remove init_on_cpu from models (#4164)

PaddleCV/icnet/eval.py

@@ -16,7 +16,6 @@ import paddle.fluid as fluid
 import numpy as np
 from utils import add_arguments, print_arguments, get_feeder_data, check_gpu
 from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
-from paddle.fluid.initializer import init_on_cpu
 from icnet import icnet
 import cityscape
 import argparse

PaddleCV/icnet/infer.py

@@ -25,7 +25,6 @@ import paddle
 from icnet import icnet
 from utils import add_arguments, print_arguments, get_feeder_data, check_gpu
 from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
-from paddle.fluid.initializer import init_on_cpu
 import numpy as np
 
 IMG_MEAN = np.array((103.939, 116.779, 123.68), dtype=np.float32)

PaddleCV/icnet/train.py

@@ -26,7 +26,6 @@ import paddle.fluid as fluid
 import numpy as np
 from utils import add_arguments, print_arguments, get_feeder_data, check_gpu
 from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
-from paddle.fluid.initializer import init_on_cpu
 
 if 'ce_mode' in os.environ:
     np.random.seed(10)
@@ -71,9 +70,8 @@ def create_loss(predict, label, mask, num_classes):
 
 def poly_decay():
     global_step = _decay_step_counter()
-    with init_on_cpu():
-        decayed_lr = LEARNING_RATE * (fluid.layers.pow(
-            (1 - global_step / TOTAL_STEP), POWER))
+    decayed_lr = LEARNING_RATE * (fluid.layers.pow(
+        (1 - global_step / TOTAL_STEP), POWER))
     return decayed_lr
 
 

PaddleCV/image_classification/utils/optimizer.py

@@ -20,7 +20,6 @@ import math
 
 import paddle.fluid as fluid
 import paddle.fluid.layers.ops as ops
-from paddle.fluid.initializer import init_on_cpu
 from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
 
 
@@ -30,10 +29,9 @@ def cosine_decay(learning_rate, step_each_epoch, epochs=120):
     """
     global_step = _decay_step_counter()
 
-    with init_on_cpu():
-        epoch = ops.floor(global_step / step_each_epoch)
-        decayed_lr = learning_rate * \
-                     (ops.cos(epoch * (math.pi / epochs)) + 1)/2
+    epoch = ops.floor(global_step / step_each_epoch)
+    decayed_lr = learning_rate * \
+                v(ops.cos(epoch * (math.pi / epochs)) + 1)/2
     return decayed_lr
 
 
@@ -53,17 +51,16 @@ def cosine_decay_with_warmup(learning_rate, step_each_epoch, epochs=120):
     warmup_epoch = fluid.layers.fill_constant(
         shape=[1], dtype='float32', value=float(5), force_cpu=True)
 
-    with init_on_cpu():
-        epoch = ops.floor(global_step / step_each_epoch)
-        with fluid.layers.control_flow.Switch() as switch:
-            with switch.case(epoch < warmup_epoch):
-                decayed_lr = learning_rate * (global_step /
-                                              (step_each_epoch * warmup_epoch))
-                fluid.layers.tensor.assign(input=decayed_lr, output=lr)
-            with switch.default():
-                decayed_lr = learning_rate * \
-                    (ops.cos((global_step - warmup_epoch * step_each_epoch) * (math.pi / (epochs * step_each_epoch))) + 1)/2
-                fluid.layers.tensor.assign(input=decayed_lr, output=lr)
+    epoch = ops.floor(global_step / step_each_epoch)
+    with fluid.layers.control_flow.Switch() as switch:
+        with switch.case(epoch < warmup_epoch):
+            decayed_lr = learning_rate * (global_step /
+                                          (step_each_epoch * warmup_epoch))
+            fluid.layers.tensor.assign(input=decayed_lr, output=lr)
+        with switch.default():
+            decayed_lr = learning_rate * \
+                (ops.cos((global_step - warmup_epoch * step_each_epoch) * (math.pi / (epochs * step_each_epoch))) + 1)/2
+            fluid.layers.tensor.assign(input=decayed_lr, output=lr)
     return lr
 
 def exponential_decay_with_warmup(learning_rate, step_each_epoch, decay_epochs, decay_rate=0.97, warm_up_epoch=5.0):
@@ -80,17 +77,16 @@ def exponential_decay_with_warmup(learning_rate, step_each_epoch, decay_epochs,
     warmup_epoch = fluid.layers.fill_constant(
         shape=[1], dtype='float32', value=float(warm_up_epoch), force_cpu=True)
 
-    with init_on_cpu():
-        epoch = ops.floor(global_step / step_each_epoch)
-        with fluid.layers.control_flow.Switch() as switch:
-            with switch.case(epoch < warmup_epoch):
-                decayed_lr = learning_rate * (global_step / (step_each_epoch * warmup_epoch))
-                fluid.layers.assign(input=decayed_lr, output=lr)
-            with switch.default():
-                div_res = (global_step - warmup_epoch * step_each_epoch) / decay_epochs
-                div_res = ops.floor(div_res)
-                decayed_lr = learning_rate * (decay_rate ** div_res)
-                fluid.layers.assign(input=decayed_lr, output=lr)
+    epoch = ops.floor(global_step / step_each_epoch)
+    with fluid.layers.control_flow.Switch() as switch:
+        with switch.case(epoch < warmup_epoch):
+            decayed_lr = learning_rate * (global_step / (step_each_epoch * warmup_epoch))
+            fluid.layers.assign(input=decayed_lr, output=lr)
+        with switch.default():
+            div_res = (global_step - warmup_epoch * step_each_epoch) / decay_epochs
+            div_res = ops.floor(div_res)
+            decayed_lr = learning_rate * (decay_rate ** div_res)
+            fluid.layers.assign(input=decayed_lr, output=lr)
 
     return lr
 

PaddleCV/ocr_recognition/crnn_ctc_model.py

@@ -16,7 +16,6 @@ from __future__ import division
 from __future__ import print_function
 import paddle.fluid as fluid
 from paddle.fluid.layers.learning_rate_scheduler import _decay_step_counter
-from paddle.fluid.initializer import init_on_cpu
 import math
 import six