it took a week to have all numbers up and down. \n<br>\n<br>So here is the result: \n<div class=\"pull-left\"><img src=\"https://cdn.steemitimages.com/DQmbR4aVhfNC8RZcyate5UbiTRHTjZzKzUYid8k9ma8rRwU/percentiles_chartPNG.PNG\" /></div>\n<br>This set of chart shows how rapid is change split by percentiles to have distribution. Red are drops, green - ups.  The outcome is dynamics of growth and decrease are similar in order of magnitude with acceptable differences and I may decide to use 90 percentile change time for my AI later.<br>\n<br>The same result may be show also as table. For some people it may be confusing as there are 2 kinds of % values - to indicate growth type and to indicate how many results match specific dynamics. The table basically shows how much time you need to wait to have for example 5% growth. In 10% cases it will happen just in 4.4 minutes. Median for 5% is 52.6 minutes.\n<br>\n<table style=\"width:100%\">\n<tr>\n<th>Percentile</th><th>+3%</th><th>+5%</th><th>+8%</th><th>+13%</th><th>+21%</th><th>-3%</th><th>-5%</th><th>-8%</th><th>-13%</th><th>-21%</th>\n</tr>\n<tr>\n<td>10</td><td>1.6</td><td>4.4</td><td>12.5</td><td>42.7</td><td>202.9</td><td>1.9</td><td>5.7</td><td>17.6</td><td>52.9</td><td>181.7</td>\n</tr>\n<tr>\n<td>20</td><td>3.4</td><td>10.7</td><td>34.1</td><td>88.2</td><td>358.7</td><td>3.8</td><td>12.3</td><td>32.8</td><td>92.8</td><td>520.4</td>\n</tr>\n<tr>\n<td>30</td><td>6.4</td><td>20.9</td><td>70.7</td><td>175.6</td><td>961.9</td><td>6.5</td><td>20.0</td><td>53.8</td><td>189.0</td><td>867.4</td>\n</tr>\n<tr>\n<td>40</td><td>11.1</td><td>36.2</td><td>106.0</td><td>267.6</td><td>1429.5</td><td>10.3</td><td>28.8</td><td>78.6</td><td>298.0</td><td>923.0</td>\n</tr>\n<tr>\n<td>50</td><td>16.8</td><td>52.6</td><td>163.5</td><td>484.2</td><td>1502.1</td><td>15.3</td><td>45.4</td><td>125.5</td><td>591.3</td><td>1154.8</td>\n</tr>\n<tr>\n<td>60</td><td>27.9</td><td>85.2</td><td>266.6</td><td>692.7</td><td>1904.6</td><td>23.2</td><td>66.5</td><td>179.6</td><td>731.3</td><td>1582.0</td>\n</tr>\n<tr>\n<td>70</td><td>44.9</td><td>144.5</td><td>469.5</td><td>1232.4</td><td>2871.2</td><td>35.1</td><td>105.3</td><td>281.0</td><td>1053.5</td><td>2052.6</td>\n</tr>\n<tr>\n<td>80</td><td>74.4</td><td>235.3</td><td>821.0</td><td>1858.4</td><td>3571.4</td><td>52.4</td><td>167.4</td><td>426.6</td><td>1314.8</td><td>2924.0</td>\n</tr>\n<tr>\n<td>90</td><td>153.5</td><td>482.8</td><td>1297.0</td><td>3788.3</td><td>6586.2</td><td>99.8</td><td>326.1</td><td>1081.0</td><td>2727.3</td><td>6411.1</td>\n</tr>\n<tr>\n<td>100</td><td>1974.1</td><td>2246.0</td><td>4382.2</td><td>10301.1</td><td>9068.4</td><td>1094.2</td><td>4107.9</td><td>5492.0</td><td>4704.8</td><td>11202.1</td>\n</tr>\n</table>\n\nAlso I see estimation of how many chances there are to make profit or to avoid loss.<br>\n<table style=\"width:100%\">\n<tr>\n<th>+3%</th><th>+5%</th><th>+8%</th><th>+13%</th><th>+21%</th><th>-3%</th><th>-5%</th><th>-8%</th><th>-13%</th><th>-21%</th>\n</tr>\n<tr>\n<td>1556</td><td>593</td><td>230</td><td>102</td><td>34</td><td>1390</td><td>513</td><td>175</td><td>65</td><td>24</td>\n</tr>\n</table>\nIt means for example there were 1556 spots of unique growth of 3% last 12 months. Another example there were unique 24 drops of 21% to avoid.\n<br>\n<br>\nSeems that make sense to move forward.<br>",
      "json_metadata": "{\"tags\":[\"btc\",\"cryptocurrency\",\"bigdata\",\"analysis\"],\"image\":[\"https://cdn.steemitimages.com/DQmR1odNnLaG2H3P8sikxyToKBxxic6RaxkQcJUyZ5151bN/ai.PNG\",\"https://cdn.steemitimages.com/DQmPeQ3Czb4dmBA1hEzDhVQpYTGKz4jnvViAGto5cVU7nff/overlaps.PNG\",\"https://cdn.steemitimages.com/DQmbR4aVhfNC8RZcyate5UbiTRHTjZzKzUYid8k9ma8rRwU/percentiles_chartPNG.PNG\"],\"links\":[\"https://en.wikipedia.org/wiki/Tay_\"],\"app\":\"steemit/0.1\",\"format\":\"markdown\"}"
    }
  ]
}

{
  "trx_id": "c71e0d3bc7d9d95f0c5081bf51cdf78555829105",
  "block": 26319196,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-27T19:11:24",
  "op": [
    "comment",
    {
      "parent_author": "djlemonskull",
      "parent_permlink": "re-anonymint-why-social-media-software-sucks-20180923t174809317z",
      "author": "anonymint",
      "permlink": "re-djlemonskull-re-anonymint-why-social-media-software-sucks-20180927t190707766z",
      "title": "",
      "body": "There’s Reddit. But yeah we need it on a decentralized ledger so censorship is impossible. I’m [working on this](https://steemit.com/cryptocurrencies/@anonymint/re-quillfiller-re-anonymint-re-phelim-re-anonymint-re-anonymint-name-your-decentralized-social-network-20180919t061859684z).",
      "json_metadata": "{\"tags\":[\"politics\"],\"links\":[\"https://steemit.com/cryptocurrencies/@anonymint/re-quillfiller-re-anonymint-re-phelim-re-anonymint-re-anonymint-name-your-decentralized-social-network-20180919t061859684z\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "5a2df5ed0dc967a563cd6926da34a4f18b7f230e",
  "block": 26319141,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-27T19:08:39",
  "op": [
    "vote",
    {
      "voter": "anonymint",
      "author": "djlemonskull",
      "permlink": "re-anonymint-why-social-media-software-sucks-20180923t174809317z",
      "weight": 1000
    }
  ]
}

{
  "trx_id": "a8429a88c65d990aea3d1dda0ac01ba6ad3b911b",
  "block": 26290286,
  "trx_in_block": 1,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-26T19:05:00",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "rok-sivante",
      "permlink": "hard-fork-up-the",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "564bc2ab1980d25c13a284d24dca4e375a989833",
  "block": 26286923,
  "trx_in_block": 11,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-26T16:16:45",
  "op": [
    "custom_json",
    {
      "required_auths": [],
      "required_posting_auths": [
        "djlemonskull"
      ],
      "id": "follow",
      "json": "[\"follow\",{\"follower\":\"djlemonskull\",\"following\":\"komodoplatform\",\"what\":[\"blog\"]}]"
    }
  ]
}

{
  "trx_id": "6453ace03db1ff98bdc381b5596fb69490b62564",
  "block": 26286921,
  "trx_in_block": 5,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-26T16:16:39",
  "op": [
    "custom_json",
    {
      "required_auths": [],
      "required_posting_auths": [
        "djlemonskull"
      ],
      "id": "follow",
      "json": "[\"follow\",{\"follower\":\"djlemonskull\",\"following\":\"jl777\",\"what\":[\"blog\"]}]"
    }
  ]
}

{
  "trx_id": "4ead8eda7444a5d973cbb1eefa13d60ee44a0f49",
  "block": 26286914,
  "trx_in_block": 10,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-26T16:16:18",
  "op": [
    "custom_json",
    {
      "required_auths": [],
      "required_posting_auths": [
        "djlemonskull"
      ],
      "id": "follow",
      "json": "[\"follow\",{\"follower\":\"djlemonskull\",\"following\":\"soldat\",\"what\":[\"blog\"]}]"
    }
  ]
}

{
  "trx_id": "322da0d2e2ec3339ab88398dfb4082cfaab22f6c",
  "block": 26286645,
  "trx_in_block": 10,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-26T16:02:51",
  "op": [
    "transfer_to_vesting",
    {
      "from": "djlemonskull",
      "to": "djlemonskull",
      "amount": "13.986 STEEM"
    }
  ]
}

{
  "trx_id": "5bba14e9c39e60dc16c0beda262c6d4727c92947",
  "block": 26260402,
  "trx_in_block": 0,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-25T18:04:12",
  "op": [
    "transfer",
    {
      "from": "blocktrades",
      "to": "djlemonskull",
      "amount": "13.986 STEEM",
      "memo": ""
    }
  ]
}

{
  "trx_id": "838f992015144d11b6839a3dbf2c7f15c8a116a3",
  "block": 26246740,
  "trx_in_block": 20,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-25T06:39:33",
  "op": [
    "comment",
    {
      "parent_author": "dv8silencersteem",
      "parent_permlink": "beautiful-saturn",
      "author": "djlemonskull",
      "permlink": "re-dv8silencersteem-beautiful-saturn-20180925t063935281z",
      "title": "",
      "body": "This is pretty awsome - seems like a symetric mege-structure hidden in the clouds of Saturn...",
      "json_metadata": "{\"tags\":[\"space\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "4896ec3b60bdee6724cca7e5b288be6eaae9f9a4",
  "block": 26229825,
  "trx_in_block": 17,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T16:32:12",
  "op": [
    "comment",
    {
      "parent_author": "djlemonskull",
      "parent_permlink": "re-tradealert-re-wildtrader-will-eos-kill-ethereum-part-i-20180924t152222097z",
      "author": "tradealert",
      "permlink": "re-djlemonskull-re-tradealert-re-wildtrader-will-eos-kill-ethereum-part-i-20180924t163210860z",
      "title": "",
      "body": "No mate I am afraid you got me wrong, what I meant is in the long run both EOS and ETH will be equally scalable but right now Ethereum has a huge head-start in terms of development where EOS  (and all the other smart contracts platforms ) is still very barebone. \n\n> In fact there is already a race among many platform wanting to take over dapps, some of which already solved problems Eth didn't (scalability+privacy+having DEX)\n\nI don't agree with this, Ethereum has layer 2 solutions for scalability, many DEXes (0x, IDEX, Airswap, Bancor, Kyber) and a privacy roadmap with zk-snark.\n\nAgain, given time Ethereum and EOS will both have the same features in terms of scalability, privacy, exchanges etc, but Ethereum will be decentralized and EOS controlled by block producers (look what DPOS did to LISK).",
      "json_metadata": "{\"tags\":[\"cryptocurrency\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "18e6222ea29dce9b173c123e1e4757b5ecdb3a7d",
  "block": 26228947,
  "trx_in_block": 31,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:48:06",
  "op": [
    "comment",
    {
      "parent_author": "djlemonskull",
      "parent_permlink": "re-honarparvar-bitcoin-price-prediction-by-the-end-of-2018-20180924t153313685z",
      "author": "honarparvar",
      "permlink": "re-djlemonskull-re-honarparvar-bitcoin-price-prediction-by-the-end-of-2018-20180924t154806686z",
      "title": "",
      "body": "you are right, not all predictions are correct. \nI draw lines for trends only, for analysis I use RSI, Bollinger Bands, Fibonacci levels (especially the 0.618) Moving averages, elliot waves etc",
      "json_metadata": "{\"tags\":[\"bitcoin\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "d225352c7fa7bedb00be3e5e55677ecacf78666a",
  "block": 26228699,
  "trx_in_block": 6,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:35:42",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "getting-the-data-and-select-coding-language",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "36cc862ef057446f6538a73c100fd4a4bb58e43e",
  "block": 26228696,
  "trx_in_block": 44,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:35:33",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "re-anonymint-why-social-media-software-sucks-20180923t174809317z",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "535b544645da00cf7b4f7c77664686f09b092866",
  "block": 26228695,
  "trx_in_block": 15,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:35:30",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "re-honarparvar-bitcoin-price-prediction-by-the-end-of-2018-20180924t153313685z",
      "weight": 10000
    }
  ]
}

{
  "trx_id": "2cedb68488567ace2d49426be3b50dc5828b41ff",
  "block": 26228650,
  "trx_in_block": 1,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:33:15",
  "op": [
    "comment",
    {
      "parent_author": "honarparvar",
      "parent_permlink": "bitcoin-price-prediction-by-the-end-of-2018",
      "author": "djlemonskull",
      "permlink": "re-honarparvar-bitcoin-price-prediction-by-the-end-of-2018-20180924t153313685z",
      "title": "",
      "body": "I hope your predictions will be right, however I somehow believe tech analysis is rather good to interpret the past (so after the change really happened) :P and the lines seem more and art then sci to me (not your lines - tech analysis lines in general). \nPS did you ever check if tech analysis make sense at all this way: take random piece of BTC prices from the past, draw your lines (seeing only that historical part), and then uncover what was there later and if those lines match? And experiment like this? :) I wonder what would be a % of correct predictions...",
      "json_metadata": "{\"tags\":[\"bitcoin\"],\"app\":\"steemit/0.1\"}"
    }
  ]
}

{
  "trx_id": "274b4421c4408650b3965b2402d23f1c2b17a92a",
  "block": 26228440,
  "trx_in_block": 8,
  "op_in_trx": 0,
  "virtual_op": 0,
  "timestamp": "2018-09-24T15:22:45",
  "op": [
    "vote",
    {
      "voter": "djlemonskull",
      "author": "djlemonskull",
      "permlink": "re-tradealert-re-wildtrader-will-eos-kill-ethereum-part-i-20180924t152222097z",
      "weight": 10000
    }
  ]
}