Merge pull request #4968 from LDOUBLEV/sdmgr

add ips

configs/kie/kie_unet_sdmgr.yml

 Global:
   use_gpu: True
-  epoch_num: 300
+  epoch_num: 60
   log_smooth_window: 20
   print_batch_step: 50
   save_model_dir: ./output/kie_5/
@@ -13,7 +13,7 @@ Global:
   #    you should set load_static_weights as False.
   load_static_weights: False
   cal_metric_during_train: False
-  pretrained_model: ./output/kie_4/best_accuracy
+  pretrained_model: 
   checkpoints:
   save_inference_dir:
   use_visualdl: False
@@ -108,4 +108,4 @@ Eval:
     shuffle: False
     drop_last: False
     batch_size_per_card: 1 # must be 1
-    num_workers: 4
+    num_workers: 4
\ No newline at end of file

tools/infer_kie.py

@@ -33,8 +33,9 @@ import paddle
 
 from ppocr.data import create_operators, transform
 from ppocr.modeling.architectures import build_model
-from ppocr.utils.save_load import init_model
+from ppocr.utils.save_load import load_model
 import tools.program as program
+import time
 
 
 def read_class_list(filepath):
@@ -80,7 +81,8 @@ def draw_kie_result(batch, node, idx_to_cls, count):
     vis_img = np.ones((h, w * 3, 3), dtype=np.uint8) * 255
     vis_img[:, :w] = img
     vis_img[:, w:] = pred_img
-    save_kie_path = os.path.dirname(config['Global']['save_res_path']) + "/kie_results/"
+    save_kie_path = os.path.dirname(config['Global'][
+        'save_res_path']) + "/kie_results/"
     if not os.path.exists(save_kie_path):
         os.makedirs(save_kie_path)
     save_path = os.path.join(save_kie_path, str(count) + ".png")
@@ -93,7 +95,7 @@ def main():
 
     # build model
     model = build_model(config['Architecture'])
-    init_model(config, model, logger)
+    load_model(config, model)
 
     # create data ops
     transforms = []
@@ -111,10 +113,15 @@ def main():
         os.makedirs(os.path.dirname(save_res_path))
 
     model.eval()
+
+    warmup_times = 0
+    count_t = []
     with open(save_res_path, "wb") as fout:
         with open(config['Global']['infer_img'], "rb") as f:
             lines = f.readlines()
             for index, data_line in enumerate(lines):
+                if index == 10:
+                    warmup_t = time.time()
                 data_line = data_line.decode('utf-8')
                 substr = data_line.strip("\n").split("\t")
                 img_path, label = data_dir + "/" + substr[0], substr[1]
@@ -122,16 +129,23 @@ def main():
                 with open(data['img_path'], 'rb') as f:
                     img = f.read()
                     data['image'] = img
+                st = time.time()
                 batch = transform(data, ops)
                 batch_pred = [0] * len(batch)
                 for i in range(len(batch)):
                     batch_pred[i] = paddle.to_tensor(
                         np.expand_dims(
                             batch[i], axis=0))
+                st = time.time()
                 node, edge = model(batch_pred)
                 node = F.softmax(node, -1)
+                count_t.append(time.time() - st)
                 draw_kie_result(batch, node, idx_to_cls, index)
     logger.info("success!")
+    logger.info("It took {} s for predict {} images.".format(
+        np.sum(count_t), len(count_t)))
+    ips = len(count_t[warmup_times:]) / np.sum(count_t[warmup_times:])
+    logger.info("The ips is {} images/s".format(ips))
 
 
 if __name__ == '__main__':