I was expecting a contest analysis by now but didn't see one so I'll try this way. Though I didn't participate in round 1A I tried to solve the Good Luck problem but only get ~10% correct with the parameters of the second input and wanted to share my approach and ask for comments. First I find all distinct combinations of selecting a set of digits, how many of each digit (~18K for second input) For each combination I find all distinct subsets this puts some strain on the memory (had to increase maximum heap size) but still fits in memory and can be calculated in a few seconds. for each subset I store the product and calculate combinatoriclly how many such subsets exist and store a map with the product as the key on each set of digits merging subsets with the same product. Then for each input of a list of products we go over all 18K possible digit combinations and lookup the product get the number of matching subsets and multiply them together selecting the combination with highest value, and product which has zero possible subsets will obviously zero the product and prevent that combination from being selected. This approach however gets 10% accuracy which is double then randomly selecting a valid combination but doesn't meat the required 14%.
Is this approach valid and I have some flaw in my combinatorics or does this approach fundamentally not select the most likely combination of digits? Can someone explain a better aproach? -- You received this message because you are subscribed to the Google Groups "Google Code Jam" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msg/google-code/-/3_scbrvjxL8J. For more options, visit https://groups.google.com/groups/opt_out.
