从for循环转向矩阵计算：大规模条件验证的向量化实现咨询

I've written code to validate if a list of objects meets certain requirements. First, here's my object class:

class S:
    def __init__(self, f, t, tf, timeline):
        self.f = f
        self.t = t
        self.tf = tf
        self.timeline = timeline

I have a validation function that takes a list of N S objects and returns True/False based on whether they meet a specific condition:

def c1(input1):
    # input1 is a list like [S_1, ..., S_N]
    if condition_c1_valid:
        return True
    else:
        return False

Right now, I'm generating all possible combinations of these objects and checking them one by one:

import itertools

# Generate all possible individual S objects
possible_objects = [S(f, t, tf, timeline) 
                    for f in [...] 
                    for t in [...] 
                    # ... other parameter loops ...
                   ]

# Generate combinations (here, 5 elements per input)
inputs_to_check = list(itertools.combinations_with_replacement(possible_objects, 5))

results = []
for inp in inputs_to_check:
    if c1(inp):
        results.append(inp)

The problem is that the scale of inputs to check is enormous—even with generator optimizations, it's handling large sets with 6/7 nested loops, leading to huge computational overhead.

I'm wondering if I can vectorize this entire validation process using a matrix/array approach to run it on a GPU with TensorFlow. My rough pseudocode idea looks like this:

# Hypothetical structure
Data[input, c1, ..., cN]
return all(inputs where (c1 AND ... AND cN) is True)

Is this feasible? Could you provide a concrete example of how to implement this? My goal is to port this computation to GPU to speed things up significantly.

内容的提问来源于stack exchange，提问作者Mathieu