1 00:00:00,000 --> 00:00:11,200 So is this your new place? 2 00:00:11,200 --> 00:00:15,560 Well I am actually visiting Oklahoma City. 3 00:00:15,560 --> 00:00:17,600 CannaCon is this week. 4 00:00:17,600 --> 00:00:27,120 So tomorrow, Thursday, May 27th and then Friday May 28th there's CannaCon in Oklahoma. 5 00:00:27,120 --> 00:00:36,320 So I thought it was a good opportunity to see how this market is developing, meet some 6 00:00:36,320 --> 00:00:47,880 people in Oklahoma and try to talk with some laboratories here. 7 00:00:47,880 --> 00:00:48,880 Good morning Jake. 8 00:00:48,880 --> 00:00:49,880 Hey how's it going? 9 00:00:49,880 --> 00:00:54,920 Welcome to the Canva Data Science Meetup Group. 10 00:00:54,920 --> 00:00:55,920 Nice to meet you both. 11 00:00:55,920 --> 00:00:58,400 So it's great to be here. 12 00:00:58,400 --> 00:01:02,440 Well just to introduce myself, so my name is Keegan. 13 00:01:02,440 --> 00:01:08,880 My background, I started working as an analyst at a cannabis testing laboratory and then 14 00:01:08,880 --> 00:01:11,800 I found there were a lot of rooms for improvement. 15 00:01:11,800 --> 00:01:17,720 So for example there's data entry that needed to be automated, certificates needed to be 16 00:01:17,720 --> 00:01:24,320 created easy and then just a website and client portal. 17 00:01:24,320 --> 00:01:32,240 So as a software developer I realized there's a lot of laboratories that need software solutions 18 00:01:32,240 --> 00:01:38,600 so I launched CannaLytics, a company to help provide software solutions to laboratories 19 00:01:38,600 --> 00:01:40,920 to make cannabis testing simple and easy. 20 00:01:40,920 --> 00:01:46,960 Because I've got a lot of knowledge so I may as well share with you. 21 00:01:46,960 --> 00:01:55,080 I guess I'll let Charles and yourself introduce yourselves. 22 00:01:55,080 --> 00:01:58,680 Charles I think you were here first, go ahead if you want to. 23 00:01:58,680 --> 00:02:06,920 So yeah I have like 27 years of software development experience and I'm moving into like data science 24 00:02:06,920 --> 00:02:10,400 and machine learning now. 25 00:02:10,400 --> 00:02:19,240 And this group has some really good data to explore and do some interesting things with. 26 00:02:19,240 --> 00:02:25,560 And so I've been working on that, I've been working on some Kaggle competitions. 27 00:02:25,560 --> 00:02:28,560 Oh cool, that's awesome. 28 00:02:28,560 --> 00:02:37,160 How about yourself Jake, how did you hear about the group and... 29 00:02:37,160 --> 00:02:40,280 So I guess it's kind of two factors. 30 00:02:40,280 --> 00:02:52,120 So I've been interested in cannabis, I do like stock analytics data, like investing analytics. 31 00:02:52,120 --> 00:02:56,260 And cannabis is not an industry I'm current with, it's one of two industries I'm really 32 00:02:56,260 --> 00:02:57,260 looking at. 33 00:02:57,260 --> 00:03:00,920 But I'm also just interested in data science generally, so like that piece. 34 00:03:00,920 --> 00:03:08,200 And then the second thing is what I do most of my time is I'm building this data science 35 00:03:08,200 --> 00:03:15,600 tool and I'm sort of trying to explore what industries it could be interesting in. 36 00:03:15,600 --> 00:03:20,240 And we've had like one or two people that I've just met with recently who were doing 37 00:03:20,240 --> 00:03:24,520 like analytics in the cannabis or at cannabis companies I believe. 38 00:03:24,520 --> 00:03:28,520 And so I thought you know maybe this would be cool people to meet here in that space 39 00:03:28,520 --> 00:03:29,520 as well. 40 00:03:29,520 --> 00:03:35,200 That's awesome, if you're open to talk about it I would love to hear about the data science 41 00:03:35,200 --> 00:03:36,560 tool you're building. 42 00:03:36,560 --> 00:03:41,560 But if you want, if you guys want to try it out I could... 43 00:03:41,560 --> 00:03:42,560 It's a... 44 00:03:42,560 --> 00:03:43,560 If you want to... 45 00:03:43,560 --> 00:03:48,200 Well honestly today was going to be a bit more of an open day, so I'm traveling to Oklahoma 46 00:03:48,200 --> 00:03:50,840 City to visit Canacon. 47 00:03:50,840 --> 00:03:57,080 So I haven't had as much time to prepare things this week as I normally do, so the whole point 48 00:03:57,080 --> 00:03:59,760 of the group is to let everybody have a chance to share. 49 00:03:59,760 --> 00:04:02,680 So yes, if you want, it would be... 50 00:04:02,680 --> 00:04:06,720 Yeah why don't I... what I'll do here, I'll... this is a... 51 00:04:06,720 --> 00:04:08,400 That'd be really cool actually. 52 00:04:08,400 --> 00:04:09,400 Why don't I... 53 00:04:09,400 --> 00:04:11,240 I'll send you both a link. 54 00:04:11,240 --> 00:04:18,720 So I have a hosted version and a local version, but as I'm sure you're aware like downloading... 55 00:04:18,720 --> 00:04:23,960 It's a Python tool, so like downloading Python packages is a brutal, brutal process. 56 00:04:23,960 --> 00:04:31,160 So I'll send you both this posted version. 57 00:04:31,160 --> 00:04:33,520 And if you want to... 58 00:04:33,520 --> 00:04:35,000 You can just log in there. 59 00:04:35,000 --> 00:04:36,000 Oh do you... 60 00:04:36,000 --> 00:04:37,000 Are you still there? 61 00:04:37,000 --> 00:04:38,000 I think you might have froze. 62 00:04:38,000 --> 00:04:45,560 Sorry, I just had to get reconnected real quick. 63 00:04:45,560 --> 00:04:46,560 Oh no worries. 64 00:04:46,560 --> 00:04:53,600 So if you go to that link that I just sent, it's a Jupiter Hub link. 65 00:04:53,600 --> 00:04:55,080 So if you essentially... 66 00:04:55,080 --> 00:04:59,840 It's like a hosted Jupiter notebook essentially, or Jupiter lab really. 67 00:04:59,840 --> 00:05:03,360 You have to put your email in to access it, but you're welcome to put in like a fake email 68 00:05:03,360 --> 00:05:05,680 if you don't want to put in a real email. 69 00:05:05,680 --> 00:05:06,680 I understand. 70 00:05:06,680 --> 00:05:11,480 Oh sweet, yeah I can sort of guide you through. 71 00:05:11,480 --> 00:05:16,240 Yes, so I'm just going to share my screen that way. 72 00:05:16,240 --> 00:05:19,160 So just a heads up, so we were recording to that way people... 73 00:05:19,160 --> 00:05:21,680 Oh yeah, no worries at all. 74 00:05:21,680 --> 00:05:22,680 Totally cool. 75 00:05:22,680 --> 00:05:25,480 If anything, if more people get to see it, that's cool. 76 00:05:25,480 --> 00:05:29,320 How long have you been running this group? 77 00:05:29,320 --> 00:05:36,720 So the data science group kicked off in February this year, so I think our first meetup was 78 00:05:36,720 --> 00:05:41,840 February 24th or 26th. 79 00:05:41,840 --> 00:05:42,840 Okay cool. 80 00:05:42,840 --> 00:05:43,840 And... 81 00:05:43,840 --> 00:05:44,840 Oh interesting, so... 82 00:05:44,840 --> 00:05:51,560 So I'll give a little background. 83 00:05:51,560 --> 00:05:54,880 So what it is, it's a visual... 84 00:05:54,880 --> 00:06:01,160 It's supposed to be like a visual Python representation and sort of visualizing Python as a spreadsheet. 85 00:06:01,160 --> 00:06:05,720 So we'd imagine everything you'll do in the spreadsheet will generate the equivalent Python 86 00:06:05,720 --> 00:06:06,720 for you. 87 00:06:06,720 --> 00:06:12,440 So if you run this cell right here, that first cell... 88 00:06:12,440 --> 00:06:17,040 Yeah you can go from there. 89 00:06:17,040 --> 00:06:19,240 That should do it. 90 00:06:19,240 --> 00:06:23,240 You're going to have to put your email in again. 91 00:06:23,240 --> 00:06:34,080 But I sort of just knew I could just do this to hopefully be able to get in touch with 92 00:06:34,080 --> 00:06:35,080 people. 93 00:06:35,080 --> 00:06:36,080 Just formally? 94 00:06:36,080 --> 00:06:39,080 Yeah, yeah, you're welcome. 95 00:06:39,080 --> 00:06:44,840 And we're not taking anything. 96 00:06:44,840 --> 00:06:46,720 You can just put like NA here. 97 00:06:46,720 --> 00:06:47,720 This is... 98 00:06:47,720 --> 00:06:50,960 I kind of have this question here for like self-service users, but obviously since we're 99 00:06:50,960 --> 00:06:56,920 on a call, you're welcome to do that. 100 00:06:56,920 --> 00:06:57,920 Cool. 101 00:06:57,920 --> 00:06:58,920 Oh sweet. 102 00:06:58,920 --> 00:07:08,640 I didn't know you could do that. 103 00:07:08,640 --> 00:07:09,640 So this is the tool. 104 00:07:09,640 --> 00:07:10,640 So essentially... 105 00:07:10,640 --> 00:07:14,960 So are you a Python user at all? 106 00:07:14,960 --> 00:07:15,960 Oh definitely. 107 00:07:15,960 --> 00:07:19,720 Both Charles and I love Python. 108 00:07:19,720 --> 00:07:25,320 So I fell in love with Python five, six years ago. 109 00:07:25,320 --> 00:07:29,000 And it's my weapon of choice. 110 00:07:29,000 --> 00:07:31,400 And so I'm fascinated by this. 111 00:07:31,400 --> 00:07:33,600 So I've done a little work. 112 00:07:33,600 --> 00:07:40,960 So this looks almost like something you would build like with PyQt, but obviously it's running 113 00:07:40,960 --> 00:07:42,400 here in the browser. 114 00:07:42,400 --> 00:07:47,520 So honestly, I'm kind of curious how you built this, if that's something you're willing to 115 00:07:47,520 --> 00:07:50,280 share, but it also looks like a cool tool. 116 00:07:50,280 --> 00:08:02,200 Yeah I have a co-creator, who I made this with, especially my twin brother. 117 00:08:02,200 --> 00:08:07,660 And so essentially it's a Python backend. 118 00:08:07,660 --> 00:08:10,920 So let me actually, it's easier to explain once I show you a little bit of the functionality. 119 00:08:10,920 --> 00:08:12,760 So let's import a data set really quick. 120 00:08:12,760 --> 00:08:22,360 You can click the import button and then you can add files from your folder. 121 00:08:22,360 --> 00:08:28,600 Because you're on the hosted version, you have to upload your file to the system first 122 00:08:28,600 --> 00:08:31,400 and then you can import it into the sheet with that button we just clicked. 123 00:08:31,400 --> 00:08:36,400 If you're on a local version, then you can just sort of, you're in your local files, 124 00:08:36,400 --> 00:08:43,280 you can connect to whatever files you want. 125 00:08:43,280 --> 00:08:46,960 But if you want to go back and click that button again, so it is add files from current 126 00:08:46,960 --> 00:08:54,040 folder, that's just going to put in our sort of the demo data airport pets. 127 00:08:54,040 --> 00:08:57,560 And I'll just show you, so let's do like, you can close that little window on the right 128 00:08:57,560 --> 00:08:58,560 there. 129 00:08:58,560 --> 00:09:03,120 The one that you're, the one you had the import window, you can close that. 130 00:09:03,120 --> 00:09:04,120 This one? 131 00:09:04,120 --> 00:09:05,800 No, the, if you scroll up. 132 00:09:05,800 --> 00:09:07,800 Yeah, it works as import CSV. 133 00:09:07,800 --> 00:09:09,000 You can close that. 134 00:09:09,000 --> 00:09:13,320 And then now we, so essentially the gist of it is it's a spreadsheet and everything you 135 00:09:13,320 --> 00:09:16,680 do in the spreadsheet is going to generate the equivalent code for you. 136 00:09:16,680 --> 00:09:23,480 So if you filter, maybe, so apply, try applying a filter to a column. 137 00:09:23,480 --> 00:09:29,560 Like is it possible to import our own data set here real quick? 138 00:09:29,560 --> 00:09:30,560 Yeah, yeah, totally. 139 00:09:30,560 --> 00:09:31,560 For sure. 140 00:09:31,560 --> 00:09:34,120 Also all you have to, so one of two ways to do it. 141 00:09:34,120 --> 00:09:37,760 One is you could, if you have it as like a CSV file, you could just upload it in here. 142 00:09:37,760 --> 00:09:38,760 Okay, sweet. 143 00:09:38,760 --> 00:09:39,760 So yeah. 144 00:09:39,760 --> 00:09:46,360 Because essentially, you know, just put together some data here. 145 00:09:46,360 --> 00:09:48,360 Yeah, totally, totally. 146 00:09:48,360 --> 00:09:53,760 So I think the best way, did you click the upload button yet or are you just in your 147 00:09:53,760 --> 00:09:54,760 local file? 148 00:09:54,760 --> 00:09:58,760 Let me just get some data for you real quick. 149 00:09:58,760 --> 00:10:01,760 Yeah, for sure. 150 00:10:01,760 --> 00:10:02,760 No. 151 00:10:02,760 --> 00:10:05,800 And then how do we upload data? 152 00:10:05,800 --> 00:10:07,440 So yeah, so you want, yeah, exactly. 153 00:10:07,440 --> 00:10:18,680 Click that button there, select your data. 154 00:10:18,680 --> 00:10:22,000 And then now, so if we, let's rerun that top. 155 00:10:22,000 --> 00:10:24,520 Oh, was that a, did you upload it as a CSV? 156 00:10:24,520 --> 00:10:27,840 It was an Excel, does it need to be a CSV? 157 00:10:27,840 --> 00:10:28,840 Yes. 158 00:10:28,840 --> 00:10:32,000 I mean, it doesn't have to be, but if you want to do it sort of through the UI, it has 159 00:10:32,000 --> 00:10:33,000 to be a CSV. 160 00:10:33,000 --> 00:10:37,600 You can, the other thing you can do is you can import any data frame into the Mito just 161 00:10:37,600 --> 00:10:40,800 by putting it into the argument to the Mito sheet dot sheet call. 162 00:10:40,800 --> 00:10:44,080 So for example, if you're working in a notebook and you have a data frame you're working with 163 00:10:44,080 --> 00:10:48,840 and you want to quickly call those in, you can do that. 164 00:10:48,840 --> 00:10:50,960 Cool. 165 00:10:50,960 --> 00:10:58,000 So now that top cell in the notebook where we call the Mito sheet, if you want to run 166 00:10:58,000 --> 00:11:05,880 that again and then click the import button, you should be able to select your file. 167 00:11:05,880 --> 00:11:07,920 But I think this is. 168 00:11:07,920 --> 00:11:09,440 That's yeah, that's the wrong data. 169 00:11:09,440 --> 00:11:13,760 But if you click, there you go. 170 00:11:13,760 --> 00:11:14,760 You can close that. 171 00:11:14,760 --> 00:11:16,760 It's a little tutorial. 172 00:11:16,760 --> 00:11:17,760 Interesting. 173 00:11:17,760 --> 00:11:18,760 So. 174 00:11:18,760 --> 00:11:27,680 And if you want to build that original data. 175 00:11:27,680 --> 00:11:37,120 Yeah, you can see this little drop down next to where it says airport pets at the bottom. 176 00:11:37,120 --> 00:11:38,120 You can just click delete. 177 00:11:38,120 --> 00:11:43,920 Because yes, we're all about, you know, working with actual data here. 178 00:11:43,920 --> 00:11:49,080 So we just have a licensee data here. 179 00:11:49,080 --> 00:11:51,080 I'm tired of doing the same demo data. 180 00:11:51,080 --> 00:11:52,080 So this is great. 181 00:11:52,080 --> 00:11:56,440 But so yeah, let me, so let me show you maybe like a few things. 182 00:11:56,440 --> 00:11:59,960 And I just love to hear your thoughts or if you want to, if you have thoughts of how you 183 00:11:59,960 --> 00:12:02,160 might want to play with that, that'd be cool. 184 00:12:02,160 --> 00:12:05,400 For example, so let's do like a filter, for example. 185 00:12:05,400 --> 00:12:10,280 So if we go to the top or if we go to the actual the sheet, and then if you click, so 186 00:12:10,280 --> 00:12:14,960 each it's, you know, it's the same feel as a spreadsheet or it's, you know, aspiring 187 00:12:14,960 --> 00:12:15,960 to be that. 188 00:12:15,960 --> 00:12:19,600 So for every column, there's a little filter icon next to it. 189 00:12:19,600 --> 00:12:22,560 So if you want to filter. 190 00:12:22,560 --> 00:12:25,940 So like a good example here, yeah, we can add a filter. 191 00:12:25,940 --> 00:12:27,960 See we have that NaN value. 192 00:12:27,960 --> 00:12:28,960 This is kind of useful. 193 00:12:28,960 --> 00:12:32,800 If you want one of the filter conditions is you can do, yeah, you can filter to a city 194 00:12:32,800 --> 00:12:38,400 or you could filter like out null values by doing is not empty. 195 00:12:38,400 --> 00:12:41,560 The last option. 196 00:12:41,560 --> 00:12:46,860 You got rid of your null values. 197 00:12:46,860 --> 00:12:54,600 And then if you scroll down to below the, yeah, let's see. 198 00:12:54,600 --> 00:12:56,280 So this is code we're generating. 199 00:12:56,280 --> 00:13:01,520 So that final step, step four, that is the equivalent code for dropping null values from 200 00:13:01,520 --> 00:13:02,520 the column. 201 00:13:02,520 --> 00:13:03,520 Yeah. 202 00:13:03,520 --> 00:13:07,800 So we generate the equivalent code for each edit. 203 00:13:07,800 --> 00:13:11,120 And it kind of goes back to why I made the tool, which is like, I think you mentioned 204 00:13:11,120 --> 00:13:15,000 like if you was a five or six years ago, you started learning Python, like I started doing 205 00:13:15,000 --> 00:13:17,560 Python maybe like two years ago. 206 00:13:17,560 --> 00:13:23,480 And I just found that like memorizing the syntax and getting the syntax right all the time 207 00:13:23,480 --> 00:13:27,480 was just like was such a maybe it just wasn't my skill set, but it was just such a learning 208 00:13:27,480 --> 00:13:30,200 curve for me. 209 00:13:30,200 --> 00:13:37,600 So I kind of wanted with my brother like make a way to make Python more accessible essentially. 210 00:13:37,600 --> 00:13:44,600 Question, do we need mydosheader or is this just all on pandas? 211 00:13:44,600 --> 00:13:45,600 So that's a good question. 212 00:13:45,600 --> 00:13:48,960 So everything you have right there is pandas. 213 00:13:48,960 --> 00:13:51,280 Where you will need mydosheader is one specific place. 214 00:13:51,280 --> 00:13:54,920 And it's the reason we have that. 215 00:13:54,920 --> 00:14:00,760 So if you scroll up, if you do so what you can actually you can do like spreadsheet formulas 216 00:14:00,760 --> 00:14:01,760 in the sheet. 217 00:14:01,760 --> 00:14:17,360 So like if you want to try adding a column, for example, you can just click add column 218 00:14:17,360 --> 00:14:30,920 button in the toolbar at the top. 219 00:14:30,920 --> 00:14:34,920 And then in there you could use a form do a formula that's like normal spreadsheet syntax. 220 00:14:34,920 --> 00:14:39,440 So if you want to use like a concat function that you would do in Excel, you could do it 221 00:14:39,440 --> 00:14:45,000 in there and sort of concat the two. 222 00:14:45,000 --> 00:14:59,040 So just write concat and you'll see a suggestion pop up. 223 00:14:59,040 --> 00:15:02,960 And then you can select the columns you want. 224 00:15:02,960 --> 00:15:07,720 Ideally I'm trying to like let's say. 225 00:15:07,720 --> 00:15:10,840 Oh, yeah, pretty soon. 226 00:15:10,840 --> 00:15:13,920 I don't know why it wasn't working like that before. 227 00:15:13,920 --> 00:15:17,480 Maybe if you try, if you close that menu on the side, I think it might make it a little 228 00:15:17,480 --> 00:15:18,480 easier. 229 00:15:18,480 --> 00:15:21,680 That A where it says AB, if you close that. 230 00:15:21,680 --> 00:15:28,880 Yeah, but also, yeah, if you know the names of your columns, it will suggest it to you. 231 00:15:28,880 --> 00:15:36,720 We use this thing called Q grid, which is like a no, it's our AG grid, which is like 232 00:15:36,720 --> 00:15:42,640 a it's a JavaScript like table library essentially. 233 00:15:42,640 --> 00:15:46,920 And they kind of help us like get a lot of these features out of the box, which is cool. 234 00:15:46,920 --> 00:15:52,560 But yeah, so there's your. 235 00:15:52,560 --> 00:16:13,640 If you wanted to put like. 236 00:16:13,640 --> 00:16:17,600 And so, yeah, so it's actually kind of what you're doing right now is like formatting. 237 00:16:17,600 --> 00:16:21,360 It's like one of the main uses I think people use this for and then you can rename it. 238 00:16:21,360 --> 00:16:24,240 But then you to your question about like, can you just like copy this code into any 239 00:16:24,240 --> 00:16:26,700 other environment? 240 00:16:26,700 --> 00:16:29,520 You can't except for when you use something like. 241 00:16:29,520 --> 00:16:33,800 Where do we have it? 242 00:16:33,800 --> 00:16:34,800 Scroll down to the bottom. 243 00:16:34,800 --> 00:16:35,800 Exactly. 244 00:16:35,800 --> 00:16:40,400 But that function, it is pandas that we're using. 245 00:16:40,400 --> 00:16:46,000 But we it's like, we had to use a complex combination of pandas to like, get that to 246 00:16:46,000 --> 00:16:47,000 work. 247 00:16:47,000 --> 00:16:51,520 So we sort of wrapped it in our own function. 248 00:16:51,520 --> 00:16:53,240 But I thought I saw you opening spider. 249 00:16:53,240 --> 00:16:55,160 Were you about to try and copy this into spider? 250 00:16:55,160 --> 00:16:56,160 Yes, actually. 251 00:16:56,160 --> 00:16:57,160 Yes, that'd be cool. 252 00:16:57,160 --> 00:16:59,160 I've actually haven't done a ton of testing with that. 253 00:16:59,160 --> 00:17:01,200 So it'd be interesting to see how that goes. 254 00:17:01,200 --> 00:17:04,680 What I'm thinking is like the functions, you don't need the UI. 255 00:17:04,680 --> 00:17:07,920 The UI won't pop up in spider or at least it won't work well. 256 00:17:07,920 --> 00:17:14,640 But I'm thinking if you copy, if you also copy the Mito sheet package into spider, the 257 00:17:14,640 --> 00:17:16,240 syntax might actually work still. 258 00:17:16,240 --> 00:17:18,240 We haven't done a lot of testing on that. 259 00:17:18,240 --> 00:17:19,240 I'll need to. 260 00:17:19,240 --> 00:17:20,240 This your project. 261 00:17:20,240 --> 00:17:21,240 Yes. 262 00:17:21,240 --> 00:17:34,240 Are you looking to what are you looking to do in the local install? 263 00:17:34,240 --> 00:17:37,000 I believe so, right? 264 00:17:37,000 --> 00:17:39,520 Because I'll need Mito sheet to. 265 00:17:39,520 --> 00:17:41,480 Yeah, you will. 266 00:17:41,480 --> 00:17:42,480 So this is the thing. 267 00:17:42,480 --> 00:17:43,480 It's the local install. 268 00:17:43,480 --> 00:17:44,480 Yeah, it should. 269 00:17:44,480 --> 00:17:45,480 It should work. 270 00:17:45,480 --> 00:17:51,400 So you just it's so the thing, the way Mito works is a is a the back end and then there's 271 00:17:51,400 --> 00:17:54,160 a front end, which is a JupyterLab extension. 272 00:17:54,160 --> 00:17:58,320 But I think you probably if you actually don't think you'll have to even download the extension 273 00:17:58,320 --> 00:18:00,320 part, just do the just install the package. 274 00:18:00,320 --> 00:18:01,320 You should be fine. 275 00:18:01,320 --> 00:18:02,320 So yeah, that'd be cool if you want to try that. 276 00:18:02,320 --> 00:18:24,760 And I guess. 277 00:18:24,760 --> 00:18:34,800 It is your principal audiences or data scientists. 278 00:18:34,800 --> 00:18:37,000 I think it definitely is. 279 00:18:37,000 --> 00:18:41,640 I think I'm I'm definitely doing a lot of thinking right now about like trying to specify 280 00:18:41,640 --> 00:18:43,200 for like a more specific audience. 281 00:18:43,200 --> 00:18:51,480 But yes, certainly right now, I think it's like data scientists who are like beginner 282 00:18:51,480 --> 00:18:53,000 to intermediate at Python. 283 00:18:53,000 --> 00:18:58,640 I'd say like the lowest level is like like a lot of people think like, oh, this would 284 00:18:58,640 --> 00:19:00,680 be a good learning tool. 285 00:19:00,680 --> 00:19:01,680 And I see that. 286 00:19:01,680 --> 00:19:06,120 But I disagree because I think it would actually wouldn't it wouldn't let you like it wouldn't 287 00:19:06,120 --> 00:19:07,920 make you become better at Python. 288 00:19:07,920 --> 00:19:11,200 Like it would just make you be able to do some of the things we want to do in Python 289 00:19:11,200 --> 00:19:12,200 more quickly. 290 00:19:12,200 --> 00:19:17,680 So I think it's actually better for like someone who's like six months to like three years 291 00:19:17,680 --> 00:19:19,960 experience with Python. 292 00:19:19,960 --> 00:19:23,360 And they know what they want to do, and they know how it works, but they're just like they're 293 00:19:23,360 --> 00:19:26,760 tired of constantly going to Stack Overflow or like Googling syntax. 294 00:19:26,760 --> 00:19:31,520 It's just easier to do some things very quickly in here like a pivot table, for example, pivoting 295 00:19:31,520 --> 00:19:33,440 in pandas is like a pretty common thing to do. 296 00:19:33,440 --> 00:19:38,600 But I've talked to like even really advanced data scientists who just like never remember 297 00:19:38,600 --> 00:19:41,960 the pandas pivot syntax, and they're constantly Googling that. 298 00:19:41,960 --> 00:19:48,480 And so our tool is like, you know, quickly do your pivot table in our sheet. 299 00:19:48,480 --> 00:19:51,520 And so one is like the ability to just generate the syntax more quickly. 300 00:19:51,520 --> 00:19:54,320 And then two is just the visual aspect. 301 00:19:54,320 --> 00:19:58,440 A lot of people just like, you know, looking at their data during their analysis more in 302 00:19:58,440 --> 00:20:01,360 a more dynamic way than just like printing out a data frame. 303 00:20:01,360 --> 00:20:04,120 Which I think is cool. 304 00:20:04,120 --> 00:20:11,360 That's definitely one of the things I like about it. 305 00:20:11,360 --> 00:20:14,080 Did the package install? 306 00:20:14,080 --> 00:20:19,160 We're still installing here. 307 00:20:19,160 --> 00:20:22,720 Yes the other thing is like it's installing Plotly, which is like for graphing, but like 308 00:20:22,720 --> 00:20:25,000 you're not going to use graphing if you're not using the font. 309 00:20:25,000 --> 00:20:26,960 So it's kind of annoying. 310 00:20:26,960 --> 00:20:30,920 How about you Charles? 311 00:20:30,920 --> 00:20:36,480 Do you have thoughts so far or? 312 00:20:36,480 --> 00:20:39,320 No, this is really interesting. 313 00:20:39,320 --> 00:20:47,760 This is, you know, you might be targeting people who use like, was it Power BI? 314 00:20:47,760 --> 00:20:49,920 Yeah, I see that. 315 00:20:49,920 --> 00:20:57,000 People who do data science stuff, but don't really know programming. 316 00:20:57,000 --> 00:21:02,680 That would be, you know, I think that would be a good audience. 317 00:21:02,680 --> 00:21:03,680 That's interesting. 318 00:21:03,680 --> 00:21:04,680 Yeah. 319 00:21:04,680 --> 00:21:08,720 What do you think about the cannabis space? 320 00:21:08,720 --> 00:21:12,000 So I think one thing, one of my interests as a cannabis, obviously it's like, you know, 321 00:21:12,000 --> 00:21:14,940 one from an investing standpoint and the two from like a standpoint of this tool maybe 322 00:21:14,940 --> 00:21:20,720 is like a lot of places I've seen people enjoy the tool is like these like transitional industries 323 00:21:20,720 --> 00:21:25,640 where it's like they're typically using things like Excel, more analog tools. 324 00:21:25,640 --> 00:21:28,840 And because of like they have much larger data sizes, so they want to do more advanced 325 00:21:28,840 --> 00:21:30,160 analyses now. 326 00:21:30,160 --> 00:21:32,800 They're transitioning to Python. 327 00:21:32,800 --> 00:21:37,960 So we I've seen that in like finance a little bit, like pharma, like bio research. 328 00:21:37,960 --> 00:21:38,960 I'm seeing that. 329 00:21:38,960 --> 00:21:46,480 Do you think that cannabis is a place where, you know, there is a push to get these like 330 00:21:46,480 --> 00:21:49,400 non technical people more technical? 331 00:21:49,400 --> 00:21:59,360 Well, I what I have heard is people do want a hold of their data and to be able to analyze 332 00:21:59,360 --> 00:22:00,360 it. 333 00:22:00,360 --> 00:22:12,880 So that's, you know, something that I'm not sure if it's their design or, you know, a 334 00:22:12,880 --> 00:22:15,360 lot of people kind of are holding that hostage these days. 335 00:22:15,360 --> 00:22:19,800 So it seems well, that's what I've heard from the industry. 336 00:22:19,800 --> 00:22:24,880 So it seems like people are having a hard time getting their data out of whatever program 337 00:22:24,880 --> 00:22:25,880 they have it in. 338 00:22:25,880 --> 00:22:30,640 They just want to do statistics on their own. 339 00:22:30,640 --> 00:22:34,520 I just kind of heard that just word of mouth yesterday. 340 00:22:34,520 --> 00:22:44,000 So I think there is a demand for just people to explore their own data. 341 00:22:44,000 --> 00:22:45,000 Right. 342 00:22:45,000 --> 00:22:49,920 So I think people are becoming increasingly more sophisticated. 343 00:22:49,920 --> 00:22:58,080 And so people want to get a hold of their data and look at it themselves. 344 00:22:58,080 --> 00:23:06,960 Charles, in your work as a data scientist, do you think people are wanting to get a hold 345 00:23:06,960 --> 00:23:11,320 of their data more and more these days? 346 00:23:11,320 --> 00:23:12,320 Yeah, I think so. 347 00:23:12,320 --> 00:23:14,640 I mean, I'm not involved in the cannabis industry. 348 00:23:14,640 --> 00:23:26,360 So I can't speak to that, but there are a lot of people that do data analysis but don't 349 00:23:26,360 --> 00:23:31,320 really know like Python or any sort of programming languages. 350 00:23:31,320 --> 00:23:36,120 And again, they use tools like, you know, they use Excel, they use Power BI, they understand 351 00:23:36,120 --> 00:23:37,920 how to use those things. 352 00:23:37,920 --> 00:23:42,480 So this would be kind of good for people who aren't as technical but still do a lot of 353 00:23:42,480 --> 00:23:46,560 data analysis. 354 00:23:46,560 --> 00:23:53,360 So my other question is how large of a data set have you tried this on? 355 00:23:53,360 --> 00:23:56,800 Yeah, that's a good question. 356 00:23:56,800 --> 00:24:00,240 I'm confident that it will... 357 00:24:00,240 --> 00:24:03,840 So the idea is that what you're doing is really in the sheet, you're editing a data frame. 358 00:24:03,840 --> 00:24:07,440 So you're either passing a data frame as an argument and that will populate the sheet. 359 00:24:07,440 --> 00:24:11,000 Or if you're passing in a CSV, the first thing it automatically does is turn your CSV into 360 00:24:11,000 --> 00:24:12,000 a data frame. 361 00:24:12,000 --> 00:24:16,680 So as long as you can fit the data frame... 362 00:24:16,680 --> 00:24:19,680 You can fit... 363 00:24:19,680 --> 00:24:23,440 As long as you can fit that set into a data frame, which I think that most of you have 364 00:24:23,440 --> 00:24:25,400 done is like 15 million rows. 365 00:24:25,400 --> 00:24:28,240 I'm not sure how many columns that was. 366 00:24:28,240 --> 00:24:29,720 So it's more about obviously the area. 367 00:24:29,720 --> 00:24:38,240 But then I think it's definitely usable and of course way faster than doing it in other 368 00:24:38,240 --> 00:24:42,320 visual tools like Excel, for example. 369 00:24:42,320 --> 00:24:46,200 Excel is just really slow. 370 00:24:46,200 --> 00:24:47,800 But then there's also like... 371 00:24:47,800 --> 00:24:50,200 It's actually a really whole space of the data science I'm not too familiar with. 372 00:24:50,200 --> 00:24:52,600 It's like these people doing... 373 00:24:52,600 --> 00:24:56,640 I think it's like PySpark where it's like they have too much data. 374 00:24:56,640 --> 00:24:59,160 It's like they can't even fit in data frames. 375 00:24:59,160 --> 00:25:02,800 It's this whole other dimension of data science that I really am not too experienced in. 376 00:25:02,800 --> 00:25:08,760 So I think we're more focused on this... 377 00:25:08,760 --> 00:25:12,720 I think the people who have these a bit more smaller data sets maybe are doing more exploratory 378 00:25:12,720 --> 00:25:14,400 data analysis to start. 379 00:25:14,400 --> 00:25:18,800 Smaller, meaning still 5 million rows potentially. 380 00:25:18,800 --> 00:25:28,280 But we're definitely not there for huge productionized machine learning models for insurance companies 381 00:25:28,280 --> 00:25:29,280 and stuff like that. 382 00:25:29,280 --> 00:25:33,000 So we're just like crazy crazy, terabytes of data. 383 00:25:33,000 --> 00:25:40,720 If you want, I just started to do this brief analysis on comparing Eastern and Western 384 00:25:40,720 --> 00:25:41,720 Washington rainfall. 385 00:25:41,720 --> 00:25:45,600 So that affects how cannabis yields. 386 00:25:45,600 --> 00:25:53,000 So if you want, could I show you that brief analysis and see if that may be something 387 00:25:53,000 --> 00:25:57,880 that could be recreated in your tool? 388 00:25:57,880 --> 00:26:03,480 Would you want to try uploading that into the tool? 389 00:26:03,480 --> 00:26:06,480 Whatever you want to do. 390 00:26:06,480 --> 00:26:14,840 Yes, so here, why don't I run through it real quick and then we can try it with your program. 391 00:26:14,840 --> 00:26:18,600 Sweet, that'd be awesome. 392 00:26:18,600 --> 00:26:29,360 And that way we can try to replicate the results. 393 00:26:29,360 --> 00:26:32,120 Just a brief overview. 394 00:26:32,120 --> 00:26:37,080 Essentially we're just doing sort of a difference in difference. 395 00:26:37,080 --> 00:26:42,440 So just kind of comparing one region to the next. 396 00:26:42,440 --> 00:26:50,200 And so essentially we're just seeing, okay, it looks like there's quite a different climate 397 00:26:50,200 --> 00:26:56,680 in these counties compared to these counties. 398 00:26:56,680 --> 00:27:03,560 So over here it looks wet and incredibly wet. 399 00:27:03,560 --> 00:27:06,640 And then over here it's much drier. 400 00:27:06,640 --> 00:27:11,520 And so does that have any effect on cannabis yields? 401 00:27:11,520 --> 00:27:17,600 Is everybody doing things inside and temperature and humidity controlled rooms? 402 00:27:17,600 --> 00:27:20,720 And it may not have any effect whatsoever. 403 00:27:20,720 --> 00:27:25,320 Or maybe people over here are having a hard time keeping their crops cool. 404 00:27:25,320 --> 00:27:34,360 So we actually will have to do some statistics to see if there's any fundamental difference 405 00:27:34,360 --> 00:27:36,640 in the cannabis yields. 406 00:27:36,640 --> 00:27:41,560 So it's not actually that tricky of an analysis here. 407 00:27:41,560 --> 00:27:43,560 It's pretty short really. 408 00:27:43,560 --> 00:27:47,760 So essentially we're just going to use dummy variables. 409 00:27:47,760 --> 00:27:57,680 So one, if you're in Eastern Washington, zero if you're in Western Washington. 410 00:27:57,680 --> 00:27:59,680 Or I forget which. 411 00:27:59,680 --> 00:28:00,680 Exactly. 412 00:28:00,680 --> 00:28:11,760 So one, if you're in Eastern Washington, zero if you're in Western Washington. 413 00:28:11,760 --> 00:28:19,960 And so the past few weeks we've been working with lab result data in Washington state as 414 00:28:19,960 --> 00:28:28,480 well as earlier year in also what is the name of your program? 415 00:28:28,480 --> 00:28:31,240 Hold on, MITO. 416 00:28:31,240 --> 00:28:40,000 So in MITO we have already read in the licensees. 417 00:28:40,000 --> 00:28:50,640 And so we're joining that data with the lab results. 418 00:28:50,640 --> 00:28:57,320 Cool. 419 00:28:57,320 --> 00:29:01,440 Okay so. 420 00:29:01,440 --> 00:29:03,160 And you're reading this from an Excel file? 421 00:29:03,160 --> 00:29:04,160 Oh that's the basis. 422 00:29:04,160 --> 00:29:06,320 Did you say that one more time please? 423 00:29:06,320 --> 00:29:07,480 You're reading this from an Excel file? 424 00:29:07,480 --> 00:29:10,560 I think I saw a line in there that said read Excel. 425 00:29:10,560 --> 00:29:13,880 Yes, I think I may have moved this file. 426 00:29:13,880 --> 00:29:18,440 So one second and I am going to make a... 427 00:29:18,440 --> 00:29:26,440 Let me pause this for one second and I'm going to... 428 00:29:26,440 --> 00:29:27,440 Yeah no worries. 429 00:29:27,440 --> 00:29:28,440 All right. 430 00:29:28,440 --> 00:29:29,440 Hey. 431 00:29:29,440 --> 00:29:30,440 Hold on. 432 00:29:30,440 --> 00:29:34,440 I almost forgot Paul here. 433 00:29:34,440 --> 00:29:37,440 Hey Paul. 434 00:29:37,440 --> 00:29:40,440 Hey Paul. 435 00:29:40,440 --> 00:29:46,440 Hey guys, how you doing? 436 00:29:46,440 --> 00:29:47,440 Good. 437 00:29:47,440 --> 00:29:48,440 Excellent. 438 00:29:48,440 --> 00:29:56,840 So you actually joined in a good time because I'm sort of getting the data connected here. 439 00:29:56,840 --> 00:30:05,680 So we've got Jake today who's building a data science tool and we're using it to explore 440 00:30:05,680 --> 00:30:08,440 some of the cannabis data we're working with. 441 00:30:08,440 --> 00:30:09,440 Oh very cool. 442 00:30:09,440 --> 00:30:10,440 Nice to meet you Paul. 443 00:30:10,440 --> 00:30:11,440 How you doing Jake? 444 00:30:11,440 --> 00:30:12,440 Doing well. 445 00:30:12,440 --> 00:30:13,440 Where are you located? 446 00:30:13,440 --> 00:30:14,440 Southeast Michigan. 447 00:30:14,440 --> 00:30:15,440 Where are you? 448 00:30:15,440 --> 00:30:16,440 Cool. 449 00:30:16,440 --> 00:30:17,440 I'm in... 450 00:30:17,440 --> 00:30:18,440 Philadelphia. 451 00:30:18,440 --> 00:30:19,440 I'm actually a bit Jersey Shore right now though. 452 00:30:19,440 --> 00:30:20,440 Oh okay. 453 00:30:20,440 --> 00:30:21,440 Okay. 454 00:30:21,440 --> 00:30:31,440 So what we're doing Paul, so this is sort of merging both tools. 455 00:30:31,440 --> 00:30:41,440 We're basically doing an analysis on Eastern versus Western Washington lab results. 456 00:30:41,440 --> 00:30:47,600 So just a simple analysis just to demonstrate working with the data. 457 00:30:47,600 --> 00:30:50,440 Is it kind of like an A-B test or something? 458 00:30:50,440 --> 00:30:51,440 Essentially. 459 00:30:51,440 --> 00:31:00,600 So this was just sort of an example of merging the data sets and just running a regression. 460 00:31:00,600 --> 00:31:01,600 Right on. 461 00:31:01,600 --> 00:31:08,360 But I'm just getting the data pulled up here real quick. 462 00:31:08,360 --> 00:31:12,080 And while you're doing that, Charles thanks again for helping me out this week. 463 00:31:12,080 --> 00:31:13,080 I appreciate it. 464 00:31:13,080 --> 00:31:22,280 I haven't had a chance to get into the parquet files yet but I'm still trying to use PowerShell 465 00:31:22,280 --> 00:31:25,480 to get some of the conversions over and it seems to be working. 466 00:31:25,480 --> 00:31:27,600 It takes a long time though. 467 00:31:27,600 --> 00:31:28,600 Yes. 468 00:31:28,600 --> 00:31:35,920 It's converting it in pandas was... 469 00:31:35,920 --> 00:31:36,920 It was slow going. 470 00:31:36,920 --> 00:31:38,880 It was very slow going. 471 00:31:38,880 --> 00:31:39,880 Pretty painful. 472 00:31:39,880 --> 00:31:40,880 Yeah. 473 00:31:40,880 --> 00:31:42,560 So that's interesting. 474 00:31:42,560 --> 00:31:43,560 I was going to... 475 00:31:43,560 --> 00:31:44,560 I thought about like... 476 00:31:44,560 --> 00:31:49,200 I don't know a lot about R but I thought about trying to do some of this analysis in R and 477 00:31:49,200 --> 00:31:52,680 thinking oh maybe that would be better but you're running into the same problems I'm 478 00:31:52,680 --> 00:31:53,680 running into. 479 00:31:53,680 --> 00:31:54,680 Yeah. 480 00:31:54,680 --> 00:31:55,680 Yeah. 481 00:31:55,680 --> 00:32:03,400 And after you're mentioning GOSH in Python, what is the name of the chunking application 482 00:32:03,400 --> 00:32:05,400 you referred to? 483 00:32:05,400 --> 00:32:06,400 Dask? 484 00:32:06,400 --> 00:32:07,400 Dask. 485 00:32:07,400 --> 00:32:14,240 Which I don't know anything about but within R there was a library that does essentially 486 00:32:14,240 --> 00:32:15,520 actually is like three libraries. 487 00:32:15,520 --> 00:32:21,360 I've tried two of them and those two didn't work but this other one seems to be kind of 488 00:32:21,360 --> 00:32:26,080 promising where it just essentially takes the data and chunks it out into files and 489 00:32:26,080 --> 00:32:28,960 kind of distributes them and you do some parallel processing on it. 490 00:32:28,960 --> 00:32:34,760 So I'm going to give that a whirl and see what happens and yeah just test it out. 491 00:32:34,760 --> 00:32:35,760 Cool. 492 00:32:35,760 --> 00:32:36,760 Sorry about that. 493 00:32:36,760 --> 00:32:37,760 I can focus now. 494 00:32:37,760 --> 00:32:48,760 So that's good and I've read your emails or your messages that you're working to get the 495 00:32:48,760 --> 00:32:52,720 UTF-16 converted to UTF-8. 496 00:32:52,720 --> 00:32:59,720 So correct me if I'm wrong but I think you've got a solution. 497 00:32:59,720 --> 00:33:08,760 Yeah it's essentially just using PowerShell and run a small script and it just chunks 498 00:33:08,760 --> 00:33:17,240 through the whole file and then gives me a UTF-8 TSV file on the other side. 499 00:33:17,240 --> 00:33:22,560 But then that's another script I can run that will convert the TSV to CSV and then I've 500 00:33:22,560 --> 00:33:26,280 got the full CSV file and then what I've been doing is I'm in the process of doing this 501 00:33:26,280 --> 00:33:32,200 right now is uploading them to Google Cloud with the free instance that I've got right 502 00:33:32,200 --> 00:33:38,200 now and then just to get some of these larger files up there to see if this tests out the 503 00:33:38,200 --> 00:33:40,880 process and then run some small queries against them. 504 00:33:40,880 --> 00:33:42,880 That's what I'm trying to do. 505 00:33:42,880 --> 00:33:44,840 Excellent work. 506 00:33:44,840 --> 00:33:53,600 So just to show just a demonstration of this data to Jake and just some statistics that 507 00:33:53,600 --> 00:33:56,720 I was beginning to tinker on. 508 00:33:56,720 --> 00:34:00,520 So just always just start with the question. 509 00:34:00,520 --> 00:34:08,240 So my question was does rainfall, so does like the amount of moisture affect the cannabinoid 510 00:34:08,240 --> 00:34:09,240 results? 511 00:34:09,240 --> 00:34:16,960 And so I kind of wanted to look at all of the different results maybe some fail more 512 00:34:16,960 --> 00:34:23,120 for moisture content, maybe the water activity is higher on some than the others. 513 00:34:23,120 --> 00:34:32,360 So the way Washington is set up, it works well for essentially just a quick and simple 514 00:34:32,360 --> 00:34:34,720 dummy variable analysis. 515 00:34:34,720 --> 00:34:40,000 So we're just going to get. 516 00:34:40,000 --> 00:34:45,560 I think we just lost him. 517 00:34:45,560 --> 00:34:50,720 Hey Keegan, I don't know if you can hear us, but you froze up. 518 00:34:50,720 --> 00:34:53,680 At least it's a nice image to look at. 519 00:34:53,680 --> 00:34:55,840 He's got his power fist going, so that's good. 520 00:34:55,840 --> 00:34:58,840 Oh, there you are Keegan. 521 00:34:58,840 --> 00:34:59,840 You came back. 522 00:34:59,840 --> 00:35:00,840 My apologies. 523 00:35:00,840 --> 00:35:07,160 So I actually should let you know Jake, so I am working at the Oklahoma City Library. 524 00:35:07,160 --> 00:35:14,120 I'm in Oklahoma City for a CannaCon tomorrow and Friday. 525 00:35:14,120 --> 00:35:23,040 So I'm going to be meeting people in the cannabis industry here, see what their challenges are, 526 00:35:23,040 --> 00:35:26,280 how things are going, how the market is developing. 527 00:35:26,280 --> 00:35:28,280 That's really cool. 528 00:35:28,280 --> 00:35:33,280 So back to the work here, Eastern Western Washington. 529 00:35:33,280 --> 00:35:48,480 What I'm doing here is just reading in the lab result data and the licensee data and 530 00:35:48,480 --> 00:35:52,200 then merging it. 531 00:35:52,200 --> 00:36:06,080 And so, for example. 532 00:36:06,080 --> 00:36:15,680 So those are the variables for licensees and we can merge these with the licensees global 533 00:36:15,680 --> 00:36:25,760 ID and then the lab result for MME ID. 534 00:36:25,760 --> 00:36:42,200 So you can look at the data points, but that way we can merge them. 535 00:36:42,200 --> 00:36:49,440 And then we have a lot of data points for this one lab result. 536 00:36:49,440 --> 00:36:53,400 So now we have this is a lab result here. 537 00:36:53,400 --> 00:36:58,480 It is with this piece of inventory. 538 00:36:58,480 --> 00:37:07,520 So we could actually look up the inventory at this point and get even more data points. 539 00:37:07,520 --> 00:37:15,720 For now, our analysis doesn't need those data points, but in the future, we'll want to just 540 00:37:15,720 --> 00:37:19,920 get all the data points that we can together. 541 00:37:19,920 --> 00:37:25,680 And then essentially what we've done here is just added on the licensees data. 542 00:37:25,680 --> 00:37:33,360 So now we know, okay, this licensee is in Bellingham, Washington. 543 00:37:33,360 --> 00:37:38,480 And they're in Whatcom County. 544 00:37:38,480 --> 00:37:46,200 So now it's pretty easy just to create a dummy variable if they're in Eastern or Western 545 00:37:46,200 --> 00:37:47,200 Washington. 546 00:37:47,200 --> 00:37:54,320 Pandas has tricks for creating dummy variables, but I actually just wrote a quick script, 547 00:37:54,320 --> 00:38:00,680 you know, just wrote a quick, you know, quick little loop here just to create that. 548 00:38:00,680 --> 00:38:03,200 Can I jump in for a sec? 549 00:38:03,200 --> 00:38:06,760 So I think what's interesting there is you wrote, so you wrote that little script there, 550 00:38:06,760 --> 00:38:09,600 you know, it's an if statement essentially. 551 00:38:09,600 --> 00:38:14,680 So what you would do in my tool is you would just use like, you would just make a new column 552 00:38:14,680 --> 00:38:17,400 and then you could do an if statement in that column. 553 00:38:17,400 --> 00:38:23,120 Just like if this, that, if x1, if y0, it's kind of interesting. 554 00:38:23,120 --> 00:38:24,120 Yes. 555 00:38:24,120 --> 00:38:26,840 So Paul, this is Jake's tool. 556 00:38:26,840 --> 00:38:32,760 And so this is sort of combined as a user interface into a Python code generator. 557 00:38:32,760 --> 00:38:33,760 Okay. 558 00:38:33,760 --> 00:38:34,760 This would be... 559 00:38:34,760 --> 00:38:35,760 That's called Mito sheet. 560 00:38:35,760 --> 00:38:36,760 Yeah. 561 00:38:36,760 --> 00:38:49,760 Here, I'll put the, I'll put some links in the chat. 562 00:38:49,760 --> 00:38:56,640 So just for everybody's sake, I'll run through mine real quick, but then essentially I see 563 00:38:56,640 --> 00:39:03,920 you would basically, you know, create your dummy variable here, sort of Excel style. 564 00:39:03,920 --> 00:39:11,040 And then you'll have, you know, the code automatically generated. 565 00:39:11,040 --> 00:39:14,040 Yeah, totally, totally. 566 00:39:14,040 --> 00:39:16,080 Was it cool? 567 00:39:16,080 --> 00:39:21,200 So we've got, we've got that dummy variable now. 568 00:39:21,200 --> 00:39:25,360 And so, I mean, so this is actually interesting. 569 00:39:25,360 --> 00:39:34,000 This is actually the first time I've calculated this statistic, but like, so this is an interesting 570 00:39:34,000 --> 00:39:37,220 statistic that maybe not a lot of people even know. 571 00:39:37,220 --> 00:39:50,800 So it looks like, you know, 62% of licensees are in Eastern Washington and, you know, you're 572 00:39:50,800 --> 00:40:00,000 about 38% are in the other, you know, Eastern Washington. 573 00:40:00,000 --> 00:40:06,760 And so what you could do is remember, I always say, well, I'm not sure if you've heard me 574 00:40:06,760 --> 00:40:13,760 say this, but what I would say is you can get real interesting by just doing conditional, 575 00:40:13,760 --> 00:40:16,740 conditional averages. 576 00:40:16,740 --> 00:40:26,440 So if you just, just took the mean conditional on, okay, if it's a producer or processor, 577 00:40:26,440 --> 00:40:35,000 you can now find out, okay, are there more producers in Western Washington or Eastern 578 00:40:35,000 --> 00:40:36,880 Washington? 579 00:40:36,880 --> 00:40:48,600 Because my, you know, hypothesis is that there are a lot of retailers in Eastern Washington 580 00:40:48,600 --> 00:40:55,480 and a lot of cultivators in Western Washington. 581 00:40:55,480 --> 00:41:03,960 And wonder, wonder what the implications would be on transportation for that. 582 00:41:03,960 --> 00:41:13,920 Well, it depends on the way the laboratory, well, actually, well, there's a lot of factors. 583 00:41:13,920 --> 00:41:14,920 So you're right. 584 00:41:14,920 --> 00:41:21,440 So the people will end up, and that's the thing in Washington is essentially transporting 585 00:41:21,440 --> 00:41:29,480 the cannabis from Western Washington to Eastern Washington, you know, first to test it at 586 00:41:29,480 --> 00:41:30,800 a laboratory. 587 00:41:30,800 --> 00:41:37,520 Here are some laboratories in Western Washington, and then to send it to all their retailers. 588 00:41:37,520 --> 00:41:43,800 So just to understand the flow a little bit, because I'm new to this, so any grower has 589 00:41:43,800 --> 00:41:47,960 to, there's some formality, they have to get it tested at a specific type of lab before 590 00:41:47,960 --> 00:41:49,320 they can sell it. 591 00:41:49,320 --> 00:41:50,320 They can't do any of that testing in-house. 592 00:41:50,320 --> 00:41:51,320 Exactly. 593 00:41:51,320 --> 00:42:00,920 So they need to get their products tested at a third party independent laboratory. 594 00:42:00,920 --> 00:42:01,920 That's interesting. 595 00:42:01,920 --> 00:42:02,920 I wonder why. 596 00:42:02,920 --> 00:42:03,920 Yeah, that's interesting. 597 00:42:03,920 --> 00:42:07,680 I think it's for, well, I'm just speculating. 598 00:42:07,680 --> 00:42:14,280 I'm new like you, but I would think it's for overlapping interests, right? 599 00:42:14,280 --> 00:42:16,360 Because you could probably cut a lot of corners. 600 00:42:16,360 --> 00:42:21,800 If you're a transporter and a producer, I wonder if some of the checks and balances 601 00:42:21,800 --> 00:42:24,800 would be kind of compromised by doing that. 602 00:42:24,800 --> 00:42:25,800 Yeah, totally. 603 00:42:25,800 --> 00:42:31,720 It was interesting if you could, like, if you could, there's a way you could, I don't 604 00:42:31,720 --> 00:42:35,320 know, there's a way you could have it so you could do that testing in-house. 605 00:42:35,320 --> 00:42:37,560 You could potentially save a lot of money. 606 00:42:37,560 --> 00:42:40,200 I imagine it's the growers who are paying for that transportation. 607 00:42:40,200 --> 00:42:48,080 Yeah, but they have a big problem with people trying to fake results. 608 00:42:48,080 --> 00:42:49,080 Right, right. 609 00:42:49,080 --> 00:42:58,720 Yeah, I mean, there's a lot of growers who are using a lot of pesticides and agents 610 00:42:58,720 --> 00:43:01,720 to stop mold. 611 00:43:01,720 --> 00:43:05,600 It wasn't a safe product to sell anymore. 612 00:43:05,600 --> 00:43:09,760 So that's why you have independent labs to do that kind of stuff. 613 00:43:09,760 --> 00:43:18,360 Also, we should actually be careful with the statistics that I was calculating because 614 00:43:18,360 --> 00:43:25,640 I was only reading in 10,000 and there are about 2 million lab results. 615 00:43:25,640 --> 00:43:29,160 So that is not even going to be close to accurate. 616 00:43:29,160 --> 00:43:32,480 Yeah, you can do a random sample if you don't want to read it all in. 617 00:43:32,480 --> 00:43:33,480 Quick question, Charles. 618 00:43:33,480 --> 00:43:45,400 Do you know how to read in a CSV with just a random number of samples? 619 00:43:45,400 --> 00:43:46,400 No. 620 00:43:46,400 --> 00:43:56,320 Yeah, but so you may just have to hitch this, but this is sort of just a demo of it, an 621 00:43:56,320 --> 00:43:57,320 analysis. 622 00:43:57,320 --> 00:43:58,320 That's pretty cool. 623 00:43:58,320 --> 00:44:02,920 I mean, we were talking about this before, just the fact that you're going through this 624 00:44:02,920 --> 00:44:07,320 and answering some of those high level stats questions gives you that baseline of understanding 625 00:44:07,320 --> 00:44:10,840 of what the market's like and how the market operates too. 626 00:44:10,840 --> 00:44:15,000 So just having that information, you'll be way ahead of a lot of people. 627 00:44:15,000 --> 00:44:16,000 Exactly. 628 00:44:16,000 --> 00:44:21,160 So my red flag was, that's what you do when you're looking at data. 629 00:44:21,160 --> 00:44:25,280 You've got to look at the data because there could be some red flags. 630 00:44:25,280 --> 00:44:32,760 And so what my red flag was, was 100% of the cultivators are in Eastern Washington. 631 00:44:32,760 --> 00:44:40,880 And so that's just because I just read in the first 10,000 observations. 632 00:44:40,880 --> 00:44:44,720 One thing that comes to mind is you're showing me this, and we talked about this a little 633 00:44:44,720 --> 00:44:51,800 bit last time, but like Washington and other states, they're in this kind of weird situation 634 00:44:51,800 --> 00:44:53,640 where they're trying to grow an industry, right? 635 00:44:53,640 --> 00:44:57,640 They're really trying to get it off the ground so they can get their tax money out of it. 636 00:44:57,640 --> 00:45:03,200 But they're also trying to rein it in enough to where it's controlled and people aren't 637 00:45:03,200 --> 00:45:07,440 getting sick from bad product or what have you, try to mitigate the risks. 638 00:45:07,440 --> 00:45:09,600 So they're kind of this weird balancing act. 639 00:45:09,600 --> 00:45:14,480 But at some point, I would imagine that they're going to have to do some auditing and compliance, 640 00:45:14,480 --> 00:45:20,360 like pretty stringent auditing and compliance checks. 641 00:45:20,360 --> 00:45:28,800 Or also checks for if there's any way to check for what's the word I'm looking for? 642 00:45:28,800 --> 00:45:29,800 Shoot. 643 00:45:29,800 --> 00:45:32,960 That will come to me in a second. 644 00:45:32,960 --> 00:45:37,720 But to position yourself to where you can provide data that would help with auditing 645 00:45:37,720 --> 00:45:42,840 and compliance would probably be a pretty good strategic move. 646 00:45:42,840 --> 00:45:52,440 That's interesting because I think you're right, LEAF Data Systems is collecting all 647 00:45:52,440 --> 00:45:57,440 this data just to look at it. 648 00:45:57,440 --> 00:46:03,640 Rule number one about data, look at the data. 649 00:46:03,640 --> 00:46:05,640 So you're right. 650 00:46:05,640 --> 00:46:09,640 And it's one of those things where like I was saying, there's such a high demand for 651 00:46:09,640 --> 00:46:12,600 this and such a low supply. 652 00:46:12,600 --> 00:46:20,040 So LEAF Data Systems is probably trying to do their best to supply visualizations and 653 00:46:20,040 --> 00:46:25,400 reports and statistics, but they may be underwater. 654 00:46:25,400 --> 00:46:33,840 There may be such a high demand, they just can't fill it. 655 00:46:33,840 --> 00:46:35,640 So you're right. 656 00:46:35,640 --> 00:46:42,200 So that's where other people in the industry can kind of help out. 657 00:46:42,200 --> 00:46:45,960 So I want to give a shout out to Jim McCray. 658 00:46:45,960 --> 00:46:53,240 And so he is actually the person who originally provided this data set to us. 659 00:46:53,240 --> 00:46:55,240 And that's essentially what he does. 660 00:46:55,240 --> 00:46:58,240 He's sort of the canary in the coal mine. 661 00:46:58,240 --> 00:47:01,120 So he, the canary in the data mine. 662 00:47:01,120 --> 00:47:09,640 So he is digging around with Kim's data, essentially doing exactly what you're talking about, just 663 00:47:09,640 --> 00:47:19,360 seeing if there's any oddities or being an extra eye on. 664 00:47:19,360 --> 00:47:20,920 So who is Jim McCray? 665 00:47:20,920 --> 00:47:23,880 Is he with the State of Washington? 666 00:47:23,880 --> 00:47:30,400 I think he, I'm not 100% sure the organization he's associated with, but he's a researcher 667 00:47:30,400 --> 00:47:31,400 in Washington. 668 00:47:31,400 --> 00:47:38,120 And he's done a lot of research on the cannabis industry. 669 00:47:38,120 --> 00:47:43,720 So you may be able to find him on LinkedIn. 670 00:47:43,720 --> 00:47:53,960 That if you could honestly be the best way to get a hold of him, make him hear his LinkedIn. 671 00:47:53,960 --> 00:47:54,960 I was just curious. 672 00:47:54,960 --> 00:48:00,920 So yeah, so he's trying to, in a researchy kind of way, try and keep a pulse on the market 673 00:48:00,920 --> 00:48:08,880 and keep his eye open for things like what compliance and that sort of thing. 674 00:48:08,880 --> 00:48:09,880 Exactly. 675 00:48:09,880 --> 00:48:13,640 So there's interesting things to look at. 676 00:48:13,640 --> 00:48:23,400 So one is just the total amount of sales per lab result, because some products just sell 677 00:48:23,400 --> 00:48:25,360 a lot and others don't. 678 00:48:25,360 --> 00:48:37,360 And then you just want to keep an eye on failure rates just to make sure that there's not a 679 00:48:37,360 --> 00:48:43,240 significant, where you just be aware of a significant trend one way or the other. 680 00:48:43,240 --> 00:48:49,440 So for example, so just to give them a shout out. 681 00:48:49,440 --> 00:48:57,160 So if you want a good example of someone who's sort of, someone keeping their finger on the 682 00:48:57,160 --> 00:49:01,400 pulse, but they're doing it in California. 683 00:49:01,400 --> 00:49:14,640 So they're just seeing, okay, how many things are being tested and they plot failure rates. 684 00:49:14,640 --> 00:49:23,040 So you can see that failure rates are going down over time while the number of tests are 685 00:49:23,040 --> 00:49:27,920 going up. 686 00:49:27,920 --> 00:49:34,240 I think they may have, and this is, I think one of their best contributions is they actually 687 00:49:34,240 --> 00:49:42,120 show the seasonality in, I believe this is seasonality in failures. 688 00:49:42,120 --> 00:49:45,920 It may not necessarily be of cannabis. 689 00:49:45,920 --> 00:49:53,240 Oh, actually, no, no, no, this is pesticide use. 690 00:49:53,240 --> 00:50:00,840 So this is the analysis they've done that I think can be extended. 691 00:50:00,840 --> 00:50:13,840 So in California, they record the amount of pesticides that are used over time and it 692 00:50:13,840 --> 00:50:18,800 seems there's a cyclical use of pesticides. 693 00:50:18,800 --> 00:50:28,320 I think it would be interesting to try to correlate pesticide use with failure rates 694 00:50:28,320 --> 00:50:40,120 to see if somebody's using pesticides, does that affect the failure rate? 695 00:50:40,120 --> 00:50:48,880 But enough of that tangent. 696 00:50:48,880 --> 00:50:53,120 Sorry about that. 697 00:50:53,120 --> 00:51:00,840 So extra eye on the data. 698 00:51:00,840 --> 00:51:08,240 So we can actually use some statistics here to start answering some of these questions. 699 00:51:08,240 --> 00:51:16,240 So back to our original question of the day, does rainfall have an effect on cannabinoid 700 00:51:16,240 --> 00:51:17,240 results? 701 00:51:17,240 --> 00:51:24,680 Hedge this, that this is not a random sample, this is just the first 10,000 lab results. 702 00:51:24,680 --> 00:51:29,880 So you'd want to actually repeat this analysis with all of the lab results and then you'd 703 00:51:29,880 --> 00:51:34,280 actually get a better estimate of the effect. 704 00:51:34,280 --> 00:51:39,000 So hedge this, this is just a demonstration of the statistics. 705 00:51:39,000 --> 00:51:45,800 It's not actually representative of what we think that your estimates are. 706 00:51:45,800 --> 00:51:51,320 So Keegan, you pulled out like the first 10,000 results. 707 00:51:51,320 --> 00:51:57,960 I was able to do the same sort of thing with the files that I have. 708 00:51:57,960 --> 00:52:03,520 But to look across the entire data set, do you have the capacity to do that yourself 709 00:52:03,520 --> 00:52:11,120 or are you kind of constrained on how much data you can work with? 710 00:52:11,120 --> 00:52:13,000 I can probably pull it off. 711 00:52:13,000 --> 00:52:18,680 I just haven't yet. 712 00:52:18,680 --> 00:52:23,840 Were you thinking that you may be able to do this analysis or? 713 00:52:23,840 --> 00:52:28,920 Well, right now, like I've been kind of complaining the last couple of meetings and leaning on 714 00:52:28,920 --> 00:52:35,160 your guys's expertise, but I'm still trying just to get the entire data sets in a usable 715 00:52:35,160 --> 00:52:38,000 kind of format and I'm getting there. 716 00:52:38,000 --> 00:52:44,000 But as soon as I get it up to Google Cloud, I mean, running queries on this data is going 717 00:52:44,000 --> 00:52:46,360 to be really easy. 718 00:52:46,360 --> 00:52:47,840 Just running SQL queries on it. 719 00:52:47,840 --> 00:52:49,640 Of course, how much is going to cost? 720 00:52:49,640 --> 00:52:50,640 I don't know yet. 721 00:52:50,640 --> 00:52:52,600 So I got to be careful of that. 722 00:52:52,600 --> 00:52:59,040 But if I get those things up in place and it seems like they're intact, they're not 723 00:52:59,040 --> 00:53:03,320 corrupt because I know that Charles, you're saying that like the inventories file you 724 00:53:03,320 --> 00:53:06,500 found to be corrupted somehow. 725 00:53:06,500 --> 00:53:09,940 So I'm just saying if I can get it up there and they're in place, I'll let you know. 726 00:53:09,940 --> 00:53:13,720 And then maybe we can do something to kind of speed up some of your exploratory analysis 727 00:53:13,720 --> 00:53:14,720 a little bit. 728 00:53:14,720 --> 00:53:15,720 Definitely, definitely, definitely. 729 00:53:15,720 --> 00:53:16,720 So how about this next week? 730 00:53:16,720 --> 00:53:25,720 I can focus and then it seems that you're going to be focusing on this as well. 731 00:53:25,720 --> 00:53:30,160 Let's get this data usable because that's the first step. 732 00:53:30,160 --> 00:53:36,040 I think, you know, we've sort of mocked out how we can do some analysis. 733 00:53:36,040 --> 00:53:41,760 But you know, we've got to do our YAC shaving. 734 00:53:41,760 --> 00:53:49,400 And so we're basically stuck right here where we need to get all this data ready. 735 00:53:49,400 --> 00:53:56,080 And Charles has come up with a couple of good solutions. 736 00:53:56,080 --> 00:54:00,120 I just need to spend time and implement them. 737 00:54:00,120 --> 00:54:07,080 So this next week, that's what I'll be focusing on is getting this data usable. 738 00:54:07,080 --> 00:54:12,640 And then next week we can touch base again and I'll send you messages throughout the 739 00:54:12,640 --> 00:54:13,640 week. 740 00:54:13,640 --> 00:54:22,040 And that will be essentially the next step in our data analysis journey. 741 00:54:22,040 --> 00:54:28,120 Just for this, just so Jake knows, I'm actually just finished up in a master's in data science 742 00:54:28,120 --> 00:54:30,240 and I'm getting ready to do my capstone project. 743 00:54:30,240 --> 00:54:36,920 So I kind of lucked out and hooked up with the guys here. 744 00:54:36,920 --> 00:54:42,460 And I thought it would be kind of interesting to do a project around the cannabis industry 745 00:54:42,460 --> 00:54:45,120 just because it's a novel data set. 746 00:54:45,120 --> 00:54:50,320 And you can apply it like Keegan was saying the first time I was talking to him that because 747 00:54:50,320 --> 00:54:55,660 it's a new data set, there's a really good low hanging fruit here with some basic analysis. 748 00:54:55,660 --> 00:54:57,960 So that was kind of that's my focus. 749 00:54:57,960 --> 00:55:02,760 And I have to actually try and get a data set together pretty quick because my project's 750 00:55:02,760 --> 00:55:04,760 due by the beginning of August. 751 00:55:04,760 --> 00:55:08,280 And there's a bunch of other stuff I've got to do as well. 752 00:55:08,280 --> 00:55:11,880 So my goal is to try and get this data in a usable format as soon as possible. 753 00:55:11,880 --> 00:55:16,480 And of course, whatever success I have, I'll share with you guys. 754 00:55:16,480 --> 00:55:18,560 And then we can all kind of pound on it. 755 00:55:18,560 --> 00:55:19,560 Cool. 756 00:55:19,560 --> 00:55:24,040 I sound like the chap, if the tool, it's of course free. 757 00:55:24,040 --> 00:55:27,400 If it's at all helpful to you in the process, definitely. 758 00:55:27,400 --> 00:55:29,600 I put the docs in the chat if you want to try it out. 759 00:55:29,600 --> 00:55:30,600 Awesome. 760 00:55:30,600 --> 00:55:31,600 Yeah. 761 00:55:31,600 --> 00:55:32,600 Thanks, Jake. 762 00:55:32,600 --> 00:55:33,600 I appreciate that. 763 00:55:33,600 --> 00:55:39,400 Just to take it home real quick and then we'll close out. 764 00:55:39,400 --> 00:55:46,360 Basically we just ran a regression here, the THC on Eastern Washington. 765 00:55:46,360 --> 00:55:56,560 So you know, you could interpret this coefficient as, you know, the THC difference in Eastern 766 00:55:56,560 --> 00:55:58,520 Washington versus Western Washington. 767 00:55:58,520 --> 00:56:05,440 Keep in mind, we have read in all the data, but you would interpret this if you were confident 768 00:56:05,440 --> 00:56:08,240 that that was about a 5% difference. 769 00:56:08,240 --> 00:56:16,480 But there were some red flags, R squared, and principally R squared is a pretty big 770 00:56:16,480 --> 00:56:19,480 red flag. 771 00:56:19,480 --> 00:56:29,920 So, Jake, this is essentially the final piece that would need to be done in your program 772 00:56:29,920 --> 00:56:30,920 if possible. 773 00:56:30,920 --> 00:56:40,960 So, you know, we can get all the way through here where we've created the dummy variable. 774 00:56:40,960 --> 00:56:47,920 And so that could take us all the way to there and we can be content with that. 775 00:56:47,920 --> 00:56:53,960 And then if you stats models, or if you're interested in doing future work, you know, 776 00:56:53,960 --> 00:56:58,280 this may be something you'd be interested in looking at is, you know, just a way where 777 00:56:58,280 --> 00:57:01,920 you could run simple regressions. 778 00:57:01,920 --> 00:57:04,160 Yeah, totally. 779 00:57:04,160 --> 00:57:07,240 That's definitely something I think I want to add for sure. 780 00:57:07,240 --> 00:57:10,360 We're sort of, you know, we're building it out. 781 00:57:10,360 --> 00:57:15,920 I think right now we have like the base functionality you need, which is like transformations and 782 00:57:15,920 --> 00:57:19,000 pivot tables and merging and filtering. 783 00:57:19,000 --> 00:57:23,240 And then I think the question from there is like, do we go more into like the visualization 784 00:57:23,240 --> 00:57:28,720 space or the statistical space or the modeling space, you know, like machine learning, because 785 00:57:28,720 --> 00:57:34,800 obviously there's all these amazing packages we could try and develop this front end for. 786 00:57:34,800 --> 00:57:36,920 And so that's, yeah, that's just kind of the crossroads we're at. 787 00:57:36,920 --> 00:57:40,920 But definitely something like that, like running a simple regression, like, you know, it's 788 00:57:40,920 --> 00:57:41,920 not simple at all. 789 00:57:41,920 --> 00:57:48,160 But you kind of hit the nail on the head because we all agree that, you know, with data science, 790 00:57:48,160 --> 00:57:55,560 love it or hate it, you end up spending about 90% of your time or more, sometimes 99% of 791 00:57:55,560 --> 00:57:59,360 your time just cleaning the data and getting it into the checklist. 792 00:57:59,360 --> 00:58:05,160 And then, but that's the beauty of it because you spend all this work and then all of a 793 00:58:05,160 --> 00:58:08,000 sudden your data set's actually manageable. 794 00:58:08,000 --> 00:58:11,040 And then all you have to do is run a regression. 795 00:58:11,040 --> 00:58:17,400 And if it's good, clean data and you have all the data points you need, often you can 796 00:58:17,400 --> 00:58:23,920 just do an ordinary least squares regression and that'll be incredibly informative. 797 00:58:23,920 --> 00:58:24,920 Yeah, totally. 798 00:58:24,920 --> 00:58:25,920 Yeah, definitely. 799 00:58:25,920 --> 00:58:35,240 So, yeah, no, Jake, Keegan has a bunch of data and I'd be interested, you know, it'd 800 00:58:35,240 --> 00:58:41,880 be interesting for you to like try and load that into your program and see, because we've 801 00:58:41,880 --> 00:58:46,760 had a lot of problems with this data. 802 00:58:46,760 --> 00:58:50,240 And so if you're going to make it, you know, if it's going to be something sort of like 803 00:58:50,240 --> 00:58:56,120 an easy to use kind of tool, these are the kind of problems people are going to run into. 804 00:58:56,120 --> 00:58:58,280 Yeah, that'd be cool. 805 00:58:58,280 --> 00:59:04,000 Keegan, if you're up for it, I'd love maybe, or wherever, I'd love to maybe just, if you 806 00:59:04,000 --> 00:59:07,880 wanted to have like a 20 minute call sometime, maybe just try and go through some of this 807 00:59:07,880 --> 00:59:08,880 together would be cool. 808 00:59:08,880 --> 00:59:09,880 Absolutely. 809 00:59:09,880 --> 00:59:17,040 And I also put the data in the chat because I think Charles was right. 810 00:59:17,040 --> 00:59:27,880 Like if you want a data set to really, you know, what's the word, crash test your program, 811 00:59:27,880 --> 00:59:32,000 that could be a good data set for you. 812 00:59:32,000 --> 00:59:38,000 If I want to reach out to you, is this, this like contact cannabis email, is that good 813 00:59:38,000 --> 00:59:39,000 for you? 814 00:59:39,000 --> 00:59:44,000 You can actually reach me at keegan at canlyrics.com. 815 00:59:44,000 --> 00:59:47,000 Cool, cool, cool, cool. 816 00:59:47,000 --> 00:59:49,000 I'll definitely do that. 817 00:59:49,000 --> 00:59:50,000 All right. 818 00:59:50,000 --> 00:59:51,000 Great. 819 00:59:51,000 --> 00:59:54,000 Well, everyone, thank you for joining today. 820 00:59:54,000 --> 00:59:59,000 And we've got a big week ahead of data wrangling. 821 00:59:59,000 --> 01:00:06,000 So I'm excited, Paul's got a grin, Charles got a grin, so Jake, I'll look forward to 822 01:00:06,000 --> 01:00:09,000 talking with you and Charles as well. 823 01:00:09,000 --> 01:00:10,000 All right. 824 01:00:10,000 --> 01:00:11,000 Good talking guys. 825 01:00:11,000 --> 01:00:12,000 Nice to meet you, Jake. 826 01:00:12,000 --> 01:00:13,000 You as well. 827 01:00:13,000 --> 01:00:14,000 How are we going, guys? 828 01:00:14,000 --> 01:00:15,000 All right, take care. 829 01:00:15,000 --> 01:00:16,000 Have a good week. 830 01:00:16,000 --> 01:00:17,000 You too. 831 01:00:17,000 --> 01:00:18,000 Bye. 832 01:00:18,000 --> 01:00:31,880 Bye.