1 00:00:00,000 --> 00:00:19,000 Welcome to the Cannabis Data Science Meetup Group. You're in for a treat today. We're going to look at a little bit of research done by some group members and then introduce you to the first ever cannabis data market. 2 00:00:19,000 --> 00:00:31,000 And then we'll start visualizing that data. So without further ado, I'm going to go ahead and share my screen with you. Awesome. Welcome to the group. 3 00:00:31,000 --> 00:00:46,000 For those of you who are new to the Cannabis Data Science Meetup Group, you can always find the source code on GitHub. So I'm going to go ahead and commit today's code. 4 00:00:46,000 --> 00:00:58,000 And then you can follow along. 5 00:00:58,000 --> 00:01:06,000 And code for yourself in Python. So I'm going to move pretty quick through the code. So it's not the end of the world if 6 00:01:06,000 --> 00:01:16,000 you come along and look at this later on down the road here. So this will be the script that we'll be working with today. 7 00:01:16,000 --> 00:01:25,000 Before we dive into this, I just wanted to share with you the first ever cannabis data science publication. 8 00:01:25,000 --> 00:01:36,000 And so this is an article that a group member, Paul Kitko, who's got his hands busy because he now has a master's in data science. 9 00:01:36,000 --> 00:01:45,000 And he wrote an article using sales data. So cannabis sales data from Washington State. 10 00:01:45,000 --> 00:01:58,000 And so he did a fantastic analysis where he looked at the combinations of goods that consumers purchase together. 11 00:01:58,000 --> 00:02:10,000 So I highly encourage you all to check this out. So I'll share a link with you. 12 00:02:10,000 --> 00:02:20,000 And then you're free to read up on Paul's work when you have a chance. It's exciting stuff. 13 00:02:20,000 --> 00:02:29,000 And here's just the data science group as well. 14 00:02:29,000 --> 00:02:34,000 And welcome to the group, Julia. 15 00:02:34,000 --> 00:02:42,000 Well, we can talk, do introductions and whatnot here in a second, just sort of introducing some of the new material. 16 00:02:42,000 --> 00:02:47,000 Long story short, awesome paper by Paul that you should check out. 17 00:02:47,000 --> 00:02:55,000 And then if you want to get your hands on some data, and we'll be pawing through it today, 18 00:02:55,000 --> 00:03:00,000 or if you want to publish your own data, you can check out the Cantlytics data market. 19 00:03:00,000 --> 00:03:07,000 And so here we're making these data sets easily available. 20 00:03:07,000 --> 00:03:20,000 Everything is open source. So making all of the source code available so that way you can compile these data sets yourself. 21 00:03:20,000 --> 00:03:26,000 But it can be a non-trivial task to put together a data set. 22 00:03:26,000 --> 00:03:34,000 So if you want to save yourself the heartache, then you can come check out the data set that's ready to go. 23 00:03:34,000 --> 00:03:39,000 Or, you know, you can put it together for yourself. 24 00:03:39,000 --> 00:03:50,000 And so what's awesome about the economics of a data market is it really gets the hands efficiently into consumers. 25 00:03:50,000 --> 00:03:58,000 So it's great for suppliers and consumers. So centrally, in the long run, 26 00:03:58,000 --> 00:04:06,000 you would expect that the prices will start approaching average fixed cost here. And why do I say that? 27 00:04:06,000 --> 00:04:15,000 Well, if you have a better idea of what this data could be worth to the consumer, 28 00:04:15,000 --> 00:04:22,000 so say you think you can sell this to two consumers at $400 a share, 29 00:04:22,000 --> 00:04:29,000 then it would be prudent of you to get this data, publish it yourself, 30 00:04:29,000 --> 00:04:43,000 and then you can get it into the hands of the consumers that you think have demand for this product. 31 00:04:43,000 --> 00:04:52,000 So for the time being, just send me a message. But this is just going to be an open source, open data data market. 32 00:04:52,000 --> 00:04:58,000 And it should efficiently get data into the hands of people that demand it. 33 00:04:58,000 --> 00:05:05,000 And also, it should facilitate the supply of data collection algorithms. 34 00:05:05,000 --> 00:05:22,000 So here in the group, if you're more interested in algorithms, here I'm just standardizing some of the work that we're doing here in the Cannabis Data Science Meetup Group. 35 00:05:22,000 --> 00:05:28,000 So check it out. 36 00:05:28,000 --> 00:05:36,000 And you're welcome to contribute or browse the data. 37 00:05:36,000 --> 00:05:42,000 So just letting you know that's there and Paul's paper's there. 38 00:05:42,000 --> 00:05:48,000 So real quick, before we get into the code here, 39 00:05:48,000 --> 00:05:55,000 just would be interested in just doing a quick round of introductions before just start zoning on there. 40 00:05:55,000 --> 00:06:02,000 So in my top corner, Julia, would you be interested in introducing yourself to the group? 41 00:06:02,000 --> 00:06:12,000 Sure. So my name is Julia. I'm currently a PhD candidate and I'm starting to transition more into the data science world. 42 00:06:12,000 --> 00:06:17,000 So I'm here to try to practice the data science, potentially learn stuff that I don't know. 43 00:06:17,000 --> 00:06:22,000 I'm a marketing entrepreneur and executive. 44 00:06:22,000 --> 00:06:31,000 I live in upstate New York and we have three major cannabis production and distribution facilities going in in my county. 45 00:06:31,000 --> 00:06:36,000 So I'm just doing a lot of research and gathering a lot of information. 46 00:06:36,000 --> 00:06:41,000 I'm sure there will be a lot of interest in whatever data can come out of this. 47 00:06:41,000 --> 00:06:48,000 I do some of my own research. Very interested in that report you showed this morning and what I can do with that. 48 00:06:48,000 --> 00:06:52,000 So that's why I'm here. You're cool. Definitely check it out. 49 00:06:52,000 --> 00:06:58,000 Paul did an awesome analysis on retail, one of the more rigorous studies you can find out there. 50 00:06:58,000 --> 00:07:03,000 He is a whiz. So to check it out, highly rated. 51 00:07:03,000 --> 00:07:11,000 Allie, would you be interested in introducing yourself? Yeah, I'm Allie and I'm in accounting right now. 52 00:07:11,000 --> 00:07:21,000 So not data science, but not too, too far off. My husband and I are looking at starting a business in the cannabis industry. 53 00:07:21,000 --> 00:07:27,000 And we're both really passionate about the expansion of the cannabis industry, especially in a nonprofit standpoint. 54 00:07:27,000 --> 00:07:34,000 And I think data science is the beginning to that transition. 55 00:07:34,000 --> 00:07:42,000 Brilliant. There's a high demand for data science and. 56 00:07:42,000 --> 00:07:48,000 I'd be thrilled to see the work you do, so I'll definitely be following your work. So that's too cool. 57 00:07:48,000 --> 00:07:55,000 And then M&R, would you be interested in introducing yourself? 58 00:07:55,000 --> 00:08:01,000 Sure. Yeah, I just finished a masters for public health, actually. 59 00:08:01,000 --> 00:08:05,000 And I'm working on like a survey statistician role in Florida. 60 00:08:05,000 --> 00:08:09,000 And so data science is what I want to do for my career. 61 00:08:09,000 --> 00:08:18,000 I'm thinking about even getting a second master's in data science possibly as a I don't know if the PhD route would be necessary, especially with the job I have. 62 00:08:18,000 --> 00:08:26,000 And so cannabis is a super interesting emerging industry and I'm interested in learning more about both those things. 63 00:08:26,000 --> 00:08:34,000 Too cool. You're well positioned as well. Long story short, we've got this data augmentation. 64 00:08:34,000 --> 00:08:45,000 And just to, I guess, give you just a brief background of where we're coming from. 65 00:08:45,000 --> 00:08:50,000 You know, I always think it's helpful to just start with a research question. 66 00:08:50,000 --> 00:09:05,000 So we don't actually have too much of a question here. So it's more just can we discover any cultivation patterns just by essentially augmenting existing data and then visualizing? 67 00:09:05,000 --> 00:09:07,000 Simple enough. 68 00:09:07,000 --> 00:09:21,000 So we've talked about augmenting. So the value that can be had. And that's the reason why you go through the rigmarole of getting the data sets and publishing them on the data market. 69 00:09:21,000 --> 00:09:34,000 And that's the value that you're paying for in the data market. And so then once you have those time for some insights, time to visualize the data. 70 00:09:34,000 --> 00:09:40,000 So we'll just look at some of these useful plots. Want to give a shout out to the Boston Python group. 71 00:09:40,000 --> 00:09:45,000 They can talk about these plots much better than I. 72 00:09:45,000 --> 00:09:52,000 So today's just going to be a rudimentary introduction, but we're just going to look at a few common plots here. 73 00:09:52,000 --> 00:10:07,000 Some box plots, some bubble plots and scatter plots, just to start adding some dimensionality to the data. 74 00:10:07,000 --> 00:10:09,000 Okay. 75 00:10:09,000 --> 00:10:18,000 So first things first, we're just going to read in some tools. 76 00:10:18,000 --> 00:10:26,000 And then what you can do is go ahead and define your plot style at the top. 77 00:10:26,000 --> 00:10:35,000 And so a pro tip, check out this. You can just print out all of the parameters that you can find with Matplotlib. 78 00:10:35,000 --> 00:10:44,000 So I was just exploring these and there is just a world that you can do with these. 79 00:10:44,000 --> 00:10:53,000 So now let's start exploring some of the data here. So with this data set, you can pick and choose the variables that you need. 80 00:10:53,000 --> 00:11:02,000 So here I've defined a lot of the variables and we'll mainly just be looking at cannabinoids today, 81 00:11:02,000 --> 00:11:09,000 but you can look through all of the solvents, the microbes, the micro toxins. 82 00:11:09,000 --> 00:11:14,000 So we can go ahead and read in the lab results here. 83 00:11:14,000 --> 00:11:20,000 And then the name of the game today is data augmentation. 84 00:11:20,000 --> 00:11:35,000 So what better than to just go ahead and combine both of these data sets here on the data market? 85 00:11:35,000 --> 00:11:44,000 So we're just going to combine the lab results with the geocoded licensee data. 86 00:11:44,000 --> 00:11:50,000 And all of these scripts are just to save us the trouble. 87 00:11:50,000 --> 00:11:55,000 I've gone ahead and gotten a copy of the licensees data. 88 00:11:55,000 --> 00:12:00,000 And so we can just read that in with the latitude and longitude data. 89 00:12:00,000 --> 00:12:08,000 So that'll save you, you know, an hour's worth of work, hopefully. 90 00:12:08,000 --> 00:12:17,000 We can just go ahead and read in the licensees data here. 91 00:12:17,000 --> 00:12:21,000 We can see how many. 92 00:12:21,000 --> 00:12:36,000 We now have about 400,000 geocoded lab results, and those are geocoded by the producer that ordered the lab test. 93 00:12:36,000 --> 00:12:39,000 So now we can start looking at more dimensions here. 94 00:12:39,000 --> 00:12:41,000 Just to look at some of these things. 95 00:12:41,000 --> 00:12:43,000 So this is going to take a hot second. 96 00:12:43,000 --> 00:13:11,000 So while I start calculating the averages here, does anyone have any data points that they're interested in looking at with producers or processors? 97 00:13:11,000 --> 00:13:27,000 Like, for example, Jerry, like, what are some of your, you know, KPIs that you're going to be using to determine, you know, if your production is going well in New York? 98 00:13:27,000 --> 00:13:30,000 That's not my job. 99 00:13:30,000 --> 00:13:31,000 Oh, not your job. 100 00:13:31,000 --> 00:13:32,000 OK. 101 00:13:32,000 --> 00:13:34,000 But maybe it will be in the future. 102 00:13:34,000 --> 00:13:36,000 So I'm just here to see what's possible. 103 00:13:36,000 --> 00:13:37,000 OK. 104 00:13:37,000 --> 00:13:45,000 Well, some of just the easy low hanging fruits are the ones we're calculating here. 105 00:13:45,000 --> 00:13:54,000 So for each licensee, just going to see how many total samples each licensee is testing that may be indicative of their size. 106 00:13:54,000 --> 00:13:59,000 Also going to look at the average concentration of their products. 107 00:13:59,000 --> 00:14:07,000 This data then is coming from production facilities where they're testing the potency. 108 00:14:07,000 --> 00:14:08,000 Is that where we're looking at? 109 00:14:08,000 --> 00:14:09,000 Exactly. 110 00:14:09,000 --> 00:14:17,000 And so that would be important to know to the three production facilities that are coming in. 111 00:14:17,000 --> 00:14:18,000 Exactly. 112 00:14:18,000 --> 00:14:29,000 And so that's where we began to look at last week. And we started to notice that, OK, you may be able to hit certain average concentrations with various methods. 113 00:14:29,000 --> 00:14:42,000 And so, OK, once you've got your extraction technique dialed in, it's time to start worrying about variance because, OK, yes, you may be able to hit a consistent product. 114 00:14:42,000 --> 00:14:49,000 But can you or you may be able to hit a benchmark, but can you hit it consistently? 115 00:14:49,000 --> 00:14:59,000 And so we figure that, OK, variance is another metric that especially as a processor you may be interested in. 116 00:14:59,000 --> 00:15:06,000 So these are just three simple statistics that you can just calculate right out of the gate. 117 00:15:06,000 --> 00:15:13,000 So these will just start with something simple. 118 00:15:13,000 --> 00:15:23,000 Just going to grab the producers that actually tested samples because, as you can see, there's some facilities that aren't testing at the moment. 119 00:15:23,000 --> 00:15:29,000 This is a new plot that I've been trying to think of a few different ways you can use it. 120 00:15:29,000 --> 00:15:46,000 This wasn't what I set out to do, but I think it's an interesting plot because you can really see that, wow, there is a lot of samples, a lot of flower samples being tested in, what's this, July? 121 00:15:46,000 --> 00:15:49,000 August, yeah. 122 00:15:49,000 --> 00:15:59,000 Exactly. So it's actually, you know, that's when most of the flowers arriving in Washington state. 123 00:15:59,000 --> 00:16:12,000 And it's peculiar because you notice some things like you can see that, you know, January clearly has a below average concentration in new flower samples. 124 00:16:12,000 --> 00:16:22,000 So whatever is happening, people just maybe just harvesting at slightly lower quality in January. 125 00:16:22,000 --> 00:16:27,000 Have you looked at indoor production facilities? 126 00:16:27,000 --> 00:16:30,000 Because there's a lot of that beginning to happen. 127 00:16:30,000 --> 00:16:34,000 We just have, you know, cultivation or the processor. 128 00:16:34,000 --> 00:16:46,000 So long story short, you can, you know, substitute this out, you know, you could look at various, you know, various other types. 129 00:16:46,000 --> 00:16:52,000 Another plot I wanted to show you, which this one didn't quite turn out as I was expecting. 130 00:16:52,000 --> 00:17:04,000 I was expecting to see a bit more of a correlation between average concentration and total samples or variance. 131 00:17:04,000 --> 00:17:13,000 And so I'll plot both of these with, you know, variance and total samples in both places. 132 00:17:13,000 --> 00:17:40,000 As you can see, there's not, not, you know, a ready apparent correlation here between, you know, concentration and total samples or average concentration and variance. 133 00:17:40,000 --> 00:17:49,000 So I think these, I think these bubble plots have their value. 134 00:17:49,000 --> 00:17:56,000 I just don't think I encapsulated, you know, the perfect bubble plot here. 135 00:17:56,000 --> 00:18:01,000 So, you know, I think there's there's more work to be done here. 136 00:18:01,000 --> 00:18:18,000 So long story short, I would, I would think you could still do some exploratory analysis here with with these scatter plots slash bubble plots and really try to think of something cool. 137 00:18:18,000 --> 00:18:23,000 And then last but not least, 138 00:18:23,000 --> 00:18:35,000 I want to show you that this one didn't turn out as expected, but there's a technique where you can use a scatter plot to plot the latitude and longitude. 139 00:18:35,000 --> 00:18:47,000 And so this one, for whatever reason, I think there's some oddball miscoding in here, but this one doesn't quite create the visualization that's expected. 140 00:18:47,000 --> 00:18:58,000 So long story short, there's room to expand with longitude and latitude plots. 141 00:18:58,000 --> 00:19:01,000 But 142 00:19:01,000 --> 00:19:21,000 that was sort of the main things that I wanted to hit today was, you know, wanted to show you the box plot, the bubble plot in the scatter plot, just to let you know that those are ways that you can show many dimensions in your data. 143 00:19:21,000 --> 00:19:34,000 And that was sort of what today was all about was, okay, you know, can we augment and then visualize data. And so after augmentation, you have many data points. 144 00:19:34,000 --> 00:19:56,000 So these are useful plots that you can use to create a two dimensional plot, which is easy, readily understood, they shows you three or four dimensions. So you're able to show three or four dimensions of data in two dimensions. 145 00:19:56,000 --> 00:20:05,000 So, so, and then this this plot, no good, this needs to get. I've done things that look like that. 146 00:20:05,000 --> 00:20:20,000 Exactly. So something's going on there, but that's maybe not the right plot for the job. So, for long story short, we've done some sort of ad hoc work here. 147 00:20:20,000 --> 00:20:32,000 You know, sort of some just exploratory analysis and really just what we stumbled upon was really the only thing where we can really 148 00:20:32,000 --> 00:20:53,000 tell a good story is, you know, just looking at this box plot of total of average k-n-b-n-o-y concentrations by the month. So I think that was sort of the most compelling figure of the day. 149 00:20:53,000 --> 00:21:04,000 And so, you know, I think that's sort of what we learned from this data, right? So we got a bunch of different data points, right? So we got a bunch. 150 00:21:04,000 --> 00:21:10,000 Start opening this up to a bit more of a discussion since having a bit of an Internet issue here. 151 00:21:10,000 --> 00:21:24,000 But I was basically just saying we've done our augmentation. We've gotten our variables of interest from the lab results. We've got our variables of interest from the licensees. 152 00:21:24,000 --> 00:21:40,000 We started playing around with some various plots and we calculated just some simple statistics, right? Total, mean, standard deviation, variance. 153 00:21:40,000 --> 00:22:05,000 So these are quite simple statistics and we were just able to, you know, you saw we tried a bunch of plots. We weren't really able to tell much of a story with any of the plots, except for this box plot here, which tells the most enlightening story, which is that there appears to be some cyclic. 154 00:22:05,000 --> 00:22:16,000 We were just able to, you know, just use a subset of the lab results and a subset of the licensee results. 155 00:22:16,000 --> 00:22:29,000 And with a simple box plot, we were able to tell, you know, an interesting story here that is hopefully insightful to people. 156 00:22:29,000 --> 00:22:38,000 So that's sort of the lesson of the day. Got to have the right tools for the job. That includes the right data for the job. 157 00:22:38,000 --> 00:22:49,000 And a little data augmentation can go a long way. It's the same for a visualization, right? A little data augmentation and visualization can go along. 158 00:22:49,000 --> 00:22:57,000 Awesome. Well, just like to give one last final shout out here to Paul. 159 00:22:57,000 --> 00:23:04,000 Can you talk a little bit more about what the data market is and how we get you paid in it? 160 00:23:04,000 --> 00:23:13,000 Exactly. So the idea is to just have a marketplace for buyers and sellers to come and sell data. 161 00:23:13,000 --> 00:23:27,000 So if you've got an interesting cannabis data market, I mean, if you've got an interesting cannabis data set that you want to list on the market, then you can put it up for sale. 162 00:23:27,000 --> 00:23:36,000 And anyone who has a demand for that data set can buy it. And, you know, I'm just taking advantage of the marketplace. 163 00:23:36,000 --> 00:23:46,000 So I just published these first two data sets. And so these are just the data sets that we use today. 164 00:23:46,000 --> 00:23:57,000 So, you know, today I basically just got a copy of the lab results and the licensees data. 165 00:23:57,000 --> 00:24:08,000 And, you know, this comes. So what is for sale on the marketplace is the data itself or the reports generated from the data? 166 00:24:08,000 --> 00:24:15,000 It is the data itself. So essentially. 167 00:24:15,000 --> 00:24:28,000 We went through this rigmarole here of downloading the data, which is quite large. So combined. 168 00:24:28,000 --> 00:24:37,000 All of the data is around. So it's not really open source, then it's proprietary. Well. 169 00:24:37,000 --> 00:24:48,000 Open source. There's a saying that, you know, it's free as in freedom, not as in free lunch. 170 00:24:48,000 --> 00:24:54,000 So once you get a copy of the data, you're free to do with it whatever you'd like. 171 00:24:54,000 --> 00:24:59,000 So as long as you, you know. 172 00:24:59,000 --> 00:25:02,000 You know, reference Ken Liddick says you're. 173 00:25:02,000 --> 00:25:07,000 As a source, as a source, basically. Exactly. 174 00:25:07,000 --> 00:25:10,000 And so you're free to use the data however you please. 175 00:25:10,000 --> 00:25:17,000 So it's essentially, you know, open data in that, you know, you you can get it and there's no strings attached. 176 00:25:17,000 --> 00:25:22,000 And so the idea is. 177 00:25:22,000 --> 00:25:28,000 Putting it on the data market will drive the price towards its fixed cost. 178 00:25:28,000 --> 00:25:38,000 So, you know, there is just a fixed cost involved of, you know, downloading the data and processing it. 179 00:25:38,000 --> 00:25:41,000 And, you know, that's all done with this script here. 180 00:25:41,000 --> 00:25:45,000 And so, you know, you're welcome to just do this yourself. 181 00:25:45,000 --> 00:25:54,000 And I just hypothesize that I think, you know, that amount of toil would be about, you know, 182 00:25:54,000 --> 00:26:01,000 about $500 a toil for a business and about $900 a toil for a student. 183 00:26:01,000 --> 00:26:04,000 So if you're up to. 184 00:26:04,000 --> 00:26:11,000 Up to process the data yourself, you know, you're you're more than welcome to. 185 00:26:11,000 --> 00:26:15,000 So all the the algorithms are open source. 186 00:26:15,000 --> 00:26:20,000 And the idea is the price will go towards fixed costs. 187 00:26:20,000 --> 00:26:26,000 So that way, you know, you would actually benefit if you're. 188 00:26:26,000 --> 00:26:34,000 In one of these groups to just buy the data because it'll be cheaper to buy than to process it yourself. 189 00:26:34,000 --> 00:26:43,000 Or if you see an arbitrage opportunity, then you can buy the data and resell it at a slightly lower price. 190 00:26:43,000 --> 00:26:50,000 And the idea is, if you know the demand better, then you can just drive the price down. 191 00:26:50,000 --> 00:26:57,000 And it's better for both the suppliers and the producers because I mean, it's better for the suppliers and consumers 192 00:26:57,000 --> 00:27:05,000 because the consumers, the more consumers there are, the lower the average fixed cost. 193 00:27:05,000 --> 00:27:18,000 So the more consumers that use the data marketplace, the lower and lower want to get a buck for their toil of processing data. 194 00:27:18,000 --> 00:27:26,000 Because that's sort of the idea is, you know, this algorithm provides value. 195 00:27:26,000 --> 00:27:36,000 So, you know, it's a non-trivial thing to simply run the algorithm because you have to download the data 196 00:27:36,000 --> 00:27:41,000 and you have to commit the resources of your hard drive. 197 00:27:41,000 --> 00:27:51,000 So if you're up to do that, then by all means, if you just want to get the data, then you can simply buy it. 198 00:27:51,000 --> 00:27:53,000 Does that answer your question, Jerry? 199 00:27:53,000 --> 00:27:55,000 Pretty much. 200 00:27:55,000 --> 00:27:56,000 Exactly. 201 00:27:56,000 --> 00:28:00,000 And, you know, the idea is, you know, this is just here for anyone. 202 00:28:00,000 --> 00:28:12,000 So if you're a data publisher, you can try to, you know, price your data accordingly and get that into the hands of consumers. 203 00:28:12,000 --> 00:28:28,000 Pretty much the data has been gathered from Washington State source and then just cleaned up and put into a usable form or a standard form. 204 00:28:28,000 --> 00:28:33,000 Exactly. And that's sort of where the value has been added. 205 00:28:33,000 --> 00:28:45,000 So if you just look at the raw data, it's just in these giant zip files. 206 00:28:45,000 --> 00:28:47,000 So you have to zip them and look at them. 207 00:28:47,000 --> 00:28:48,000 Exactly. 208 00:28:48,000 --> 00:28:51,000 So there's a lot of data massaging that's going on. 209 00:28:51,000 --> 00:28:58,000 Exactly. You've got to substantially massage about 50 gigabytes of data. 210 00:28:58,000 --> 00:29:01,000 And so I can do that overnight. 211 00:29:01,000 --> 00:29:05,000 Oh, yes. And so that's why it really... 212 00:29:05,000 --> 00:29:07,000 It's a lot of work. It's a lot of work. 213 00:29:07,000 --> 00:29:11,000 But it matches the, you know, the consumer with the supplier. 214 00:29:11,000 --> 00:29:16,000 So that's what makes it so efficient is, like you said, you can do that overnight. 215 00:29:16,000 --> 00:29:21,000 So if you can do that at a low cost, then you can do that at a... 216 00:29:21,000 --> 00:29:26,000 It's a time-intentioned process. So you want to be paid for it. I got you. 217 00:29:26,000 --> 00:29:29,000 Oh, yes. But that's sort of the brilliance of it. 218 00:29:29,000 --> 00:29:38,000 So, you know, if you, like, for example, if you want to process the data and then list it, then, you know, you can do so. 219 00:29:38,000 --> 00:29:46,000 And then you can also do so in novel ways. So I've just listed the lab results. 220 00:29:46,000 --> 00:29:57,000 But if you wanted to, say, process all of the sales data and augment the sales data, then that would actually probably... 221 00:29:57,000 --> 00:29:59,000 Something I could put on the marketplace then. 222 00:29:59,000 --> 00:30:09,000 Exactly. And it may be more...the sales data would, you would imagine, would be more valuable than just the lab results data. 223 00:30:09,000 --> 00:30:17,000 For a manufacturer or producer, I would want to see the lab results. And I pay a pretty price for it. 224 00:30:17,000 --> 00:30:19,000 Exactly. And so if you know... 225 00:30:19,000 --> 00:30:27,000 But just to get the data, unless they have the ability to do the analytics on the data, it's not very meaningful. 226 00:30:27,000 --> 00:30:32,000 So I could have them buy the data and hire me to do the analytics for them. 227 00:30:32,000 --> 00:30:40,000 Exactly. And so feel free to try to think of some data sets that you may want to publish. 228 00:30:40,000 --> 00:30:49,000 And, you know, I'll make that process easy for you. That way you can get paid for your data cleaning. 229 00:30:49,000 --> 00:31:00,000 Awesome. Well, Ami, just go ahead and stop presenting for now. 230 00:31:00,000 --> 00:31:10,000 Are there any final questions or comments for the day? 231 00:31:10,000 --> 00:31:22,000 All right. Well, for next week, I'm going to get this Internet situation under control and may even try to do something special midway. 232 00:31:22,000 --> 00:31:24,000 I think Discord might be better. 233 00:31:24,000 --> 00:31:28,000 Discord. OK. I'll try some of the... 234 00:31:28,000 --> 00:31:35,000 I've used Discord. I've used Slack. You know, of course, Zoom works better, but you got to pay for it. 235 00:31:35,000 --> 00:31:42,000 OK. I'll try some of these platforms out and then I'll settle on something that works. 236 00:31:42,000 --> 00:31:49,000 And like I said, if I get this figured out, then maybe we can do something special later on early or mid week. 237 00:31:49,000 --> 00:31:59,000 So keep an eye out for a message from me and I'll try to make it up to you since today was kind of cut short and a little choppy. 238 00:31:59,000 --> 00:32:07,000 So I'll try to make it up one way or the other. Thank you. Awesome, Jerry. OK. M&R. All right. Thank you. 239 00:32:07,000 --> 00:32:14,000 I'm going to wrap it up for today, but it was awesome having you both. So thank you for coming to the Cannabis Data Science Meetup Group. 240 00:32:14,000 --> 00:32:19,000 See you again. All right. See you next week.