1 00:00:00,000 --> 00:00:14,320 And it used to start at 10 o'clock, but now they start at 930. 2 00:00:14,320 --> 00:00:18,360 Well that's good. 3 00:00:18,360 --> 00:00:25,060 There's also a letter engage seminar at 930 today and I was going to share that with the 4 00:00:25,060 --> 00:00:30,720 group and I was going to conclude around 925 today. 5 00:00:30,720 --> 00:00:35,920 That way people can make their way to their next event. 6 00:00:35,920 --> 00:00:37,960 Okay, great. 7 00:00:37,960 --> 00:00:38,960 Awesome. 8 00:00:38,960 --> 00:00:43,360 So we'll just give it about five minutes for people to dig in here. 9 00:00:43,360 --> 00:00:55,040 But I saw your GitHub repository where you've started to work on some Python, Jupyter work 10 00:00:55,040 --> 00:00:58,200 books. 11 00:00:58,200 --> 00:01:03,400 Would you be happy to talk a bit about the work you've done there? 12 00:01:03,400 --> 00:01:09,280 Yeah, actually I've added a bunch of starter notebooks now. 13 00:01:09,280 --> 00:01:11,080 Welcome Jake. 14 00:01:11,080 --> 00:01:13,520 Hi Jake. 15 00:01:13,520 --> 00:01:17,120 Hello. 16 00:01:17,120 --> 00:01:21,720 So I did figure out how to the inventories and the plants. 17 00:01:21,720 --> 00:01:32,920 I was able to read those in to pandas by just specifying the valid columns and not reading 18 00:01:32,920 --> 00:01:34,440 in the invalid columns. 19 00:01:34,440 --> 00:01:35,800 And so that works now. 20 00:01:35,800 --> 00:01:38,180 That was a sticking point. 21 00:01:38,180 --> 00:01:42,800 And so I put those up in the GitHub repository. 22 00:01:42,800 --> 00:01:50,800 And then did you maybe see like what are like the common inventory types and the not so 23 00:01:50,800 --> 00:01:54,880 common ones by chance? 24 00:01:54,880 --> 00:02:00,360 I don't remember that right now. 25 00:02:00,360 --> 00:02:03,680 You can download that workbook and look. 26 00:02:03,680 --> 00:02:08,640 You can maybe dig into that next time. 27 00:02:08,640 --> 00:02:10,520 But you're just in this opening bit. 28 00:02:10,520 --> 00:02:15,200 I'm going to go ahead and share some links with you here just while we're waiting for 29 00:02:15,200 --> 00:02:16,880 some people. 30 00:02:16,880 --> 00:02:22,680 So welcome Nick. 31 00:02:22,680 --> 00:02:27,760 Hi Nick. 32 00:02:27,760 --> 00:02:34,560 Let's see here. 33 00:02:34,560 --> 00:02:46,880 So David Busby over at OpenTHC put together this SQLite file for business to consumer 34 00:02:46,880 --> 00:02:47,880 sales. 35 00:02:47,880 --> 00:02:57,800 So if you were having trouble working with the large sales file, this is essentially 36 00:02:57,800 --> 00:03:05,320 the sales file in a large SQL database if that's easier for you to work with. 37 00:03:05,320 --> 00:03:09,680 And so I just received that in the past couple of days. 38 00:03:09,680 --> 00:03:12,600 So I'm going to be taking a look at it myself. 39 00:03:12,600 --> 00:03:20,160 And perhaps next week we can do some work with SQL taking some poke at the sales data. 40 00:03:20,160 --> 00:03:24,440 But just keep in mind, was there a question or something? 41 00:03:24,440 --> 00:03:30,120 I was wondering if we could use Dask to deal with those really large files. 42 00:03:30,120 --> 00:03:34,040 Can you tell us about Dask please? 43 00:03:34,040 --> 00:03:37,280 Dask sits. 44 00:03:37,280 --> 00:03:42,360 Dask allows you to multitask, like multi-thread things. 45 00:03:42,360 --> 00:03:48,640 And then also it reads pandas files, but it doesn't, it chunks them up. 46 00:03:48,640 --> 00:03:54,840 And if there's not enough room in memory, it stores them on disk and so it saves them 47 00:03:54,840 --> 00:03:57,320 like in pages. 48 00:03:57,320 --> 00:04:00,160 But you don't, and then you can just use them. 49 00:04:00,160 --> 00:04:07,880 It integrates with pandas, scikit-learn, most python. 50 00:04:07,880 --> 00:04:10,400 And it just spelled Dask? 51 00:04:10,400 --> 00:04:11,400 Yeah. 52 00:04:11,400 --> 00:04:15,060 Well, it's definitely worth checking out. 53 00:04:15,060 --> 00:04:19,240 So I'll put that on the agenda for next week. 54 00:04:19,240 --> 00:04:26,600 So would that be useful for the raw files or? 55 00:04:26,600 --> 00:04:29,760 I haven't used it that much. 56 00:04:29,760 --> 00:04:34,720 I know it works well with pandas and scikit-learn. 57 00:04:34,720 --> 00:04:39,840 And also if you have multiple machines, you can run Dask on each one of the machines and 58 00:04:39,840 --> 00:04:47,040 it'll use the memory and processor on each one of the machines. 59 00:04:47,040 --> 00:04:51,400 And that's, you may need a nice setup if you're working with this data. 60 00:04:51,400 --> 00:04:58,880 So like David was telling me, he's basically, he's running this locally on his server. 61 00:04:58,880 --> 00:05:07,360 So he's just got this giant SQL database just sitting there locally. 62 00:05:07,360 --> 00:05:12,120 So it'd be kind of expensive to put this in the cloud. 63 00:05:12,120 --> 00:05:24,680 But actually, I want to check out the SQL because there's a lot of things I feel like 64 00:05:24,680 --> 00:05:30,000 you could do, a lot of queries and that with the SQL would be great. 65 00:05:30,000 --> 00:05:31,000 Exactly. 66 00:05:31,000 --> 00:05:37,040 So we can get our hands into that for next week. 67 00:05:37,040 --> 00:05:42,040 And then I guess just go ahead and kick it off here. 68 00:05:42,040 --> 00:05:47,080 Just for the people who are new here or who just got here in the past like five minutes, 69 00:05:47,080 --> 00:05:53,320 I think we're going to go to about 9.25 today because Charles is heading to an event. 70 00:05:53,320 --> 00:05:58,600 And then I'm going to be listening into this Flutter and Gauge at 9.30. 71 00:05:58,600 --> 00:06:05,980 I'm not sure if any of you use Flutter, but it's sort of a fun tool. 72 00:06:05,980 --> 00:06:10,760 So it's going to be interesting to see. 73 00:06:10,760 --> 00:06:17,920 I would like to see that get pretty popular because it's still pretty new. 74 00:06:17,920 --> 00:06:22,640 Anywho, just sharing that. 75 00:06:22,640 --> 00:06:31,520 All right. 76 00:06:31,520 --> 00:06:36,960 So I guess I can just go ahead and share my screen and just show you some of the things 77 00:06:36,960 --> 00:06:50,760 I'm doing and then we can talk about it along the way if that sounds good to you. 78 00:06:50,760 --> 00:06:58,560 Great. 79 00:06:58,560 --> 00:07:04,280 So essentially we were starting to look at the different sample types last week. 80 00:07:04,280 --> 00:07:12,800 And there's probably going to be different types of cannabinoids for the different sample 81 00:07:12,800 --> 00:07:13,800 types. 82 00:07:13,800 --> 00:07:17,960 So oils, of course, are going to have a higher distribution than flour. 83 00:07:17,960 --> 00:07:23,600 So I just thought this would be an interesting thing to look at. 84 00:07:23,600 --> 00:07:36,760 So over in Spider, or you could use your favorite text editor and your favorite programming 85 00:07:36,760 --> 00:07:37,760 language. 86 00:07:37,760 --> 00:07:48,600 But once again, I'm just working with the same LCB data dump just because the variables 87 00:07:48,600 --> 00:07:50,000 are relatively the same. 88 00:07:50,000 --> 00:07:58,760 So I figured if I can master it for one month, then it should be fairly easy to generalize 89 00:07:58,760 --> 00:08:03,920 to the full data set. 90 00:08:03,920 --> 00:08:19,760 So just go ahead and I'm not sure if this is actually going to be in the right location. 91 00:08:19,760 --> 00:08:42,400 So never mind this. 92 00:08:42,400 --> 00:08:57,600 Okay. 93 00:08:57,600 --> 00:09:04,240 So essentially we're just going to read this data in here and we're going to start looking 94 00:09:04,240 --> 00:09:09,320 at some of these sample types that we looked at last week. 95 00:09:09,320 --> 00:09:21,640 So real quick here, I'm actually just going to go ahead and just commit everything over 96 00:09:21,640 --> 00:09:33,200 to GitHub just so that everybody can follow along. 97 00:09:33,200 --> 00:09:53,040 And then just so everybody can follow along with all the material. 98 00:09:53,040 --> 00:10:10,160 You know, I'm trying to keep most things here in the cannabis data science repository. 99 00:10:10,160 --> 00:10:16,320 So that'll need to get cleaned up a little bit, but so far just sort of archiving things 100 00:10:16,320 --> 00:10:20,400 there. 101 00:10:20,400 --> 00:10:31,240 And then, okay, so we've got our data read in. 102 00:10:31,240 --> 00:10:40,000 And so the tool that I'll be using today is just Seaborn, which is sort of an extension 103 00:10:40,000 --> 00:10:41,720 of Matplotlib. 104 00:10:41,720 --> 00:10:53,320 So Seaborn requires Matplotlib, so you can see it more of an extension or an enhancement. 105 00:10:53,320 --> 00:10:57,040 We'll go ahead and grab those packages. 106 00:10:57,040 --> 00:11:10,120 And essentially, you know, let's find out what columns we even have here. 107 00:11:10,120 --> 00:11:21,360 So it's sort of a handful, but essentially, we want to look at some of these cannabinoids 108 00:11:21,360 --> 00:11:25,780 that may be of interest to people. 109 00:11:25,780 --> 00:11:33,120 So we could maybe create a total cannabinoid calculation, but at the moment, let's just 110 00:11:33,120 --> 00:11:44,440 pick a, you know, this first cannabinoid here, THCA. 111 00:11:44,440 --> 00:11:49,120 And so, you know, in the lab, you could call this an analyte. 112 00:11:49,120 --> 00:11:55,480 So you could say your analyte is THCA. 113 00:11:55,480 --> 00:12:06,160 And so, you know, now let's just basically look at the distribution of THCA in the lab 114 00:12:06,160 --> 00:12:07,160 data. 115 00:12:07,160 --> 00:12:10,880 And let's see what happens. 116 00:12:10,880 --> 00:12:20,360 Okay, so that's what happened. 117 00:12:20,360 --> 00:12:35,640 I think we want... 118 00:12:35,640 --> 00:13:03,040 So we'll just look at the kernel density of this analyte. 119 00:13:03,040 --> 00:13:24,440 So this may take a minute since there's a lot of data. 120 00:13:24,440 --> 00:13:47,080 And essentially, you know, you can add the hue, which is the actual variable type. 121 00:13:47,080 --> 00:13:58,000 So the actual, you know, inventory type, remember, is the intermediate type. 122 00:13:58,000 --> 00:14:10,040 So we'll eventually check out the hue with intermediate type. 123 00:14:10,040 --> 00:14:22,760 So we may have to get just a subset of our data here. 124 00:14:22,760 --> 00:14:45,400 I wonder if we can just do a sample of it. 125 00:14:45,400 --> 00:14:58,120 Okay, so now we've just got a random sample of the data. 126 00:14:58,120 --> 00:15:17,280 So now maybe it'll be a bit more approachable. 127 00:15:17,280 --> 00:15:20,320 So we've at least got a chart now. 128 00:15:20,320 --> 00:15:27,440 And so this is just an estimated distribution of THC. 129 00:15:27,440 --> 00:15:37,760 So not the most informative chart ever. 130 00:15:37,760 --> 00:15:41,640 It looks like there's a grouping lower down towards zero. 131 00:15:41,640 --> 00:15:48,600 You know, perhaps that's edibles or things of that sort, or miscoding, or it's hard to 132 00:15:48,600 --> 00:15:50,120 say at this point. 133 00:15:50,120 --> 00:15:52,800 And then there's a grouping up here. 134 00:15:52,800 --> 00:15:57,480 And you can see there's a slight bump up here, which must be the oils. 135 00:15:57,480 --> 00:16:06,080 But you really need to separate this by the actual intermediate type for it to be informative. 136 00:16:06,080 --> 00:16:16,080 So let's look at how we can adapt this plot here. 137 00:16:16,080 --> 00:16:45,720 I'd really like to use this plot, but it looks like my Seaborn version may be outdated. 138 00:16:45,720 --> 00:17:05,760 You could just add hue. 139 00:17:05,760 --> 00:17:09,400 So it looks like my Seaborn version may be outdated. 140 00:17:09,400 --> 00:17:15,280 So we may have to just come back to these at another point. 141 00:17:15,280 --> 00:17:21,800 However, we can maybe now change gears into something else interesting that I was going 142 00:17:21,800 --> 00:17:23,840 to show you today. 143 00:17:23,840 --> 00:17:34,120 So we've been talking about mocking up the LEAF API. 144 00:17:34,120 --> 00:17:42,920 And I thought, OK, the first step is we're going to need a database just to store some 145 00:17:42,920 --> 00:17:43,920 things. 146 00:17:43,920 --> 00:17:50,360 So I just thought, OK, we'll just go ahead and use the database that essentially I'm 147 00:17:50,360 --> 00:17:58,760 already using for some of this open source material that I'm putting out through Candletics. 148 00:17:58,760 --> 00:18:08,440 So some of the tools I'll show you today that essentially I'm using to mock up the API are 149 00:18:08,440 --> 00:18:23,280 essentially the Candletics engine and then the Candletics API. 150 00:18:23,280 --> 00:18:37,880 And I'll go ahead and commit the latest work here, which has these two mock endpoints. 151 00:18:37,880 --> 00:18:44,120 So I've gone ahead and mocked the lab results endpoint and the MMEs endpoint. 152 00:18:44,120 --> 00:18:50,720 And then we can go ahead and I'll gradually mock all of the LEAF API endpoints. 153 00:18:50,720 --> 00:18:57,760 And then you can use them as you please for testing and whatnot. 154 00:18:57,760 --> 00:19:02,320 But they'll be fairly limited with a limited amount of data. 155 00:19:02,320 --> 00:19:20,760 So 156 00:19:20,760 --> 00:19:26,840 and then I'll go ahead and share these links with you here in the chat. 157 00:19:26,840 --> 00:19:31,120 The API and the engine. 158 00:19:31,120 --> 00:19:38,160 And so essentially, I'll just go ahead and just kind of show you these. 159 00:19:38,160 --> 00:19:39,880 Nothing set in stone here. 160 00:19:39,880 --> 00:19:47,920 These are just sort of general tools that I was putting together just to help with cannabis 161 00:19:47,920 --> 00:19:49,840 analytics. 162 00:19:49,840 --> 00:19:53,380 So we're doing cannabis data science here. 163 00:19:53,380 --> 00:20:05,480 So these are just sort of open ended packages that you can use to contribute to help your 164 00:20:05,480 --> 00:20:07,560 workflow. 165 00:20:07,560 --> 00:20:21,400 So the engine is just a Python package, which you can just install with your command prompt. 166 00:20:21,400 --> 00:20:31,800 So you can just, you know, install can lytics to get the latest version. 167 00:20:31,800 --> 00:20:35,920 And we're at the moment. 168 00:20:35,920 --> 00:20:40,600 So you've now got kin lytics installed. 169 00:20:40,600 --> 00:20:51,160 And then let me actually restart spider and I'll show you a bit more. 170 00:20:51,160 --> 00:21:05,120 But essentially, at the moment, it just leverages Firebase for a database. 171 00:21:05,120 --> 00:21:11,880 So if you want to learn about Firebase, I'll put that in the chat. 172 00:21:11,880 --> 00:21:22,800 So Firebase is sort of a fork of MongoDB and I also like MongoDB. 173 00:21:22,800 --> 00:21:24,280 So that's worth checking out. 174 00:21:24,280 --> 00:21:29,880 And so these are what you'd call no SQL database. 175 00:21:29,880 --> 00:21:37,920 So it doesn't have quite the same connective structure as the SQL database. 176 00:21:37,920 --> 00:21:44,440 And just to kind of just give you a look at what the data looks like here. 177 00:21:44,440 --> 00:22:02,080 So here is just some collections that I've made with some of the MME data. 178 00:22:02,080 --> 00:22:06,600 So you have essentially a directory. 179 00:22:06,600 --> 00:22:13,320 So you can think about it as a directory of collections and documents. 180 00:22:13,320 --> 00:22:18,920 And then all documents are essentially a JSON file. 181 00:22:18,920 --> 00:22:23,160 So you can think about these as key value pairs. 182 00:22:23,160 --> 00:22:29,680 So you essentially just have a giant JSON database. 183 00:22:29,680 --> 00:22:32,640 So it has its pros and its cons. 184 00:22:32,640 --> 00:22:37,400 The cons right off the bat are going to be it's much harder. 185 00:22:37,400 --> 00:22:39,480 Well, it depends on your perspective. 186 00:22:39,480 --> 00:22:47,400 But it can be harder to do queries and aggregations than it is in an SQL database. 187 00:22:47,400 --> 00:22:53,160 And then it has some advantages in that you're just going JSON to JSON. 188 00:22:53,160 --> 00:22:59,960 And sometimes the loose structure can be beneficial. 189 00:22:59,960 --> 00:23:05,560 And it can be fast if you know what you're looking for. 190 00:23:05,560 --> 00:23:08,480 So once again, nothing set in stone. 191 00:23:08,480 --> 00:23:12,680 But I essentially just tossed the MME data. 192 00:23:12,680 --> 00:23:26,240 And I tossed about 1,000 lab results here just so that way you can just pull these to 193 00:23:26,240 --> 00:23:28,720 start doing some practice. 194 00:23:28,720 --> 00:23:56,080 So how did I get those there? 195 00:23:56,080 --> 00:24:01,560 I used the Cantlytics engine. 196 00:24:01,560 --> 00:24:03,480 I think I put that in the chat. 197 00:24:03,480 --> 00:24:07,720 Yeah, so that's the Cantlytics engine. 198 00:24:07,720 --> 00:24:14,640 And so this is just essentially a wrapper around Firebase. 199 00:24:14,640 --> 00:24:18,320 So Firebase already has a Python SDK. 200 00:24:18,320 --> 00:24:27,880 But this is an even more expedient way to interact with the data. 201 00:24:27,880 --> 00:24:33,840 And so just put some useful tools here. 202 00:24:33,840 --> 00:24:41,480 So you can initialize Firebase. 203 00:24:41,480 --> 00:24:48,520 Use a bunch of wrappers to get documents, get collections. 204 00:24:48,520 --> 00:24:50,720 You can import data. 205 00:24:50,720 --> 00:25:01,000 So this handles importing data into Firebase from CSVs and Excel. 206 00:25:01,000 --> 00:25:03,920 So a fairly way to get your data in. 207 00:25:03,920 --> 00:25:10,960 And then so that's essentially one thing that Firebase was lacking was easy imports and 208 00:25:10,960 --> 00:25:12,640 easy exports. 209 00:25:12,640 --> 00:25:15,080 So there you go. 210 00:25:15,080 --> 00:25:24,760 Now you've got an easy way to import data and an easy way to export data from Firebase. 211 00:25:24,760 --> 00:25:41,720 And if you look in the tests, you can see how you can go about spinning up your Firebase 212 00:25:41,720 --> 00:25:43,880 instance. 213 00:25:43,880 --> 00:26:02,280 And then I actually use that code right there to populate these two collections. 214 00:26:02,280 --> 00:26:04,360 So it's all open source. 215 00:26:04,360 --> 00:26:11,560 So you can use it yourself and populate your own Firebase database essentially for free 216 00:26:11,560 --> 00:26:16,120 if that's something you're looking to do. 217 00:26:16,120 --> 00:26:19,160 You do have to pay for Firebase. 218 00:26:19,160 --> 00:26:22,240 But at the moment, the cost is really low. 219 00:26:22,240 --> 00:26:28,080 So as long as you keep your usage low, then it's essentially free. 220 00:26:28,080 --> 00:26:31,080 So it's a nifty tool. 221 00:26:31,080 --> 00:26:36,080 But I'm not set in stone to any one particular tool. 222 00:26:36,080 --> 00:26:40,280 So use your favorite tool there. 223 00:26:40,280 --> 00:26:46,720 And so then we've got our data there with the CanleLytx engine. 224 00:26:46,720 --> 00:26:54,680 And so now we can retrieve it with the CanleLytx API. 225 00:26:54,680 --> 00:27:03,360 And so once again, you can find the code here on GitHub. 226 00:27:03,360 --> 00:27:12,520 But this is essentially going to be the interface to all the actions that you can do in CanleLytx. 227 00:27:12,520 --> 00:27:24,720 So if you want to interact with any of this data, then you can use the CanleLytx API, 228 00:27:24,720 --> 00:27:30,400 which includes batteries included. 229 00:27:30,400 --> 00:27:32,940 It includes the CanleLytx engine. 230 00:27:32,940 --> 00:27:39,480 So this tool is bundled up inside the API. 231 00:27:39,480 --> 00:27:42,280 So once again, nothing is set in stone. 232 00:27:42,280 --> 00:27:50,220 But it's just a classic example of dog fooding, where you eat your own dog food. 233 00:27:50,220 --> 00:27:53,760 So you use your own tools. 234 00:27:53,760 --> 00:28:02,720 So that's the idea here is just building a set of tooling to perform cannabis analysis. 235 00:28:02,720 --> 00:28:07,560 That way, we can control everything from the ground up. 236 00:28:07,560 --> 00:28:17,880 We can control database interactions so you can get pretty low level. 237 00:28:17,880 --> 00:28:22,680 And then there's actually a lot more here than just database. 238 00:28:22,680 --> 00:28:35,640 So for example, you can work with storage, so you can put files into storage. 239 00:28:35,640 --> 00:28:42,760 You can download files, rename files, delete files. 240 00:28:42,760 --> 00:28:48,480 There's also authentication, so you can create users. 241 00:28:48,480 --> 00:28:52,000 You can give the users custom claims. 242 00:28:52,000 --> 00:28:55,400 You can give them custom tokens. 243 00:28:55,400 --> 00:29:00,960 And so the token would be for if you want to log in on your front end. 244 00:29:00,960 --> 00:29:06,760 So this is essentially a backend system that can talk with your front end. 245 00:29:06,760 --> 00:29:15,360 And then there's utility functions to get your users, update a user, then delete a user 246 00:29:15,360 --> 00:29:17,720 if need be. 247 00:29:17,720 --> 00:29:25,040 So it's just a little backend service wrapper around Firebase. 248 00:29:25,040 --> 00:29:36,520 So if you need those services, then this may be an easy interface for you. 249 00:29:36,520 --> 00:29:44,020 So let's go ahead and test some of these things out. 250 00:29:44,020 --> 00:29:49,080 So you can essentially clone the repository. 251 00:29:49,080 --> 00:29:53,920 And then you'll have this repo here. 252 00:29:53,920 --> 00:30:05,960 Looks like everything went through. 253 00:30:05,960 --> 00:30:13,120 And then just to kind of walk you through, so first things first, you'd clone the repository. 254 00:30:13,120 --> 00:30:20,240 And then you can actually run the server locally. 255 00:30:20,240 --> 00:30:26,440 So this is sort of to give you an example of how you could go about doing that. 256 00:30:26,440 --> 00:30:36,600 So you'd go to the folder where you downloaded, where you cloned the API. 257 00:30:36,600 --> 00:30:41,720 You can open the command console. 258 00:30:41,720 --> 00:31:02,120 And then you can just do, so you can just run server. 259 00:31:02,120 --> 00:31:06,940 It looks like one of my endpoints is not 100% correct here. 260 00:31:06,940 --> 00:31:08,280 So we can just jump in there. 261 00:31:08,280 --> 00:31:12,440 And so just to kind of show you a little bit about the API and tell you a little bit about 262 00:31:12,440 --> 00:31:13,440 it. 263 00:31:13,440 --> 00:31:20,520 So you've learned now that, OK, the database is Firebase Firestore. 264 00:31:20,520 --> 00:31:26,200 And so the API is Python Django. 265 00:31:26,200 --> 00:31:36,760 And so the typical structure here of a Django file, so in the root directory, you have your 266 00:31:36,760 --> 00:31:39,480 requirements. 267 00:31:39,480 --> 00:31:44,040 You have your read me. 268 00:31:44,040 --> 00:31:50,640 I'm just using node just to kind of help run some helper functions. 269 00:31:50,640 --> 00:31:55,760 There's no actual node dependencies. 270 00:31:55,760 --> 00:32:05,020 And then just, and then Docker to just get it up to, it's actually running in Google 271 00:32:05,020 --> 00:32:07,000 Cloud App Engine. 272 00:32:07,000 --> 00:32:14,120 But you can run it anywhere that you desire. 273 00:32:14,120 --> 00:32:22,800 And so just to give you a quick structure, so all the code is in the Canalytics API. 274 00:32:22,800 --> 00:32:30,600 And so just to give you an idea of what a Django project looks like, you'll have a settings 275 00:32:30,600 --> 00:32:33,720 file. 276 00:32:33,720 --> 00:32:37,160 And this is typically fairly long. 277 00:32:37,160 --> 00:32:39,600 You kind of want to try to keep it short. 278 00:32:39,600 --> 00:32:47,200 And this is just where you define all your key parameters towards how everything's going 279 00:32:47,200 --> 00:32:51,960 to be functioning. 280 00:32:51,960 --> 00:32:59,020 So this is where you define any third party apps you're using. 281 00:32:59,020 --> 00:33:06,880 And we're going to be using the rest framework, since we're going to be building an API here. 282 00:33:06,880 --> 00:33:12,020 And then everything else is relatively just boilerplate. 283 00:33:12,020 --> 00:33:19,000 So the rest of this is essentially just boilerplate to make everything else run. 284 00:33:19,000 --> 00:33:27,120 And you'll need to set up essentially your example credentials. 285 00:33:27,120 --> 00:33:43,440 So if you walk through the read me, I may have to flesh that out a bit about how to 286 00:33:43,440 --> 00:33:45,920 get set up with credentials. 287 00:33:45,920 --> 00:33:56,160 But essentially, you'll need to get these credentials set to really use this in development. 288 00:33:56,160 --> 00:34:01,840 So you've got your settings. 289 00:34:01,840 --> 00:34:09,040 And then where all the beauty happens, well, so you've got your URLs. 290 00:34:09,040 --> 00:34:13,220 And so these are going to be your actual endpoints. 291 00:34:13,220 --> 00:34:19,200 So for example, your endpoint here. 292 00:34:19,200 --> 00:34:23,040 And so it didn't like that these began with a slash. 293 00:34:23,040 --> 00:34:30,520 So your endpoint here is just going to be test, leaf, MMEs. 294 00:34:30,520 --> 00:34:36,960 And so I'm pointing this to essentially a Python function. 295 00:34:36,960 --> 00:34:43,280 So I just put these in this file over here. 296 00:34:43,280 --> 00:34:54,720 And so then this can give you some ideas about what Django endpoints may look like. 297 00:34:54,720 --> 00:35:06,240 So for a REST API, it's just a Python function that takes the request as an argument. 298 00:35:06,240 --> 00:35:10,600 You check the method of the request. 299 00:35:10,600 --> 00:35:18,720 So say you wanted to add functionality here for labs to actually create results. 300 00:35:18,720 --> 00:35:24,560 Well, then you would add a post. 301 00:35:24,560 --> 00:35:49,000 And then you would say, you know, right? 302 00:35:49,000 --> 00:36:00,320 And then you may even just, you know, you could return a not implemented error. 303 00:36:00,320 --> 00:36:07,440 So eventually, we'll want to go ahead and build out that functionality. 304 00:36:07,440 --> 00:36:13,400 Since we're essentially, remember, let's not lose sight of the original goal. 305 00:36:13,400 --> 00:36:29,920 Our original goal is we're essentially just creating a mock leaf API so we can, you know, 306 00:36:29,920 --> 00:36:42,800 so we can just, you know, basically mock some of these calls ourselves. 307 00:36:42,800 --> 00:36:53,560 So let's go ahead and see, you know, see this in action. 308 00:36:53,560 --> 00:36:57,240 So you know, we can come over here. 309 00:36:57,240 --> 00:37:07,240 You know, we'll just create a blank file here. 310 00:37:07,240 --> 00:37:17,440 Great, and so just since we're working in Python, you know, we can use the request package 311 00:37:17,440 --> 00:37:25,160 and, you know, the base. 312 00:37:25,160 --> 00:37:32,280 So the production base, if you want to test out what's in production, it's going to be 313 00:37:32,280 --> 00:37:43,320 so it's going to be API.canalytics.com. 314 00:37:43,320 --> 00:37:52,440 But you know, for right now, we're just going to be using the development base. 315 00:37:52,440 --> 00:37:57,360 So that way, you know, I can make, you know, rapid changes here just for, you know, just 316 00:37:57,360 --> 00:38:02,960 to kind of show, you know, show what we can do in development. 317 00:38:02,960 --> 00:38:08,320 And that way you can learn how to do development yourself if you're interested. 318 00:38:08,320 --> 00:38:11,360 So we've got it set up. 319 00:38:11,360 --> 00:38:16,460 Remember, we're going to be hitting these endpoints. 320 00:38:16,460 --> 00:38:33,440 So let's try to get the MMEs. 321 00:38:33,440 --> 00:38:46,600 So we just, you know, we're just going to make a get request, so requests. 322 00:38:46,600 --> 00:39:05,320 Took a second, so we got a 200 status code, I don't know. 323 00:39:05,320 --> 00:39:13,600 So I guess let's just do a data dump and see what we got. 324 00:39:13,600 --> 00:39:14,760 No promises here. 325 00:39:14,760 --> 00:39:17,360 This is a live test. 326 00:39:17,360 --> 00:39:21,840 Oh, sure enough, look at that. 327 00:39:21,840 --> 00:39:25,800 So that's a little much to work with. 328 00:39:25,800 --> 00:39:29,840 So let's just look at the first observation. 329 00:39:29,840 --> 00:39:33,240 Oh, so it's not a list. 330 00:39:33,240 --> 00:39:36,920 So let's see what we're even working with here. 331 00:39:36,920 --> 00:39:40,440 So we've got a dictionary. 332 00:39:40,440 --> 00:39:50,520 Ah, so it looks like at the moment the CanLytics API, let's actually go and so this is what's 333 00:39:50,520 --> 00:39:56,560 nice from building it from the ground up is you can actually now come over here and we 334 00:39:56,560 --> 00:40:03,720 can actually look at the endpoint so we can see what the endpoints actually doing. 335 00:40:03,720 --> 00:40:08,200 So that way it's not a black box, right? 336 00:40:08,200 --> 00:40:13,200 So now we control the data import, right? 337 00:40:13,200 --> 00:40:22,240 So we've now imported data here into Firebase. 338 00:40:22,240 --> 00:40:25,800 We've got the records here, right? 339 00:40:25,800 --> 00:40:34,320 So here we have somebody at 555 Brown Street, LCB Grow. 340 00:40:34,320 --> 00:40:35,960 Okay. 341 00:40:35,960 --> 00:40:44,360 So we want to basically be able to query that with our API. 342 00:40:44,360 --> 00:40:48,640 So let's start tinkering. 343 00:40:48,640 --> 00:40:54,680 So we're getting our MME data. 344 00:40:54,680 --> 00:40:59,560 So if we make our request, it'll check to see if there's a limit. 345 00:40:59,560 --> 00:41:02,180 Otherwise there won't be one. 346 00:41:02,180 --> 00:41:13,080 So let's try, let's test this limit out. 347 00:41:13,080 --> 00:41:24,760 So the way you could do that is you're just going to do Q or question mark, limit. 348 00:41:24,760 --> 00:41:38,280 Let's just try doing a limit of two. 349 00:41:38,280 --> 00:41:56,560 Yeah, that's right. 350 00:41:56,560 --> 00:42:05,560 So now let's, it looks like we may have gotten an error. 351 00:42:05,560 --> 00:42:09,600 So we got a 500 error code. 352 00:42:09,600 --> 00:42:19,920 And what's nice is, you know, we're in development here, so we can actually see the debug error. 353 00:42:19,920 --> 00:42:23,400 So you know, so this is how you would do development. 354 00:42:23,400 --> 00:42:31,960 So you would see, okay, there was a get request to the LEAF MMEs. 355 00:42:31,960 --> 00:42:40,000 Ah, and look, here's the error. 356 00:42:40,000 --> 00:42:45,920 To has class string, but we expected an integer. 357 00:42:45,920 --> 00:42:47,000 Okay. 358 00:42:47,000 --> 00:42:52,560 So we can go and tinker on this MMEs endpoint. 359 00:42:52,560 --> 00:42:57,720 So we now know this limit has to be an integer. 360 00:42:57,720 --> 00:43:05,400 So just a hot fix here, we could do better logic later on, but you know, you could just 361 00:43:05,400 --> 00:43:10,000 say, you know, you could just do something silly like that, you know. 362 00:43:10,000 --> 00:43:17,280 So if there's a limit, you know, make sure the limit's an integer. 363 00:43:17,280 --> 00:43:23,600 You know, and so go ahead and save that. 364 00:43:23,600 --> 00:43:30,560 This should, I think this should just refresh. 365 00:43:30,560 --> 00:43:38,160 I'm not sure if I've got the live server rerunning. 366 00:43:38,160 --> 00:43:57,720 Interesting… 367 00:43:57,720 --> 00:43:58,560 So. 368 00:43:58,560 --> 00:43:59,380 60 seconds for you. 369 00:44:28,560 --> 00:44:31,040 little bit. 370 00:44:32,880 --> 00:44:38,040 And actually we may even just want to change this just to return the docs just 371 00:44:38,040 --> 00:44:45,400 outright. I think that's just a bit a bit better. And let's actually just go 372 00:44:45,400 --> 00:44:54,840 ahead and re-spin this server. When you're actually doing 373 00:44:54,840 --> 00:45:05,840 development there is a live reload package which makes things a lot better. 374 00:45:05,960 --> 00:45:08,960 Okay. 375 00:45:11,160 --> 00:45:19,120 Okay let's try this again. Okay so now we've got a fast response. We've got our 376 00:45:19,120 --> 00:45:30,000 200 status code and we've got two documents here. So now this 377 00:45:30,000 --> 00:45:37,320 little fix here to the limit that seemed to do the trick. And if we 378 00:45:37,320 --> 00:45:45,840 look at our data we've got 555 Brown Street, the LCB grow. 379 00:45:45,840 --> 00:45:54,800 And so the next step is really to set this up to actually handle queries. And 380 00:45:54,800 --> 00:46:09,840 so the nifty thing is the get collection routine is set up to handle 381 00:46:09,840 --> 00:46:18,680 filters. So we can basically pass filters of the sort. 382 00:46:18,680 --> 00:46:38,360 Key, you know, so say we wanted you know key, you know, name, operation. Right? 383 00:46:39,480 --> 00:46:48,600 So on my to-do list is to actually set up these endpoints to be able to handle 384 00:46:48,600 --> 00:46:55,320 queries. Because you know we could actually set this up you know to really 385 00:46:55,320 --> 00:47:01,040 handle any sort of query. Because it would be cool to set it up so that way 386 00:47:01,040 --> 00:47:08,480 you could query by phone number, address. You could even do something that matches 387 00:47:08,480 --> 00:47:11,880 close to the address. 388 00:47:11,880 --> 00:47:18,000 You could of course do global ID. You don't want to do if anybody's suspended. 389 00:47:18,000 --> 00:47:26,400 You want to do type, city. So all of these are fields that we could set up to 390 00:47:26,400 --> 00:47:33,440 query over here. Okay? So now let's go ahead and you know let's hit this lab 391 00:47:33,440 --> 00:47:41,400 results endpoint to kind of see if it works as well. So we'll probably have to 392 00:47:41,400 --> 00:47:47,640 do a similar thing where we fix the limit. 393 00:47:49,960 --> 00:47:58,840 And this is also not set up for queries. But let's okay so we go ahead and got a 394 00:47:58,840 --> 00:48:02,000 live reload there. 395 00:48:02,000 --> 00:48:23,360 And so now let's check out. Now we're just going to hit the lab results endpoint 396 00:48:23,360 --> 00:48:41,720 and see what happens. Okay it looks like we may have gotten an error. Okay so we 397 00:48:41,720 --> 00:48:48,600 got a 404. So that means this endpoint was not found. So we probably want to 398 00:48:48,600 --> 00:49:00,920 also set up our URLs you know to you know it's better handle 404s. 399 00:49:00,920 --> 00:49:06,120 Because you know the idea of any endpoint I mean of any API is you want 400 00:49:06,120 --> 00:49:11,560 to handle all endpoints elegantly. So you know if somebody hits your base 401 00:49:11,560 --> 00:49:17,960 endpoint you know you want to so here's just a look at some of the other 402 00:49:17,960 --> 00:49:25,520 endpoints of the CanLytics API. So you know so if somebody hits the base 403 00:49:25,520 --> 00:49:34,320 endpoint so let's just say you just hit the base here you know you still get a 404 00:49:34,320 --> 00:49:42,560 status code and you and then you know you actually get a message. So it actually 405 00:49:42,560 --> 00:49:48,360 says okay you know welcome to the CanLytics API. You know you know we 406 00:49:48,360 --> 00:49:52,000 probably want to touch this up instead of data you know we may actually want to 407 00:49:52,000 --> 00:49:59,080 change that to message. But you know you're welcome to come in here and 408 00:49:59,080 --> 00:50:05,160 tinker on this because this is just a completely open source just sort of fun 409 00:50:05,160 --> 00:50:11,160 API that is just sort of set up just to you know tinker around with cannabis 410 00:50:11,160 --> 00:50:17,880 data. And you know like I said I'm not set in stone to any technology so if 411 00:50:17,880 --> 00:50:23,120 anybody has recommendations you know fly them out you know I'm not so you know 412 00:50:23,120 --> 00:50:30,200 I'm not stuck in stone on anything. And so just to give you a look here so we 413 00:50:30,200 --> 00:50:39,840 just hit the base endpoint we just got this message and so not to get too off 414 00:50:39,840 --> 00:50:46,120 track here but the idea is you know to handle all endpoints elegantly. Okay so 415 00:50:46,120 --> 00:50:52,480 the reason this one failed is we're actually we actually want to hit the 416 00:50:52,480 --> 00:51:01,040 test endpoint the test leaf endpoint. So let's go ahead and hit that endpoint and 417 00:51:01,040 --> 00:51:03,760 see what happens. 418 00:51:03,760 --> 00:51:10,680 And I'll publish all of this afterwards so that actually I'll do that right now 419 00:51:10,680 --> 00:51:19,880 that way you can actually hit the live API. Okay so now we've got a we've got 420 00:51:19,880 --> 00:51:34,880 our response, we've got our status code, and hopefully we have two lab results. 421 00:51:41,360 --> 00:51:48,840 Okay so we're still returning lab let's actually fix that real quick so that 422 00:51:48,840 --> 00:51:54,160 actually is kind of annoying me so it may annoy other people. I think it's just 423 00:51:54,160 --> 00:51:58,240 better if this just returns 424 00:52:02,400 --> 00:52:10,080 I think it's just better if it just returns the documents as a list. 425 00:52:10,080 --> 00:52:20,440 Got our live reload and see this is what's nice about you know cooking up 426 00:52:20,440 --> 00:52:27,200 your own little API here because now you know with just a couple lines of code 427 00:52:27,200 --> 00:52:32,160 you know we've now got this is a list and if you prefer it as something else 428 00:52:32,160 --> 00:52:37,440 then you know get in there and tinker and change it. So now we you know we've 429 00:52:37,440 --> 00:52:50,760 got our list here we've got two and you know we've got these lab results and so 430 00:52:50,760 --> 00:53:01,400 this is these are the exact same you know lab results as 431 00:53:01,400 --> 00:53:13,000 these are the same lab results you know from that LCB data dump. So don't be 432 00:53:13,000 --> 00:53:16,840 here's the link to the data dump. 433 00:53:22,560 --> 00:53:27,200 Okay but long story short so we've gone a long ways now we've basically gone 434 00:53:27,200 --> 00:53:35,640 from this data dump here we've gotten the data into Firebase we did that with 435 00:53:35,640 --> 00:53:40,960 the help of the Canalytics engine and so you're welcome to go and explore over 436 00:53:40,960 --> 00:53:49,200 here you know and check out check out the Firebase module you know we've used 437 00:53:49,200 --> 00:53:55,800 the import data function and then if you want just examples of that go check out 438 00:53:55,800 --> 00:54:02,680 test because the test is you know this can this can get you started that's how 439 00:54:02,680 --> 00:54:09,040 you initialize Firebase you know create some documents and then import your data 440 00:54:09,040 --> 00:54:17,000 and then now that we've got our data in Firebase we need a way to interact with 441 00:54:17,000 --> 00:54:29,600 it so we've got the Canalytics API and that's as simple to use as just you know 442 00:54:29,600 --> 00:54:35,680 making some get requests to you know to these various endpoints and then you've 443 00:54:35,680 --> 00:54:44,120 got this full suite of data and just to go ahead and polish it off I'll just go 444 00:54:44,120 --> 00:54:53,360 ahead and round it out by just showing you so you've you've done the install 445 00:54:53,360 --> 00:54:55,400 it 446 00:54:58,200 --> 00:55:03,800 all right so you've got the Canalytics API so you've done the installation we've 447 00:55:03,800 --> 00:55:11,680 done the development we've done the testing and so now it's time to publish 448 00:55:11,680 --> 00:55:20,560 well you know that's as easy to do it's just you know NPM and so that's why I 449 00:55:20,560 --> 00:55:31,280 have node is to expedite the publishing process so what's going on here so if 450 00:55:31,280 --> 00:55:39,760 you go and look over here and publish so basically we're containerizing the API 451 00:55:39,760 --> 00:55:52,640 so we create a container of the API then we push the container up to Google Cloud 452 00:55:52,640 --> 00:56:04,440 and then we actually are then redirecting all hosting from Firebase so 453 00:56:04,440 --> 00:56:13,200 over in Firebase we've actually got hosting and you can set up multiple 454 00:56:13,200 --> 00:56:20,760 sites here and you know we've actually got the you know the Canalytics API up 455 00:56:20,760 --> 00:56:30,640 here and so you know right now we're containerizing everything and you know 456 00:56:30,640 --> 00:56:38,560 it's going to be a little bit of a build process here so we're publishing that 457 00:56:40,920 --> 00:56:46,760 so I'll just sort of minimize this and then sort of bring it all home here so 458 00:56:46,760 --> 00:57:02,520 that way we can get people out of here in time for their next seminar 459 00:57:03,720 --> 00:57:13,000 all right so sorry to just kind of go on that long spiel there but you know what 460 00:57:13,000 --> 00:57:19,280 your what's your input what's your input too complicated of a way to get data or 461 00:57:19,280 --> 00:57:30,960 a manageable way I mean it's really cool it's probably a little more than what I 462 00:57:30,960 --> 00:57:41,200 would need but it's really interesting well this just got posted over here so 463 00:57:41,200 --> 00:57:44,760 here in this last couple of minutes 464 00:57:52,520 --> 00:57:57,360 we'll see if I wanted to do like more advanced things this would be this would 465 00:57:57,360 --> 00:58:02,440 be really useful yes so now it's published so now in this last minute 466 00:58:02,440 --> 00:58:14,720 let's see if we can't move over to the production endpoint and get the same we 467 00:58:14,720 --> 00:58:17,640 looks like 468 00:58:17,640 --> 00:58:24,640 it looks like there is an error along the way that's interesting oh I need a 469 00:58:31,640 --> 00:58:34,640 but it's interesting 470 00:58:34,640 --> 00:58:53,320 oh I need a but we got 200 status codes and so if we check out our data here 471 00:58:53,320 --> 00:59:04,440 we've got the two MMEs and we have our two lab results so thank you for bearing 472 00:59:04,440 --> 00:59:13,200 with it but there's essentially you know an MVP you know a bit of a viable 473 00:59:13,200 --> 00:59:21,640 product right so we've basically taken our data we've gotten it into you know 474 00:59:21,640 --> 00:59:39,040 firebase storage anyways we've got it there using the engine and now we've 475 00:59:39,040 --> 00:59:45,040 retrieved it from the you know the analytics API so you know you're welcome 476 00:59:45,040 --> 00:59:51,840 to you know go tinker on that and 477 00:59:56,240 --> 01:00:01,240 and see you know what happens 478 01:00:02,800 --> 01:00:05,800 but 479 01:00:05,800 --> 01:00:23,280 but anyways it looks like there's still some debugging to do so I'll go ahead 480 01:00:23,280 --> 01:00:26,280 and wrap it up there 481 01:00:26,280 --> 01:00:33,200 but thank you thank you for attending and I'll fill you in and let you know 482 01:00:33,200 --> 01:00:37,920 what's going on next week and next week we'll be checking out that SQL database 483 01:00:37,920 --> 01:00:42,120 see you then awesome thanks for attending Nick 484 01:00:42,120 --> 01:00:56,960 Michael