Bienqq
/
car-detection-bayes
2
1
Fork 0
Code Issues Pull Requests Releases Wiki Activity

updates

This commit is contained in:
Glenn Jocher 2018-11-17 12:51:22 +01:00
parent 07f15b68d3
commit dd7c3d2455
2 changed files with 19 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
models.py
View File

@ -123,16 +123,16 @@ class YOLOLayer(nn.Module):
y = torch.sigmoid(p[..., 1]) # Center y y = torch.sigmoid(p[..., 1]) # Center y
# Width and height (yolo method) # Width and height (yolo method)
w = p[..., 2] # Width # w = p[..., 2] # Width
h = p[..., 3] # Height # h = p[..., 3] # Height
width = torch.exp(w.data) * self.anchor_w # width = torch.exp(w.data) * self.anchor_w
height = torch.exp(h.data) * self.anchor_h # height = torch.exp(h.data) * self.anchor_h
# Width and height (power method) # Width and height (power method)
# w = torch.sigmoid(p[..., 2]) # Width w = torch.sigmoid(p[..., 2]) # Width
# h = torch.sigmoid(p[..., 3]) # Height h = torch.sigmoid(p[..., 3]) # Height
# width = ((w.data * 2) ** 2) * self.anchor_w width = ((w.data * 2) ** 2) * self.anchor_w
# height = ((h.data * 2) ** 2) * self.anchor_h height = ((h.data * 2) ** 2) * self.anchor_h
# Add offset and scale with anchors (in grid space, i.e. 0-13) # Add offset and scale with anchors (in grid space, i.e. 0-13)
pred_boxes = FT(bs, self.nA, nG, nG, 4) pred_boxes = FT(bs, self.nA, nG, nG, 4)
@ -168,13 +168,13 @@ class YOLOLayer(nn.Module):
if nM > 0: if nM > 0:
lx = k * MSELoss(x[mask], tx[mask]) lx = k * MSELoss(x[mask], tx[mask])
ly = k * MSELoss(y[mask], ty[mask]) ly = k * MSELoss(y[mask], ty[mask])
lw = (k * 0.7) * MSELoss(w[mask], tw[mask]) lw = (k * 1) * MSELoss(w[mask], tw[mask])
lh = (k * 0.7) * MSELoss(h[mask], th[mask]) lh = (k * 1) * MSELoss(h[mask], th[mask])
# lconf = k * BCEWithLogitsLoss(pred_conf[mask], mask[mask].float()) # lconf = k * BCEWithLogitsLoss(pred_conf[mask], mask[mask].float())
lconf = (k * 5) * BCEWithLogitsLoss(pred_conf, mask.float()) lconf = (k * 10) * BCEWithLogitsLoss(pred_conf, mask.float())
lcls = (k / 20) * CrossEntropyLoss(pred_cls[mask], torch.argmax(tcls, 1)) lcls = (k / 10) * CrossEntropyLoss(pred_cls[mask], torch.argmax(tcls, 1))
# lcls = k * BCEWithLogitsLoss(pred_cls[mask], tcls.float()) # lcls = k * BCEWithLogitsLoss(pred_cls[mask], tcls.float())
else: else:
lx, ly, lw, lh, lcls, lconf = FT([0]), FT([0]), FT([0]), FT([0]), FT([0]), FT([0]) lx, ly, lw, lh, lcls, lconf = FT([0]), FT([0]), FT([0]), FT([0]), FT([0]), FT([0])

13
utils/utils.py
View File

@ -259,12 +259,12 @@ def build_targets(pred_boxes, pred_conf, pred_cls, target, anchor_wh, nA, nC, nG
ty[b, a, gj, gi] = gy - gj.float() ty[b, a, gj, gi] = gy - gj.float()
# Width and height (yolo method) # Width and height (yolo method)
tw[b, a, gj, gi] = torch.log(gw / anchor_wh[a, 0]) # tw[b, a, gj, gi] = torch.log(gw / anchor_wh[a, 0])
th[b, a, gj, gi] = torch.log(gh / anchor_wh[a, 1]) # th[b, a, gj, gi] = torch.log(gh / anchor_wh[a, 1])
# Width and height (power method) # Width and height (power method)
# tw[b, a, gj, gi] = torch.sqrt(gw / anchor_wh[a, 0]) / 2 tw[b, a, gj, gi] = torch.sqrt(gw / anchor_wh[a, 0]) / 2
# th[b, a, gj, gi] = torch.sqrt(gh / anchor_wh[a, 1]) / 2 th[b, a, gj, gi] = torch.sqrt(gh / anchor_wh[a, 1]) / 2
# One-hot encoding of label # One-hot encoding of label
tcls[b, a, gj, gi, tc] = 1 tcls[b, a, gj, gi, tc] = 1
@ -436,8 +436,9 @@ def plot_results():
import matplotlib.pyplot as plt import matplotlib.pyplot as plt
plt.figure(figsize=(16, 8)) plt.figure(figsize=(16, 8))
s = ['X', 'Y', 'Width', 'Height', 'Objectness', 'Classification', 'Total Loss', 'Precision', 'Recall', 'mAP'] s = ['X', 'Y', 'Width', 'Height', 'Objectness', 'Classification', 'Total Loss', 'Precision', 'Recall', 'mAP']
for f in ('results.txt',): for f in ('results.txt',
results = np.loadtxt(f, usecols=[2, 3, 4, 5, 6, 7, 8, 9, 10]).T ):
results = np.loadtxt(f, usecols=[2, 3, 4, 5, 6, 7, 8, 9, 10]).T # column 16 is mAP
for i in range(9): for i in range(9):
plt.subplot(2, 5, i + 1) plt.subplot(2, 5, i + 1)
plt.plot(results[i, :250], marker='.', label=f) plt.plot(results[i, :250], marker='.', label=f)