Looks like an expensive affair to me, This tool could potentially add value, maybe we do the training of the data and then we simply supply the template of the data , like a snapshot to use , this could avoid processing.
Not sure if this kind of service is available. Maybe just integrating the Google Images or Drive API, Simply Upload the Document and it can parsed in raw text which can be used for search also. On Thu, Jul 9, 2020 at 7:08 PM Ed Cable <[email protected]> wrote: > In line with the recent formation of our working group on AI/ML, I wanted > to share a document that Yash was seeking feedback on regarding the usage > of cloud-based AutoML on GCP vs MLKit which is free of charge and deployed > on-prem. This is for the Vision PPI project. > > @Yash Khare <[email protected]> based on this analysis, from a cost > savings and flexibility perspective, I would definitely support MLKit. In > terms of the accuracy and power of the model, it's on part with AutoML? > What implications would not choosing a cloud-based option have for others > trying to train and deploy model in future. Are we able to package up and > containerize what we're doing with MLKit and share with others to deploy in > their cloud or on-prem environments? > > Ed > > ---------- Forwarded message --------- > From: Yash Khare <[email protected]> > Date: Tue, Jul 7, 2020 at 3:37 AM > Subject: Usage of AutoML for the Vision PPI project > To: Nayan Ambali <[email protected]> > Cc: Ed Cable <[email protected]> > > > Hello, > > As we discussed in the last meeting, I made a document that summarizes the > pricing of the AutoML API that we are planning to use. I have also provided > an alternative to it which is free of cost, MLKit. I request you to please > go through the doc and tell me which option should we proceed with. Taking > either option will need around 2-3 weeks for implementation so time isn't a > problem since it is the amount of time I mentioned in my proposal. I have > experimented with both AutoML and MLKit and am familiar with how to > integrate them into an Android app. > > If we decide to go ahead with AutoML, we would need a few free credits for > GCP so that we can keep the models deployed. > > Here is a link to the doc with all the details: > https://docs.google.com/document/d/15zCBE0zsJSVWmH3hzmMm0B2UGXnD-BqqPTjNRMNj8CY/edit?usp=sharing > > I request you and Ed to please go through the document and let me know > with which option to proceed with. > > Looking forward to hearing from you, > Cheers, > Yash. > > > -- > *Ed Cable* > President/CEO, Mifos Initiative > [email protected] | Skype: edcable | Mobile: +1.484.477.8649 > > *Collectively Creating a World of 3 Billion Maries | *http://mifos.org > <http://facebook.com/mifos> <http://www.twitter.com/mifos> > > -- Thanks and regards, Saransh Sharma Research Partner This mail is governed by Muellners® IT policy. The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message, including any attachments. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. All messages sent to and from this e-mail address may be monitored as permitted by applicable law and regulations to ensure compliance with our internal policies and to protect our business. E-mails are not secure and cannot be guaranteed to be error free as they can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by e-mail.
