#print("num of non 0 elements in expected : ", expected[expectedIndices].nonzero().numel())
#print("num of non 0 elements in result : ", result[expectedIndices].nonzero().numel())
print('csvec accuracy :', csvec_accuracy)
print('expected :', expected)
print('result :', result)
for h_i, h in enumerate(hVals):
result = torch.zeros(len(summed), device=device)
h_curr = int(h * k)
w_0 = TopHCS(d=d,
c=c,
r=r,
h=h_curr,
numBlocks=numBlocks,
device=device)
w_0.store(vecs[0])
print(w_0.topH)
print("")
w_1 = TopHCS(d=d,
c=c,
r=r,
h=h_curr,
numBlocks=numBlocks,
device=device)
w_1.store(vecs[1])
print(w_0.topH)
print("")
w_2 = TopHCS(d=d,
c=c,
diff = recovered_1[recovered_1 != 0] - expected[recovered_1 != 0]
L1_1[k_i] = torch.median(torch.abs(diff))
L2_1[k_i] = torch.median(diff**2)
for h_i, hVal in enumerate(hVals):
h = int(k * hVal)
workers = [] #reusing
for vec in vecs:
w = TopHCS(h=h,
d=d,
c=c,
r=r,
numBlocks=numBlocks,
device=device)
w.store(vec)
workers.append(w)
assert (len(workers) == len(vecs))
recovered_2 = TopHCS.topKSum(workers, k)
indexAcc_2[
h_i,
k_i] = (expected[expectedIndices] *
recovered_2[expectedIndices]).nonzero().numel() / k
diff = recovered_2[recovered_2 != 0] - expected[recovered_2 != 0]
L1_2[h_i, k_i] = torch.median(torch.abs(diff))
L2_2[h_i, k_i] = torch.median(diff**2)
#indexAcc = (expected == recovered).nonzero().numel() / k
diff = recovered_1[diff] - summed[
diff] #ugly code, but overwriting to reduce vars
L1_1[k_i] = torch.median(torch.abs(diff))
L2_1[k_i] = torch.median(diff**2)
for h_i, hVal in enumerate(hVals):
h = int(k * hVal)
workers = [] #re-initialize
for v in vecs:
w = TopHCS(h=h,
d=d,
c=c,
r=r,
numBlocks=numBlocks,
device=device)
w.store(v)
workers.append(w)
assert (len(workers) == len(vecs))
recovered_2 = TopHCS.topKSum(workers, k)
print("Heap @ exp Indices : ", recovered_2[expectedIndices])
indexAcc_2[
h_i,
k_i] = (expected[expectedIndices] *
recovered_2[expectedIndices]).nonzero().numel() / k
diff = torch.sort(
torch.nonzero(recovered_2.view(-1).data).squeeze())[0]
assert (len(diff) == k)
print("Heap @ reported Indices : ", recovered_2[diff])
w_0 += vecs[0]
print("worker added")
w_1 = CSVec(d=d, c=c, r=r, numBlocks=numBlocks, device=device)
w_1 += vecs[1]
print("worker added")
workersSummed = w_0 + w_1
result = workersSummed.unSketch(k)
csvecAcc[k_i] = (expected[expectedIndices] * result[expectedIndices]).nonzero().numel() / k
print("k = {}".format(k))
csvecL2[k_i] = (torch.sum((result - expected)**2))**0.5
for h_i, h in enumerate(hVals):
result = torch.zeros(len(summed), device=device)
h_curr = int(h*k)
w_0 = TopHCS(d=d, c=c, r=r, h=h_curr, numBlocks=numBlocks, device=device)
w_0.store(vecs[0])
print("worker added")
w_1 = TopHCS(d=d, c=c, r=r, h=h_curr, numBlocks=numBlocks, device=device)
w_1.store(vecs[1])
print("worker added")
workers = [w_0, w_1]
result = TopHCS.topKSum(workers, k)
topHCSAcc[h_i, k_i] = (expected[expectedIndices] * result[expectedIndices]).nonzero().numel() / k
print("h_curr = {}".format(h_curr))
topHCSL2[h_i, k_i] = (torch.sum((result - expected)**2))**0.5
print('topHCS accuracy :', '\n', topHCSAcc)
print('CSVec accuracy :', '\n', csvecAcc)
numColors = len(hVals) + 1
colors = plt.cm.plasma(np.linspace(0,1,numColors))
axs[p, 0].set_xlabel("k")
print("worker added")
result = w_0.unSketch(k)
csvecAcc[k_i] = (expected[expectedIndices] *
result[expectedIndices]).nonzero().numel() / k
print("k = {}".format(k))
csvecL2[k_i] = (torch.sum((result - expected)**2))**0.5
for h_i, h in enumerate(hVals):
result = torch.zeros(len(summed), device=device)
w_0 = TopHCS(d=d,
c=c,
r=r,
h=h,
numBlocks=numBlocks,
device=device)
w_0.store(vecs[0])
print("worker added")
workers = [w_0]
result = TopHCS.topKSum(workers, k)
topHCSAcc[
h_i, k_i] = (expected[expectedIndices] *
result[expectedIndices]).nonzero().numel() / k
print("h_curr = {}".format(h_curr))
topHCSL2[h_i, k_i] = (torch.sum((result - expected)**2))**0.5
print('topHCS accuracy :', '\n', topHCSAcc)
print('CSVec accuracy :', '\n', csvecAcc)
numColors = len(hVals) + 1
colors = plt.cm.plasma(np.linspace(0, 1, numColors))
axs[p, 0].set_xlabel("k")
axs[p, 0].set_ylabel("index accuracy")
