adapt import

examples/bert/bert_classifier.py

@@ -14,14 +14,14 @@
 """BERT fine-tuning in Paddle Dygraph Mode."""
 
 import paddle.fluid as fluid
-from hapi.metrics import Accuracy
-from hapi.configure import Config
-from hapi.text.bert import BertEncoder
+from paddle.incubate.hapi.metrics import Accuracy
+from paddle.incubate.hapi.configure import Config
+from paddle.incubate.hapi.text.bert import BertEncoder
 from paddle.fluid.dygraph import Linear, Layer
-from hapi.loss import SoftmaxWithCrossEntropy
-from hapi.model import set_device, Model, Input
-import hapi.text.tokenizer.tokenization as tokenization
-from hapi.text.bert import BertConfig, BertDataLoader, BertInputExample, make_optimizer
+from paddle.incubate.hapi.loss import SoftmaxWithCrossEntropy
+from paddle.incubate.hapi.model import set_device, Model, Input
+import paddle.incubate.hapi.text.tokenizer.tokenization as tokenization
+from paddle.incubate.hapi.text.bert import BertConfig, BertDataLoader, BertInputExample, make_optimizer
 
 
 class ClsModelLayer(Model):
@@ -157,7 +157,8 @@ def main():
         labels,
         device=device)
 
-    cls_model.bert_layer.load("./bert_uncased_L-12_H-768_A-12/bert", reset_optimizer=True)
+    cls_model.bert_layer.load(
+        "./bert_uncased_L-12_H-768_A-12/bert", reset_optimizer=True)
 
     # do train
     cls_model.fit(train_data=train_dataloader.dataloader,

examples/bert_leveldb/bert_classifier.py

@@ -14,14 +14,14 @@
 """BERT fine-tuning in Paddle Dygraph Mode."""
 
 import paddle.fluid as fluid
-from hapi.metrics import Accuracy
-from hapi.configure import Config
-from hapi.text.bert import BertEncoder
+from paddle.incubate.hapi.metrics import Accuracy
+from paddle.incubate.hapi.configure import Config
+from paddle.incubate.hapi.text.bert import BertEncoder
 from paddle.fluid.dygraph import Linear, Layer
-from hapi.loss import SoftmaxWithCrossEntropy
-from hapi.model import set_device, Model, Input
-import hapi.text.tokenizer.tokenization as tokenization
-from hapi.text.bert import BertConfig, BertDataLoader, BertInputExample, make_optimizer
+from paddle.incubate.hapi.loss import SoftmaxWithCrossEntropy
+from paddle.incubate.hapi.model import set_device, Model, Input
+import paddle.incubate.hapi.text.tokenizer.tokenization as tokenization
+from paddle.incubate.hapi.text.bert import BertConfig, BertDataLoader, BertInputExample, make_optimizer
 
 
 class ClsModelLayer(Model):
@@ -159,7 +159,8 @@ def main():
         labels,
         device=device)
 
-    cls_model.bert_layer.load("./bert_uncased_L-12_H-768_A-12/bert", reset_optimizer=True)
+    cls_model.bert_layer.load(
+        "./bert_uncased_L-12_H-768_A-12/bert", reset_optimizer=True)
 
     # do train
     cls_model.fit(train_data=train_dataloader.dataloader,

examples/bmn/bmn_metric.py

@@ -20,7 +20,7 @@ import json
 
 sys.path.append('../')
 
-from hapi.metrics import Metric
+from paddle.incubate.hapi.metrics import Metric
 from bmn_utils import boundary_choose, bmn_post_processing
 
 

examples/bmn/eval.py

@@ -18,7 +18,7 @@ import sys
 import logging
 import paddle.fluid as fluid
 
-from hapi.model import set_device, Input
+from paddle.incubate.hapi.model import set_device, Input
 
 from modeling import bmn, BmnLoss
 from bmn_metric import BmnMetric

examples/bmn/modeling.py

@@ -17,9 +17,9 @@ from paddle.fluid import ParamAttr
 import numpy as np
 import math
 
-from hapi.model import Model
-from hapi.loss import Loss
-from hapi.download import get_weights_path_from_url
+from paddle.incubate.hapi.model import Model
+from paddle.incubate.hapi.loss import Loss
+from paddle.incubate.hapi.download import get_weights_path_from_url
 
 __all__ = ["BMN", "BmnLoss", "bmn"]
 

examples/bmn/predict.py

@@ -18,7 +18,7 @@ import os
 import logging
 import paddle.fluid as fluid
 
-from hapi.model import set_device, Input
+from paddle.incubate.hapi.model import set_device, Input
 
 from modeling import bmn, BmnLoss
 from bmn_metric import BmnMetric

examples/bmn/reader.py

@@ -21,7 +21,7 @@ import sys
 
 sys.path.append('../')
 
-from hapi.distributed import DistributedBatchSampler
+from paddle.incubate.hapi.distributed import DistributedBatchSampler
 from paddle.io import Dataset, DataLoader
 
 logger = logging.getLogger(__name__)

examples/bmn/train.py

@@ -18,7 +18,7 @@ import logging
 import sys
 import os
 
-from hapi.model import set_device, Input
+from paddle.incubate.hapi.model import set_device, Input
 
 from reader import BmnDataset
 from config_utils import *

examples/cyclegan/cyclegan.py

@@ -19,8 +19,8 @@ from __future__ import print_function
 import numpy as np
 
 import paddle.fluid as fluid
-from hapi.model import Model
-from hapi.loss import Loss
+from paddle.incubate.hapi.model import Model
+from paddle.incubate.hapi.loss import Loss
 
 from layers import ConvBN, DeConvBN
 

examples/cyclegan/infer.py

@@ -25,7 +25,7 @@ from PIL import Image
 from scipy.misc import imsave
 
 import paddle.fluid as fluid
-from hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.model import Model, Input, set_device
 
 from check import check_gpu, check_version
 from cyclegan import Generator, GeneratorCombine

examples/cyclegan/test.py

@@ -22,7 +22,7 @@ import numpy as np
 from scipy.misc import imsave
 
 import paddle.fluid as fluid
-from hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.model import Model, Input, set_device
 
 from check import check_gpu, check_version
 from cyclegan import Generator, GeneratorCombine

examples/cyclegan/train.py

@@ -24,7 +24,7 @@ import time
 
 import paddle
 import paddle.fluid as fluid
-from hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.model import Model, Input, set_device
 
 from check import check_gpu, check_version
 from cyclegan import Generator, Discriminator, GeneratorCombine, GLoss, DLoss
@@ -78,12 +78,12 @@ def main():
     g_AB.prepare(inputs=[input_A], device=FLAGS.device)
     g_BA.prepare(inputs=[input_B], device=FLAGS.device)
 
-    g.prepare(g_optimizer, GLoss(), inputs=[input_A, input_B],
-        device=FLAGS.device)
-    d_A.prepare(da_optimizer, DLoss(), inputs=[input_B, fake_B],
-        device=FLAGS.device)
-    d_B.prepare(db_optimizer, DLoss(), inputs=[input_A, fake_A],
-        device=FLAGS.device)
+    g.prepare(
+        g_optimizer, GLoss(), inputs=[input_A, input_B], device=FLAGS.device)
+    d_A.prepare(
+        da_optimizer, DLoss(), inputs=[input_B, fake_B], device=FLAGS.device)
+    d_B.prepare(
+        db_optimizer, DLoss(), inputs=[input_A, fake_A], device=FLAGS.device)
 
     if FLAGS.resume:
         g.load(FLAGS.resume)

examples/handwritten_number_recognition/mnist.py

@@ -19,12 +19,12 @@ import argparse
 
 from paddle import fluid
 from paddle.fluid.optimizer import Momentum
-from hapi.datasets.mnist import MNIST as MnistDataset
+from paddle.incubate.hapi.datasets.mnist import MNIST as MnistDataset
 
-from hapi.model import Input, set_device
-from hapi.loss import CrossEntropy
-from hapi.metrics import Accuracy
-from hapi.vision.models import LeNet
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.loss import CrossEntropy
+from paddle.incubate.hapi.metrics import Accuracy
+from paddle.incubate.hapi.vision.models import LeNet
 
 
 def main():

examples/image_classification/imagenet_dataset.py

@@ -18,8 +18,8 @@ import math
 import random
 import numpy as np
 
-from hapi.datasets import DatasetFolder
-from hapi.vision.transforms import transforms
+from paddle.incubate.hapi.datasets import DatasetFolder
+from paddle.incubate.hapi.vision.transforms import transforms
 from paddle import fluid
 
 

examples/image_classification/main.py

@@ -27,11 +27,11 @@ import paddle.fluid as fluid
 from paddle.fluid.dygraph.parallel import ParallelEnv
 from paddle.io import BatchSampler, DataLoader
 
-from hapi.model import Input, set_device
-from hapi.loss import CrossEntropy
-from hapi.distributed import DistributedBatchSampler
-from hapi.metrics import Accuracy
-import hapi.vision.models as models
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.loss import CrossEntropy
+from paddle.incubate.hapi.distributed import DistributedBatchSampler
+from paddle.incubate.hapi.metrics import Accuracy
+import paddle.incubate.hapi.vision.models as models
 
 from imagenet_dataset import ImageNetDataset
 

examples/ocr/eval.py

@@ -19,8 +19,8 @@ import functools
 import paddle.fluid.profiler as profiler
 import paddle.fluid as fluid
 
-from hapi.model import Input, set_device
-from hapi.vision.transforms import BatchCompose
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.vision.transforms import BatchCompose
 
 from utility import add_arguments, print_arguments
 from utility import SeqAccuracy, LoggerCallBack, SeqBeamAccuracy

examples/ocr/predict.py

@@ -25,9 +25,9 @@ from PIL import Image
 import paddle.fluid.profiler as profiler
 import paddle.fluid as fluid
 
-from hapi.model import Input, set_device
-from hapi.datasets.folder import ImageFolder
-from hapi.vision.transforms import BatchCompose
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.datasets.folder import ImageFolder
+from paddle.incubate.hapi.vision.transforms import BatchCompose
 
 from utility import add_arguments, print_arguments
 from utility import postprocess, index2word

examples/ocr/seq2seq_attn.py

@@ -19,9 +19,9 @@ import paddle.fluid as fluid
 import paddle.fluid.layers as layers
 from paddle.fluid.layers import BeamSearchDecoder
 
-from hapi.text import RNNCell, RNN, DynamicDecode
-from hapi.model import Model
-from hapi.loss import Loss
+from paddle.incubate.hapi.text import RNNCell, RNN, DynamicDecode
+from paddle.incubate.hapi.model import Model
+from paddle.incubate.hapi.loss import Loss
 
 
 class ConvBNPool(fluid.dygraph.Layer):

examples/ocr/train.py

@@ -24,8 +24,8 @@ import functools
 import paddle.fluid.profiler as profiler
 import paddle.fluid as fluid
 
-from hapi.model import Input, set_device
-from hapi.vision.transforms import BatchCompose
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.vision.transforms import BatchCompose
 
 from utility import add_arguments, print_arguments
 from utility import SeqAccuracy, LoggerCallBack

examples/ocr/utility.py

@@ -21,8 +21,8 @@ import numpy as np
 import paddle.fluid as fluid
 import six
 
-from hapi.metrics import Metric
-from hapi.callbacks import ProgBarLogger
+from paddle.incubate.hapi.metrics import Metric
+from paddle.incubate.hapi.callbacks import ProgBarLogger
 
 
 def print_arguments(args):

examples/sentiment_classification/models.py

@@ -15,9 +15,9 @@ import paddle.fluid as fluid
 from paddle.fluid.dygraph.nn import Linear, Embedding
 from paddle.fluid.dygraph.base import to_variable
 import numpy as np
-from hapi.model import Model
-from hapi.text.text import GRUEncoderLayer as BiGRUEncoder
-from hapi.text.test import BOWEncoder, CNNEncoder, GRUEncoder
+from paddle.incubate.hapi.model import Model
+from paddle.incubate.hapi.text.text import GRUEncoderLayer as BiGRUEncoder
+from paddle.incubate.hapi.text.test import BOWEncoder, CNNEncoder, GRUEncoder
 
 
 class CNN(Model):
@@ -36,14 +36,18 @@ class CNN(Model):
             dict_size=self.dict_dim + 1,
             emb_dim=self.emb_dim,
             seq_len=self.seq_len,
-            filter_size= self.win_size,
-            num_filters= self.hid_dim,
-            hidden_dim= self.hid_dim,
+            filter_size=self.win_size,
+            num_filters=self.hid_dim,
+            hidden_dim=self.hid_dim,
             padding_idx=None,
             act='tanh')
-        self._fc1 = Linear(input_dim = self.hid_dim*self.seq_len, output_dim=self.fc_hid_dim, act="softmax")
-        self._fc_prediction = Linear(input_dim = self.fc_hid_dim,
-                                 output_dim = self.class_dim,
+        self._fc1 = Linear(
+            input_dim=self.hid_dim * self.seq_len,
+            output_dim=self.fc_hid_dim,
+            act="softmax")
+        self._fc_prediction = Linear(
+            input_dim=self.fc_hid_dim,
+            output_dim=self.class_dim,
             act="softmax")
 
     def forward(self, inputs):
@@ -69,10 +73,13 @@ class BOW(Model):
             padding_idx=None,
             bow_dim=self.hid_dim,
             seq_len=self.seq_len)
-        self._fc1 = Linear(input_dim = self.hid_dim, output_dim=self.hid_dim, act="tanh")
-        self._fc2 = Linear(input_dim = self.hid_dim, output_dim=self.fc_hid_dim, act="tanh")
-        self._fc_prediction = Linear(input_dim = self.fc_hid_dim,
-                                 output_dim = self.class_dim,
+        self._fc1 = Linear(
+            input_dim=self.hid_dim, output_dim=self.hid_dim, act="tanh")
+        self._fc2 = Linear(
+            input_dim=self.hid_dim, output_dim=self.fc_hid_dim, act="tanh")
+        self._fc_prediction = Linear(
+            input_dim=self.fc_hid_dim,
+            output_dim=self.class_dim,
             act="softmax")
 
     def forward(self, inputs):
@@ -94,8 +101,10 @@ class GRU(Model):
         self.class_dim = 2
         self.batch_size = batch_size
         self.seq_len = seq_len
-        self._fc1 = Linear(input_dim=self.hid_dim, output_dim=self.fc_hid_dim, act="tanh")
-        self._fc_prediction = Linear(input_dim=self.fc_hid_dim,
+        self._fc1 = Linear(
+            input_dim=self.hid_dim, output_dim=self.fc_hid_dim, act="tanh")
+        self._fc_prediction = Linear(
+            input_dim=self.fc_hid_dim,
             output_dim=self.class_dim,
             act="softmax")
         self._encoder = GRUEncoder(
@@ -130,9 +139,11 @@ class BiGRU(Model):
             is_sparse=False)
         h_0 = np.zeros((self.batch_size, self.hid_dim), dtype="float32")
         h_0 = to_variable(h_0)
-        self._fc1 = Linear(input_dim = self.hid_dim, output_dim=self.hid_dim*3)
-        self._fc2 = Linear(input_dim = self.hid_dim*2, output_dim=self.fc_hid_dim, act="tanh")
-        self._fc_prediction = Linear(input_dim=self.fc_hid_dim,
+        self._fc1 = Linear(input_dim=self.hid_dim, output_dim=self.hid_dim * 3)
+        self._fc2 = Linear(
+            input_dim=self.hid_dim * 2, output_dim=self.fc_hid_dim, act="tanh")
+        self._fc_prediction = Linear(
+            input_dim=self.fc_hid_dim,
             output_dim=self.class_dim,
             act="softmax")
         self._encoder = BiGRUEncoder(
@@ -144,7 +155,8 @@ class BiGRU(Model):
 
     def forward(self, inputs):
         emb = self.embedding(inputs)
-        emb = fluid.layers.reshape(emb, shape=[self.batch_size, -1, self.hid_dim])
+        emb = fluid.layers.reshape(
+            emb, shape=[self.batch_size, -1, self.hid_dim])
         fc_1 = self._fc1(emb)
         encoded_vector = self._encoder(fc_1)
         encoded_vector = fluid.layers.tanh(encoded_vector)

examples/sentiment_classification/sentiment_classifier.py

@@ -13,14 +13,13 @@
 # limitations under the License.
 """Sentiment Classification in Paddle Dygraph Mode. """
 
-
 from __future__ import print_function
 import numpy as np
 import paddle.fluid as fluid
-from hapi.model import set_device, Model, CrossEntropy, Input
-from hapi.configure import Config
-from hapi.text.senta import SentaProcessor
-from hapi.metrics import Accuracy
+from paddle.incubate.hapi.model import set_device, Model, CrossEntropy, Input
+from paddle.incubate.hapi.configure import Config
+from paddle.incubate.hapi.text.senta import SentaProcessor
+from paddle.incubate.hapi.metrics import Accuracy
 from models import CNN, BOW, GRU, BiGRU
 import json
 import os
@@ -32,12 +31,14 @@ args.Print()
 device = set_device("gpu" if args.use_cuda else "cpu")
 dev_count = fluid.core.get_cuda_device_count() if args.use_cuda else 1
 
+
 def main():
     if args.do_train:
         train()
     elif args.do_infer:
         infer()
 
+
 def train():
     fluid.enable_dygraph(device)
     processor = SentaProcessor(
@@ -66,19 +67,16 @@ def train():
         epoch=args.epoch,
         shuffle=False)
     if args.model_type == 'cnn_net':
-        model = CNN( args.vocab_size, args.batch_size,
-                     args.padding_size)
+        model = CNN(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'bow_net':
-        model = BOW( args.vocab_size, args.batch_size,
-                     args.padding_size)
+        model = BOW(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'gru_net':
-        model = GRU( args.vocab_size, args.batch_size,
-                     args.padding_size)
+        model = GRU(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'bigru_net':
-        model = BiGRU( args.vocab_size, args.batch_size,
-                       args.padding_size)
+        model = BiGRU(args.vocab_size, args.batch_size, args.padding_size)
 
-    optimizer = fluid.optimizer.Adagrad(learning_rate=args.lr, parameter_list=model.parameters()) 
+    optimizer = fluid.optimizer.Adagrad(
+        learning_rate=args.lr, parameter_list=model.parameters())
 
     inputs = [Input([None, None], 'int64', name='doc')]
     labels = [Input([None, 1], 'int64', name='label')]
@@ -86,7 +84,7 @@ def train():
     model.prepare(
         optimizer,
         CrossEntropy(),
-        Accuracy(topk=(1,)),
+        Accuracy(topk=(1, )),
         inputs,
         labels,
         device=device)
@@ -99,6 +97,7 @@ def train():
               eval_freq=args.eval_freq,
               save_freq=args.save_freq)
 
+
 def infer():
     fluid.enable_dygraph(device)
     processor = SentaProcessor(
@@ -114,26 +113,19 @@ def infer():
         epoch=1,
         shuffle=False)
     if args.model_type == 'cnn_net':
-        model_infer = CNN( args.vocab_size, args.batch_size,
-                           args.padding_size)
+        model_infer = CNN(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'bow_net':
-        model_infer = BOW( args.vocab_size, args.batch_size,
-                           args.padding_size)
+        model_infer = BOW(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'gru_net':
-        model_infer = GRU( args.vocab_size, args.batch_size,
-                           args.padding_size)
+        model_infer = GRU(args.vocab_size, args.batch_size, args.padding_size)
     elif args.model_type == 'bigru_net':
-        model_infer = BiGRU( args.vocab_size, args.batch_size,
+        model_infer = BiGRU(args.vocab_size, args.batch_size,
                             args.padding_size)
 
     print('Do inferring ...... ')
     inputs = [Input([None, None], 'int64', name='doc')]
     model_infer.prepare(
-        None,
-        CrossEntropy(),
-        Accuracy(topk=(1,)),
-        inputs,
-        device=device)
+        None, CrossEntropy(), Accuracy(topk=(1, )), inputs, device=device)
     model_infer.load(args.checkpoints, reset_optimizer=True)
     preds = model_infer.predict(test_data=infer_data_generator)
     preds = np.array(preds[0]).reshape((-1, 2))
@@ -143,9 +135,15 @@ def infer():
 
             for p in range(len(preds)):
                 label = np.argmax(preds[p])
-                result = json.dumps({'index': p, 'label': label, 'probs': preds[p].tolist()})
-                w.write(result+'\n')
-        print('Predictions saved at '+os.path.join(args.output_dir, 'predictions.json'))
+                result = json.dumps({
+                    'index': p,
+                    'label': label,
+                    'probs': preds[p].tolist()
+                })
+                w.write(result + '\n')
+        print('Predictions saved at ' + os.path.join(args.output_dir,
+                                                     'predictions.json'))
+
 
 if __name__ == '__main__':
     main()

examples/seq2seq/predict.py

@@ -23,7 +23,7 @@ import paddle.fluid as fluid
 from paddle.fluid.layers.utils import flatten
 from paddle.fluid.io import DataLoader
 
-from hapi.model import Input, set_device
+from paddle.incubate.hapi.model import Input, set_device
 from args import parse_args
 from seq2seq_base import BaseInferModel
 from seq2seq_attn import AttentionInferModel

examples/seq2seq/seq2seq_attn.py

@@ -19,8 +19,8 @@ from paddle.fluid.initializer import UniformInitializer
 from paddle.fluid.dygraph import Embedding, Linear, Layer
 from paddle.fluid.layers import BeamSearchDecoder
 
-from hapi.model import Model, Loss
-from hapi.text import DynamicDecode, RNN, BasicLSTMCell, RNNCell
+from paddle.incubate.hapi.model import Model, Loss
+from paddle.incubate.hapi.text import DynamicDecode, RNN, BasicLSTMCell, RNNCell
 
 from seq2seq_base import Encoder
 

examples/seq2seq/seq2seq_base.py

@@ -19,8 +19,8 @@ from paddle.fluid.initializer import UniformInitializer
 from paddle.fluid.dygraph import Embedding, Linear, Layer
 from paddle.fluid.layers import BeamSearchDecoder
 
-from hapi.model import Model, Loss
-from hapi.text import DynamicDecode, RNN, BasicLSTMCell, RNNCell
+from paddle.incubate.hapi.model import Model, Loss
+from paddle.incubate.hapi.text import DynamicDecode, RNN, BasicLSTMCell, RNNCell
 
 
 class CrossEntropyCriterion(Loss):

examples/seq2seq/train.py

@@ -21,7 +21,7 @@ import numpy as np
 import paddle.fluid as fluid
 from paddle.fluid.io import DataLoader
 
-from hapi.model import Input, set_device
+from paddle.incubate.hapi.model import Input, set_device
 from args import parse_args
 from seq2seq_base import BaseModel, CrossEntropyCriterion
 from seq2seq_attn import AttentionModel

examples/seq2seq/utility.py

@@ -16,8 +16,8 @@ import math
 
 import paddle.fluid as fluid
 
-from hapi.metrics import Metric
-from hapi.callbacks import ProgBarLogger
+from paddle.incubate.hapi.metrics import Metric
+from paddle.incubate.hapi.callbacks import ProgBarLogger
 
 
 class TrainCallback(ProgBarLogger):

examples/sequence_tagging/eval.py

@@ -28,11 +28,11 @@ import numpy as np
 work_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 sys.path.append(os.path.join(work_dir, "../"))
 
-from hapi.model import set_device, Input
-from hapi.text.sequence_tagging import SeqTagging, ChunkEval, LacLoss
-from hapi.text.sequence_tagging import LacDataset, LacDataLoader
-from hapi.text.sequence_tagging import check_gpu, check_version
-from hapi.text.sequence_tagging import PDConfig
+from paddle.incubate.hapi.model import set_device, Input
+from paddle.incubate.hapi.text.sequence_tagging import SeqTagging, ChunkEval, LacLoss
+from paddle.incubate.hapi.text.sequence_tagging import LacDataset, LacDataLoader
+from paddle.incubate.hapi.text.sequence_tagging import check_gpu, check_version
+from paddle.incubate.hapi.text.sequence_tagging import PDConfig
 
 import paddle.fluid as fluid
 from paddle.fluid.layers.utils import flatten
@@ -65,7 +65,8 @@ def main(args):
         device=place)
     model.load(args.init_from_checkpoint, skip_mismatch=True)
 
-    eval_result = model.evaluate(eval_dataset.dataloader, batch_size=args.batch_size)
+    eval_result = model.evaluate(
+        eval_dataset.dataloader, batch_size=args.batch_size)
     print("precison: %.5f" % (eval_result["precision"][0]))
     print("recall: %.5f" % (eval_result["recall"][0]))
     print("F1: %.5f" % (eval_result["F1"][0]))

examples/sequence_tagging/predict.py

@@ -29,11 +29,11 @@ import numpy as np
 work_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 sys.path.append(os.path.join(work_dir, "../"))
 
-from hapi.text.sequence_tagging import SeqTagging
-from hapi.model import Input, set_device
-from hapi.text.sequence_tagging import LacDataset, LacDataLoader
-from hapi.text.sequence_tagging import check_gpu, check_version
-from hapi.text.sequence_tagging import PDConfig
+from paddle.incubate.hapi.text.sequence_tagging import SeqTagging
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.text.sequence_tagging import LacDataset, LacDataLoader
+from paddle.incubate.hapi.text.sequence_tagging import check_gpu, check_version
+from paddle.incubate.hapi.text.sequence_tagging import PDConfig
 
 import paddle.fluid as fluid
 from paddle.fluid.layers.utils import flatten

examples/sequence_tagging/train.py

@@ -28,11 +28,11 @@ import numpy as np
 work_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 sys.path.append(os.path.join(work_dir, "../"))
 
-from hapi.model import Input, set_device
-from hapi.text.sequence_tagging import SeqTagging, LacLoss, ChunkEval
-from hapi.text.sequence_tagging import LacDataset, LacDataLoader
-from hapi.text.sequence_tagging import check_gpu, check_version
-from hapi.text.sequence_tagging import PDConfig
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.text.sequence_tagging import SeqTagging, LacLoss, ChunkEval
+from paddle.incubate.hapi.text.sequence_tagging import LacDataset, LacDataLoader
+from paddle.incubate.hapi.text.sequence_tagging import check_gpu, check_version
+from paddle.incubate.hapi.text.sequence_tagging import PDConfig
 
 import paddle.fluid as fluid
 from paddle.fluid.optimizer import AdamOptimizer

examples/style-transfer/README.md

@@ -32,10 +32,10 @@ gram_matrix = fluid.layers.matmul(tensor, fluid.layers.transpose(tensor, [1, 0])
 import numpy as np
 import matplotlib.pyplot as plt
 
-from hapi.model import Model, Loss
+from paddle.incubate.hapi.model import Model, Loss
 
-from hapi.vision.models import vgg16
-from hapi.vision.transforms import transforms
+from paddle.incubate.hapi.vision.models import vgg16
+from paddle.incubate.hapi.vision.transforms import transforms
 from paddle import fluid
 from paddle.fluid.io import Dataset
 

examples/style-transfer/style_transfer.py

@@ -3,10 +3,10 @@ import argparse
 import numpy as np
 import matplotlib.pyplot as plt
 
-from hapi.model import Model, Loss
+from paddle.incubate.hapi.model import Model, Loss
 
-from hapi.vision.models import vgg16
-from hapi.vision.transforms import transforms
+from paddle.incubate.hapi.vision.models import vgg16
+from paddle.incubate.hapi.vision.transforms import transforms
 from paddle import fluid
 from paddle.fluid.io import Dataset
 

examples/transformer/predict.py

@@ -25,7 +25,7 @@ from paddle.fluid.layers.utils import flatten
 from utils.configure import PDConfig
 from utils.check import check_gpu, check_version
 
-from hapi.model import Input, set_device
+from paddle.incubate.hapi.model import Input, set_device
 from reader import prepare_infer_input, Seq2SeqDataset, Seq2SeqBatchSampler
 from transformer import InferTransformer
 

examples/transformer/train.py

@@ -23,8 +23,8 @@ from paddle.io import DataLoader
 from utils.configure import PDConfig
 from utils.check import check_gpu, check_version
 
-from hapi.model import Input, set_device
-from hapi.callbacks import ProgBarLogger
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.callbacks import ProgBarLogger
 from reader import create_data_loader
 from transformer import Transformer, CrossEntropyCriterion
 

examples/transformer/transformer.py

@@ -20,8 +20,8 @@ import paddle.fluid as fluid
 import paddle.fluid.layers as layers
 from paddle.fluid.dygraph import Embedding, LayerNorm, Linear, Layer, to_variable
 from paddle.fluid.dygraph.learning_rate_scheduler import LearningRateDecay
-from hapi.model import Model, CrossEntropy, Loss
-from hapi.text import TransformerBeamSearchDecoder, DynamicDecode
+from paddle.incubate.hapi.model import Model, CrossEntropy, Loss
+from paddle.incubate.hapi.text import TransformerBeamSearchDecoder, DynamicDecode
 
 
 def position_encoding_init(n_position, d_pos_vec):

examples/tsm/infer.py

@@ -19,8 +19,8 @@ import os
 import argparse
 import numpy as np
 
-from hapi.model import Input, set_device
-from hapi.vision.transforms import Compose
+from paddle.incubate.hapi.model import Input, set_device
+from paddle.incubate.hapi.vision.transforms import Compose
 
 from check import check_gpu, check_version
 from modeling import tsm_resnet50
@@ -36,9 +36,7 @@ def main():
     device = set_device(FLAGS.device)
     fluid.enable_dygraph(device) if FLAGS.dynamic else None
 
-    transform = Compose([GroupScale(),
-                         GroupCenterCrop(),
-                         NormalizeImage()])
+    transform = Compose([GroupScale(), GroupCenterCrop(), NormalizeImage()])
     dataset = KineticsDataset(
         pickle_file=FLAGS.infer_file,
         label_list=FLAGS.label_list,
@@ -46,8 +44,8 @@ def main():
         transform=transform)
     labels = dataset.label_list
 
-    model = tsm_resnet50(num_classes=len(labels),
-                         pretrained=FLAGS.weights is None)
+    model = tsm_resnet50(
+        num_classes=len(labels), pretrained=FLAGS.weights is None)
 
     inputs = [Input([None, 8, 3, 224, 224], 'float32', name='image')]
 
@@ -66,19 +64,23 @@ def main():
 if __name__ == '__main__':
     parser = argparse.ArgumentParser("CNN training on TSM")
     parser.add_argument(
-        "--data", type=str, default='dataset/kinetics',
+        "--data",
+        type=str,
+        default='dataset/kinetics',
         help="path to dataset root directory")
     parser.add_argument(
-        "--device", type=str, default='gpu',
-        help="device to use, gpu or cpu")
+        "--device", type=str, default='gpu', help="device to use, gpu or cpu")
     parser.add_argument(
-        "-d", "--dynamic", action='store_true',
-        help="enable dygraph mode")
+        "-d", "--dynamic", action='store_true', help="enable dygraph mode")
     parser.add_argument(
-        "--label_list", type=str, default=None,
+        "--label_list",
+        type=str,
+        default=None,
         help="path to category index label list file")
     parser.add_argument(
-        "--infer_file", type=str, default=None,
+        "--infer_file",
+        type=str,
+        default=None,
         help="path to pickle file for inference")
     parser.add_argument(
         "-w",

examples/tsm/main.py

@@ -22,10 +22,10 @@ import numpy as np
 from paddle import fluid
 from paddle.fluid.dygraph.parallel import ParallelEnv
 
-from hapi.model import Model, Input, set_device
-from hapi.loss import CrossEntropy
-from hapi.metrics import Accuracy
-from hapi.vision.transforms import Compose
+from paddle.incubate.hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.loss import CrossEntropy
+from paddle.incubate.hapi.metrics import Accuracy
+from paddle.incubate.hapi.vision.transforms import Compose
 
 from modeling import tsm_resnet50
 from check import check_gpu, check_version

examples/tsm/modeling.py

@@ -17,8 +17,8 @@ import paddle.fluid as fluid
 from paddle.fluid.layer_helper import LayerHelper
 from paddle.fluid.dygraph.nn import Conv2D, Pool2D, BatchNorm, Linear
 
-from hapi.model import Model
-from hapi.download import get_weights_path_from_url
+from paddle.incubate.hapi.model import Model
+from paddle.incubate.hapi.download import get_weights_path_from_url
 
 __all__ = ["TSM_ResNet", "tsm_resnet50"]
 

examples/yolov3/infer.py

@@ -24,7 +24,7 @@ from paddle import fluid
 from paddle.fluid.optimizer import Momentum
 from paddle.io import DataLoader
 
-from hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.model import Model, Input, set_device
 
 from modeling import yolov3_darknet53, YoloLoss
 from transforms import *
@@ -65,13 +65,17 @@ def main():
     device = set_device(FLAGS.device)
     fluid.enable_dygraph(device) if FLAGS.dynamic else None
 
-    inputs = [Input([None, 1], 'int64', name='img_id'),
-              Input([None, 2], 'int32', name='img_shape'),
-              Input([None, 3, None, None], 'float32', name='image')]
+    inputs = [
+        Input(
+            [None, 1], 'int64', name='img_id'), Input(
+                [None, 2], 'int32', name='img_shape'), Input(
+                    [None, 3, None, None], 'float32', name='image')
+    ]
 
     cat2name = load_labels(FLAGS.label_list, with_background=False)
 
-    model = yolov3_darknet53(num_classes=len(cat2name),
+    model = yolov3_darknet53(
+        num_classes=len(cat2name),
         model_mode='test',
         pretrained=FLAGS.weights is None)
 
@@ -106,19 +110,33 @@ if __name__ == '__main__':
     parser.add_argument(
         "-d", "--dynamic", action='store_true', help="enable dygraph mode")
     parser.add_argument(
-        "--label_list", type=str, default=None,
+        "--label_list",
+        type=str,
+        default=None,
         help="path to category label list file")
     parser.add_argument(
-        "-t", "--draw_threshold", type=float, default=0.5,
+        "-t",
+        "--draw_threshold",
+        type=float,
+        default=0.5,
         help="threshold to reserve the result for visualization")
     parser.add_argument(
-        "-i", "--infer_image", type=str, default=None,
+        "-i",
+        "--infer_image",
+        type=str,
+        default=None,
         help="image path for inference")
     parser.add_argument(
-        "-o", "--output_dir", type=str, default='output',
+        "-o",
+        "--output_dir",
+        type=str,
+        default='output',
         help="directory to save inference result if --visualize is set")
     parser.add_argument(
-        "-w", "--weights", default=None, type=str,
+        "-w",
+        "--weights",
+        default=None,
+        type=str,
         help="path to weights for inference")
     FLAGS = parser.parse_args()
     print_arguments(FLAGS)

examples/yolov3/main.py

@@ -25,9 +25,9 @@ from paddle import fluid
 from paddle.fluid.optimizer import Momentum
 from paddle.io import DataLoader
 
-from hapi.model import Model, Input, set_device
-from hapi.distributed import DistributedBatchSampler
-from hapi.vision.transforms import Compose, BatchCompose
+from paddle.incubate.hapi.model import Model, Input, set_device
+from paddle.incubate.hapi.distributed import DistributedBatchSampler
+from paddle.incubate.hapi.vision.transforms import Compose, BatchCompose
 
 from modeling import yolov3_darknet53, YoloLoss
 from coco import COCODataset

examples/yolov3/modeling.py

@@ -23,7 +23,7 @@ from paddle.fluid.regularizer import L2Decay
 from hapi.model import Model
 from hapi.loss import Loss
 from hapi.download import get_weights_path_from_url
-from hapi.vision.models import darknet53
+from darknet import darknet53
 
 __all__ = ['YoloLoss', 'YOLOv3', 'yolov3_darknet53']