fix static training bugs

ppcls/metric/metrics.py

@@ -46,7 +46,7 @@ class TopkAcc(AvgMetrics):
         for k in self.topk:
             metric_dict["top{}".format(k)] = paddle.metric.accuracy(
                 x, label, k=k)
-            self.avg_meters["top{}".format(k)].update(metric_dict["top{}".format(k)].numpy()[0], x.shape[0])
+            self.avg_meters["top{}".format(k)].update(metric_dict["top{}".format(k)], x.shape[0])
         return metric_dict
 
 

ppcls/utils/misc.py

@@ -12,6 +12,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import paddle
+
 __all__ = ['AverageMeter']
 
 
@@ -44,6 +46,8 @@ class AverageMeter(object):
 
     @property
     def avg_info(self):
+        if isinstance(self.avg, paddle.Tensor):
+            self.avg = self.avg.numpy()[0]
         return "{}: {:.5f}".format(self.name, self.avg)
 
     @property