mirrors
/
fast-reid
mirror of https://github.com/JDAI-CV/fast-reid.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

optimize jaccard distance computation and the ranking

pull/586/head
Marah Gamdou 2021-10-08 19:42:32 +02:00
parent 10a5f38aaa
commit 8969f1bd3a
2 changed files with 11 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
fastreid/evaluation/rank.py
View File

@ -81,8 +81,7 @@ def eval_cuhk03(distmat, q_pids, g_pids, q_camids, g_camids, max_rank):
# compute AP # compute AP
num_rel = raw_cmc.sum() num_rel = raw_cmc.sum()
tmp_cmc = raw_cmc.cumsum() tmp_cmc = raw_cmc.cumsum()
tmp_cmc = [x / (i + 1.) for i, x in enumerate(tmp_cmc)] tmp_cmc = (tmp_cmc/np.arange(1,len(tmp_cmc)+1)) * raw_cmc
tmp_cmc = np.asarray(tmp_cmc) * raw_cmc
AP = tmp_cmc.sum() / num_rel AP = tmp_cmc.sum() / num_rel
all_AP.append(AP) all_AP.append(AP)
num_valid_q += 1. num_valid_q += 1.

19
fastreid/utils/compute_dist.py
View File

@ -50,8 +50,8 @@ def build_dist(feat_1: torch.Tensor, feat_2: torch.Tensor, metric: str = "euclid
elif metric == "jaccard": elif metric == "jaccard":
feat = torch.cat((feat_1, feat_2), dim=0) feat = torch.cat((feat_1, feat_2), dim=0)
dist = compute_jaccard_distance(feat, k1=kwargs["k1"], k2=kwargs["k2"], search_option=0) dist = compute_jaccard_distance(feat,feat_1.shape[0],feat_2.shape[0], k1=kwargs["k1"], k2=kwargs["k2"], search_option=0)
return dist[: feat_1.size(0), feat_1.size(0):] return dist
def k_reciprocal_neigh(initial_rank, i, k1): def k_reciprocal_neigh(initial_rank, i, k1):
@ -62,7 +62,7 @@ def k_reciprocal_neigh(initial_rank, i, k1):
@torch.no_grad() @torch.no_grad()
def compute_jaccard_distance(features, k1=20, k2=6, search_option=0, fp16=False): def compute_jaccard_distance(features, N_feat_1, N_feat_2, k1=20, k2=6, search_option=0, fp16=False):
if search_option < 3: if search_option < 3:
# torch.cuda.empty_cache() # torch.cuda.empty_cache()
features = features.cuda() features = features.cuda()
@ -153,17 +153,18 @@ def compute_jaccard_distance(features, k1=20, k2=6, search_option=0, fp16=False)
for i in range(N): for i in range(N):
invIndex.append(np.where(V[:, i] != 0)[0]) # len(invIndex)=all_num invIndex.append(np.where(V[:, i] != 0)[0]) # len(invIndex)=all_num
jaccard_dist = np.zeros((N, N), dtype=mat_type) jaccard_dist = np.zeros((N_feat_1, N_feat_2), dtype=mat_type)
for i in range(N): for i in range(N_feat_1):
temp_min = np.zeros((1, N), dtype=mat_type) temp_min = np.zeros((1, N), dtype=mat_type)
indNonZero = np.where(V[i, :] != 0)[0] indNonZero = np.where(V[i, :] != 0)[0]
indImages = [invIndex[ind] for ind in indNonZero] indImages = [invIndex[ind] for ind in indNonZero]
for j in range(len(indNonZero)): for j in range(len(indNonZero)):
temp_min[0, indImages[j]] = temp_min[0, indImages[j]] + np.minimum( if indImages[j]>=N_feat_1:
V[i, indNonZero[j]], V[indImages[j], indNonZero[j]] temp_min[0, indImages[j]] = temp_min[0, indImages[j]] + np.minimum(
) V[i, indNonZero[j]], V[indImages[j], indNonZero[j]]
)
jaccard_dist[i] = 1 - temp_min / (2 - temp_min) jaccard_dist[i] = (1 - temp_min / (2 - temp_min))[:,N_feat_1:]
del invIndex, V del invIndex, V