ZF/insightface
1
0
Fork 0
mirror of https://github.com/deepinsight/insightface.git synced 2026-05-21 17:17:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

change model save logic

Browse Source
This commit is contained in:
Jia Guo
2017-12-07 14:37:27 +08:00
parent 97e98f654f
commit b0a2c35eff
2 changed files with 13 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/train.sh
View File

@@ -11,5 +11,5 @@ MODELDIR="../model-$NETWORK-$JOB"
mkdir -p "$MODELDIR"
PREFIX="$MODELDIR/model"
LOGFILE="$MODELDIR/log"
CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir $DATA_DIR --network "$NETWORK" --loss-type 0 --lr 0.1 --prefix "$PREFIX" --per-batch-size 128 --image-size '112,112' --version-input 1 --version-output E --version-unit 3 > "$LOGFILE" 2>&1 &
CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir $DATA_DIR --network "$NETWORK" --loss-type 0 --lr 0.1 --prefix "$PREFIX" --per-batch-size 128 --image-size '112,112' --version-input 1 --version-output E --version-unit 3 --use-se > "$LOGFILE" 2>&1 &

22
src/train_softmax.py
View File

@@ -504,8 +504,9 @@ def train_net(args):
print('VACC: %f'%(acc_value))
highest_acc = [0.0]
last_save_acc = [0.0]
highest_acc = []
for i in xrange(len(ver_list)):
highest_acc.append(0.0)
global_step = [0]
save_step = [0]
if len(args.lr_steps)==0:
@@ -534,15 +535,17 @@ def train_net(args):
if mbatch>=0 and mbatch%args.verbose==0:
acc_list = ver_test(mbatch)
acc = acc_list[0]
save_step[0]+=1
msave = save_step[0]
do_save = False
if acc>=highest_acc[0]:
highest_acc[0] = acc
if acc>=0.99:
do_save = True
if mbatch>lr_steps[-1] and mbatch%10000==0:
lfw_score = acc_list[0]
for i in xrange(len(acc_list)):
acc = acc_list[i]
if acc>=highest_acc[i]:
highest_acc[i] = acc
if lfw_score>=0.99:
do_save = True
if args.loss_type==1 and mbatch>lr_steps[-1] and mbatch%10000==0:
do_save = True
if do_save:
print('saving', msave, acc)
@@ -555,13 +558,12 @@ def train_net(args):
# X = np.concatenate(embeddings_list, axis=0)
# print('saving lfw npy', X.shape)
# np.save(lfw_npy, X)
print('[%d]Accuracy-Highest: %1.5f'%(mbatch, highest_acc[0]))
#print('[%d]Accuracy-Highest: %1.5f'%(mbatch, highest_acc[0]))
if mbatch<=args.beta_freeze:
_beta = args.beta
else:
move = max(0, mbatch-args.beta_freeze)
_beta = max(args.beta_min, args.beta*math.pow(1+args.gamma*move, -1.0*args.power))
#_beta = max(args.beta_min, args.beta*math.pow(0.7, move//500))
#print('beta', _beta)
os.environ['BETA'] = str(_beta)