reorg demo

demo/ernie-classification/ernie_tiny_demo.py

-import paddle.fluid as fluid
-import paddlehub as hub
-
-# Step1
-module = hub.Module(name="ernie")
-inputs, outputs, program = module.context(trainable=True, max_seq_len=128)
-
-# Step2
-dataset = hub.dataset.ChnSentiCorp()
-reader = hub.reader.ClassifyReader(
-    dataset=dataset, vocab_path=module.get_vocab_path(), max_seq_len=128)
-
-# Step3
-with fluid.program_guard(program):
-    label = fluid.layers.data(name="label", shape=[1], dtype='int64')
-
-    pooled_output = outputs["pooled_output"]
-
-    cls_task = hub.create_text_classification_task(
-        feature=pooled_output, label=label, num_classes=dataset.num_labels)
-
-# Step4
-strategy = hub.AdamWeightDecayStrategy(learning_rate=5e-5, weight_decay=0.01)
-
-config = hub.RunConfig(
-    use_cuda=True, num_epoch=3, batch_size=32, strategy=strategy)
-
-feed_list = [
-    inputs["input_ids"].name, inputs["position_ids"].name,
-    inputs["segment_ids"].name, inputs["input_mask"].name, label.name
-]
-
-hub.finetune_and_eval(
-    task=cls_task, data_reader=reader, feed_list=feed_list, config=config)

demo/ernie-classification/question_answering.py

-#   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""Finetuning on classification task """
-
-import argparse
-
-import paddle.fluid as fluid
-import paddlehub as hub
-
-# yapf: disable
-parser = argparse.ArgumentParser(__doc__)
-parser.add_argument("--num_epoch", type=int, default=3, help="Number of epoches for fine-tuning.")
-parser.add_argument("--learning_rate", type=float, default=5e-5, help="Learning rate used to train with warmup.")
-parser.add_argument("--weight_decay", type=float, default=0.01, help="Weight decay rate for L2 regularizer.")
-parser.add_argument("--data_dir", type=str, default=None, help="Path to training data.")
-parser.add_argument("--checkpoint_dir", type=str, default=None, help="Directory to model checkpoint")
-parser.add_argument("--max_seq_len", type=int, default=512, help="Number of words of the longest seqence.")
-parser.add_argument("--batch_size", type=int, default=32, help="Total examples' number in batch for training.")
-args = parser.parse_args()
-# yapf: enable.
-
-if __name__ == '__main__':
-    # Step1: load Paddlehub ERNIE pretrained model
-    module = hub.Module(name="ernie")
-    inputs, outputs, program = module.context(
-        trainable=True, max_seq_len=args.max_seq_len)
-
-    # Step2: Download dataset and use ClassifyReader to read dataset
-    dataset = hub.dataset.NLPCC_DBQA()
-    reader = hub.reader.ClassifyReader(
-        dataset=dataset,
-        vocab_path=module.get_vocab_path(),
-        max_seq_len=args.max_seq_len)
-
-    # Step3: construct transfer learning network
-    with fluid.program_guard(program):
-        label = fluid.layers.data(name="label", shape=[1], dtype='int64')
-
-        # Use "pooled_output" for classification tasks on an entire sentence.
-        # Use "sequence_output" for token-level output.
-        pooled_output = outputs["pooled_output"]
-
-        # Setup feed list for data feeder
-        # Must feed all the tensor of ERNIE's module need
-        feed_list = [
-            inputs["input_ids"].name, inputs["position_ids"].name,
-            inputs["segment_ids"].name, inputs["input_mask"].name, label.name
-        ]
-        # Define a classfication finetune task by PaddleHub's API
-        cls_task = hub.create_text_classification_task(
-            pooled_output, label, num_classes=dataset.num_labels)
-
-        # Step4: Select finetune strategy, setup config and finetune
-        strategy = hub.AdamWeightDecayStrategy(
-            weight_decay=args.weight_decay,
-            learning_rate=args.learning_rate,
-            warmup_strategy="linear_warmup_decay",
-        )
-
-        # Setup runing config for PaddleHub Finetune API
-        config = hub.RunConfig(
-            use_cuda=True,
-            num_epoch=args.num_epoch,
-            batch_size=args.batch_size,
-            checkpoint_dir=args.checkpoint_dir,
-            strategy=strategy)
-
-        # Finetune and evaluate by PaddleHub's API
-        # will finish training, evaluation, testing, save model automatically
-        hub.finetune_and_eval(
-            task=cls_task,
-            data_reader=reader,
-            feed_list=feed_list,
-            config=config)

demo/ernie-classification/question_matching.py

-#   Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""Finetuning on classification task """
-
-import argparse
-
-import paddle.fluid as fluid
-import paddlehub as hub
-
-# yapf: disable
-parser = argparse.ArgumentParser(__doc__)
-parser.add_argument("--num_epoch", type=int, default=3, help="Number of epoches for fine-tuning.")
-parser.add_argument("--learning_rate", type=float, default=5e-5, help="Learning rate used to train with warmup.")
-parser.add_argument("--weight_decay", type=float, default=0.01, help="Weight decay rate for L2 regularizer.")
-parser.add_argument("--data_dir", type=str, default=None, help="Path to training data.")
-parser.add_argument("--checkpoint_dir", type=str, default=None, help="Directory to model checkpoint")
-parser.add_argument("--max_seq_len", type=int, default=512, help="Number of words of the longest seqence.")
-parser.add_argument("--batch_size", type=int, default=32, help="Total examples' number in batch for training.")
-args = parser.parse_args()
-# yapf: enable.
-
-if __name__ == '__main__':
-    # Step1: load Paddlehub ERNIE pretrained model
-    module = hub.Module(name="ernie")
-    inputs, outputs, program = module.context(
-        trainable=True, max_seq_len=args.max_seq_len)
-
-    # Step2: Download dataset and use ClassifyReader to read dataset
-    dataset = hub.dataset.LCQMC()
-    reader = hub.reader.ClassifyReader(
-        dataset=dataset,
-        vocab_path=module.get_vocab_path(),
-        max_seq_len=args.max_seq_len)
-
-    # Step3: construct transfer learning network
-    with fluid.program_guard(program):
-        label = fluid.layers.data(name="label", shape=[1], dtype='int64')
-
-        # Use "pooled_output" for classification tasks on an entire sentence.
-        # Use "sequence_output" for token-level output.
-        pooled_output = outputs["pooled_output"]
-
-        # Setup feed list for data feeder
-        # Must feed all the tensor of ERNIE's module need
-        feed_list = [
-            inputs["input_ids"].name, inputs["position_ids"].name,
-            inputs["segment_ids"].name, inputs["input_mask"].name, label.name
-        ]
-        # Define a classfication finetune task by PaddleHub's API
-        cls_task = hub.create_text_classification_task(
-            pooled_output, label, num_classes=dataset.num_labels)
-
-        # Step4: Select finetune strategy, setup config and finetune
-        strategy = hub.AdamWeightDecayStrategy(
-            weight_decay=args.weight_decay,
-            learning_rate=args.learning_rate,
-            warmup_strategy="linear_warmup_decay",
-        )
-
-        # Setup runing config for PaddleHub Finetune API
-        config = hub.RunConfig(
-            use_cuda=True,
-            num_epoch=args.num_epoch,
-            batch_size=args.batch_size,
-            checkpoint_dir=args.checkpoint_dir,
-            strategy=strategy)
-
-        # Finetune and evaluate by PaddleHub's API
-        # will finish training, evaluation, testing, save model automatically
-        hub.finetune_and_eval(
-            task=cls_task,
-            data_reader=reader,
-            feed_list=feed_list,
-            config=config)

demo/ernie-classification/run_question_answering.sh

-export CUDA_VISIBLE_DEVICES=3
-
-CKPT_DIR="./ckpt_dbqa"
-python -u question_answering.py \
-                   --batch_size 8 \
-                   --weight_decay  0.01 \
-                   --checkpoint_dir $CKPT_DIR \
-                   --num_epoch 3 \
-                   --max_seq_len 512 \
-                   --learning_rate 2e-5

demo/ernie-classification/run_question_matching.sh

-export CUDA_VISIBLE_DEVICES=5
-
-CKPT_DIR="./ckpt_question_matching"
-python -u question_matching.py \
-                   --batch_size 32 \
-                   --weight_decay  0.0 \
-                   --checkpoint_dir $CKPT_DIR \
-                   --num_epoch 3 \
-                   --max_seq_len 128 \
-                   --learning_rate 2e-5

demo/ernie-classification/run_sentiment_cls.sh

-export CUDA_VISIBLE_DEVICES=5
-
-CKPT_DIR="./ckpt_sentiment_cls"
-python -u sentiment_cls.py \
-                   --batch_size 32 \
-                   --use_gpu=False \
-                   --weight_decay  0.01 \
-                   --checkpoint_dir $CKPT_DIR \
-                   --num_epoch 3 \
-                   --max_seq_len 128 \
-                   --learning_rate 5e-5

demo/ernie-seq-labeling/run_sequence_labeling.sh → demo/sequence-labeling/run_sequence_labeling.sh

-export CUDA_VISIBLE_DEVICES=6
+export CUDA_VISIBLE_DEVICES=0
 
 CKPT_DIR="./ckpt_sequence_labeling"
-
-python -u sequence_labeling.py \
+python -u sequence_label.py \
                    --batch_size 16 \
                    --weight_decay  0.01 \
                    --checkpoint_dir $CKPT_DIR \

demo/ernie-seq-labeling/sequence_labeling.py → demo/sequence-labeling/sequence_label.py

@@ -37,13 +37,12 @@ if __name__ == '__main__':
         trainable=True, max_seq_len=args.max_seq_len)
 
     # Step2: Download dataset and use SequenceLabelReader to read dataset
+    dataset = hub.dataset.MSRA_NER()
     reader = hub.reader.SequenceLabelReader(
-        dataset=hub.dataset.MSRA_NER(),
+        dataset=dataset,
         vocab_path=module.get_vocab_path(),
         max_seq_len=args.max_seq_len)
 
-    num_labels = len(reader.get_labels())
-
     # Step3: construct transfer learning network
     with fluid.program_guard(program):
         label = fluid.layers.data(
@@ -62,11 +61,11 @@ if __name__ == '__main__':
             seq_len
         ]
         # Define a sequence labeling finetune task by PaddleHub's API
-        seq_label_task = hub.create_seq_labeling_task(
+        seq_label_task = hub.create_seq_label_task(
             feature=sequence_output,
             labels=label,
             seq_len=seq_len,
-            num_classes=num_labels)
+            num_classes=dataset.num_labels)
 
         # Select a finetune strategy
         strategy = hub.AdamWeightDecayStrategy(

demo/ernie-classification/cls_predict.py → demo/text-classification/cls_predict.py

@@ -64,7 +64,7 @@ if __name__ == '__main__':
         ]
 
         # Define a classfication finetune task by PaddleHub's API
-        cls_task = hub.create_text_classification_task(
+        cls_task = hub.create_text_cls_task(
             feature=pooled_output, label=label, num_classes=dataset.num_labels)
 
         # classificatin probability tensor

demo/text-classification/run_classifier.sh

+export CUDA_VISIBLE_DEVICES=5
+
+# User can select senticorp, nlpcc_dbqa, lcqmc for different task
+DATASET="senticorp"
+CKPT_DIR="./ckpt_${DATASET}"
+# Recommending hyper parameters for difference task
+# ChnSentiCorp: batch_size=24, weight_decay=0.01, num_epoch=3, max_seq_len=128, lr=5e-5
+# NLPCC_DBQA: batch_size=8, weight_decay=0.01, num_epoch=3, max_seq_len=512, lr=2e-5
+# LCQMC: batch_size=32, weight_decay=0, num_epoch=3, max_seq_len=128, lr=2e-5
+
+python -u text_classifier.py \
+                   --batch_size=24 \
+                   --use_gpu=True \
+                   --dataset=${DATASET} \
+                   --checkpoint_dir=${CKPT_DIR} \
+                   --learning_rate=5e-5 \
+                   --weight_decay=0.01 \
+                   --max_seq_len=128
+                   --num_epoch=3 \

demo/ernie-classification/sentiment_cls.py → demo/text-classification/text_classifier.py

@@ -23,8 +23,10 @@ import paddlehub as hub
 parser = argparse.ArgumentParser(__doc__)
 parser.add_argument("--num_epoch", type=int, default=3, help="Number of epoches for fine-tuning.")
 parser.add_argument("--use_gpu", type=ast.literal_eval, default=False, help="Whether use GPU for finetuning, input should be True or False")
+parser.add_argument("--dataset", type=str, default="senticorp", help="Directory to model checkpoint")
 parser.add_argument("--learning_rate", type=float, default=5e-5, help="Learning rate used to train with warmup.")
 parser.add_argument("--weight_decay", type=float, default=0.01, help="Weight decay rate for L2 regularizer.")
+parser.add_argument("--warmup_proportion", type=float, default=0.0, help="Warmup proportion params for warmup strategy")
 parser.add_argument("--data_dir", type=str, default=None, help="Path to training data.")
 parser.add_argument("--checkpoint_dir", type=str, default=None, help="Directory to model checkpoint")
 parser.add_argument("--max_seq_len", type=int, default=512, help="Number of words of the longest seqence.")
@@ -40,7 +42,16 @@ if __name__ == '__main__':
         trainable=True, max_seq_len=args.max_seq_len)
 
     # Step2: Download dataset and use ClassifyReader to read dataset
-    dataset = hub.dataset.ChnSentiCorp()
+    dataset = None
+    if args.dataset.lower() == "senticorp":
+        dataset = hub.dataset.ChnSentiCorp()
+    elif args.dataset.lower() == "nlpcc_dbqa":
+        dataset = hub.dataset.NLPCC_DBQA()
+    elif args.dataset.lower() == "lcqmc":
+        dataset = hub.dataset.LCQMC()
+    else:
+        raise ValueError("%s dataset is not defined" % args.dataset)
+
     reader = hub.reader.ClassifyReader(
         dataset=dataset,
         vocab_path=module.get_vocab_path(),
@@ -72,7 +83,6 @@ if __name__ == '__main__':
         )
 
         # Setup runing config for PaddleHub Finetune API
-        print(args.use_gpu)
         config = hub.RunConfig(
             use_cuda=args.use_gpu,
             num_epoch=args.num_epoch,