I have an adjacency matrix like:

[[  0.,  15.,   0.,   7.,  10.,   0.],[ 15.,   0.,   9.,  11.,   0.,   9.],[  0.,   9.,   0.,   0.,  12.,   7.],[  7.,  11.,   0.,   0.,   8.,  14.],[ 10.,   0.,  12.,   8.,   0.,   8.],[  0.,   9.,   7.,  14.,   8.,   0.]]

How can I convert it to an adjacency list like this one down here?

graph = {'1': [{'2':'15'}, {'4':'7'}, {'5':'10'}],
'2': [{'3':'9'}, {'4':'11'}, {'6':'9'}],
'3': [{'5':'12'}, {'6':'7'}],
'4': [{'5':'8'}, {'6':'14'}],
'5': [{'6':'8'}]}

?

Keep a list of already added edges in a set edges. Those edges are stored in a frozenset, so already added pairs are not replicated.

Then build your graph by enumerating the outer list with a starting index of one, then the inner list also with a starting index of one. Zero valued entries are eliminated with the if condition on values:

from collections import defaultdict
from pprint import pprintl =[[  0.,  15.,   0.,   7.,  10.,   0.],[ 15.,   0.,   9.,  11.,   0.,   9.],[  0.,   9.,   0.,   0.,  12.,   7.],[  7.,  11.,   0.,   0.,   8.,  14.],[ 10.,   0.,  12.,   8.,   0.,   8.],[  0.,   9.,   7.,  14.,   8.,   0.]]    graph = defaultdict(list)
edges = set()for i, v in enumerate(l, 1):for j, u in enumerate(v, 1):if u != 0 and frozenset([i, j]) not in edges:edges.add(frozenset([i, j]))graph[i].append({j: u})pprint(graph)
# {1: [{2: 15.0}, {4: 7.0}, {5: 10.0}],
#  2: [{3: 9.0}, {4: 11.0}, {6: 9.0}],
#  3: [{5: 12.0}, {6: 7.0}],
#  4: [{5: 8.0}, {6: 14.0}],
#  5: [{6: 8.0}]}

Using a defaultdict which takes a list as default value will help build the list-valued dictionary on the fly.