Hey there, I hope you are fine and with good health.
I am a certified Python (PCAPP) / Machine Learning (ML) Programmer with five years of industrial experience. I understand that you want to recognise drinks in a vendor machine through images. We can two SOTA DL approaches: (i) Single Shot Detection (architectures such as RetinaNet, YOLOv3, etc.), (ii) Using a Region Proposal Network to find objects in an image and a second CNN backbone network to fine-tune the generated proposals to make predictions (two-stage networks such as RCNN, Faster RCNN). Presuming you have the dataset we can use any of the two to tackle the problem at hand. I shall also label them accoring to the sequence and row-wise. Let's chat to discuss more of your requirements clearly and get it started.
#Skills:
> Image processing, Classification, Clustering, Segmentation, Localisation & Detection, OCR.
> Time Series Analysis, Multi-Step Time series Analysis, RNNs/LSTMs.
> Training and Mentoring, Data Visualisation, Learning Curves
> ML Algorithms, ML Model Development, Deep Learning and Data/Text Mining, NLP
> Video Analytics ( Anomalies detection, Video Classification & Labeling )
> Audio Analytics ( Speech Recognition, Speaker Diarization & ASR )
> Deploying ML models as web services & on edge devices as embedded models.
> AWS (EC2), Heroku, Azure & GCP
> Dockers
#Core Libraries:
> tensorflow-keras, pytorch, opencv, pyedflib, scikit-learn, PIL, matplotlib/seaborn, tesseract, numpy, pandas, etc.