Build a machine learning model for stock chart pattern recognition

I want to develop a ML model to recognize a certain TRIANGLE chart pattern. This is a stock price pattern that basically shows how the price consolidates with time i.e. it has tighter and tighter range (high to low) until it gets very tight on the right hand side and then we have a breakout (usually to the upside but not always) . The breakout happens before the end point of the chart pattern. I want to develop a model that will recognise such a pattern as it develops with time. See attachment 00 Fully developed chart pattern
This pattern can develop on different time frames i.e. daily, weekly, hourly,5 minute etc. and takes different time to develop.

The data format to work with looks like the table below. I can provide many examples of the pattern for teaching but you need to be able to prepare the data for processing. Attachment 01 Data format example
What the model should be able to do is when provided with a new data set the model should analyze the latest data (it can be the latest 10 or the latest 150 bars) and classify if it is developing into the pattern of interest or not.

If the current developing shape matches the triangle pattern but is incomplete, then the model should be able to calculate the end point of the pattern in time (by projecting the lines and calculating the distance in time intervals).

To make sure you have read and understood the assignment I will only consider bids starting wth the code word "Carrot".

Thanks