Comments (2)
Thanks for raising this point. It's definitely true that pre-processing is important. For example, it's crucial that the tokenization used during training is as close to the tokenization when evaluating "in the wild" as possible. If you have a different tokenizer that doesn't handle edge cases the same (e.g. splitting on apostrophes, etc), you will introduce systematic errors into the parser, so it's important to take that into account when running Parsey McParseface.
Joint segmentation and parsing models can be implemented in SyntaxNet and we would encourage anyone who is interested to try it out. In particular, using a beam model has the advantage that it can maintain and compare multiple segmentations at once. We can also look into releasing some more detailed benchmarks beyond the ones we have already.
Best,
David Weiss
from models.
Closing this issue for now. Feel free to reopen if you want to add more data points on the effect of pre-processing on syntaxnet models.
from models.
Related Issues (20)
- DCN V2 pretrained model on Criteo 1TB data
- 'pip3 install tf-models-nightly Collecting tf-models-nightly' HOT 4
- tf1 upgrade to tf2,tf.distribute.MirroredStrategy core dump HOT 3
- i cant run movinet streaming official model on vs code running windows, it keep throwing an error HOT 12
- Keras 3 compatibility? AttributeError: 'LossScaleOptimizer' object has no attribute 'get_scaled_loss' HOT 7
- tensorflow.python.framework.errors_impl.NotFoundError: inference.so not found. HOT 4
- TypeError: unhashable type: 'list'. When running "model_builder_tf2_test.py" HOT 8
- ValueError: Only fixed_shape_resizeris supported with tflite. Found keep_aspect_ratio_resize HOT 2
- object_detection : python setup.py egg_info did not run successfully installing the object detection API HOT 2
- Apache Beam Pipeline cannot maximize the number of workers for criteo_preprocess.py in Google Cloud
- SSD output structure
- ImportError: cannot import name 'eval_pb2' from 'object_detection.protos' HOT 6
- Unable to install tf-models-official HOT 5
- GradCAM for MoViNet HOT 2
- models/research/object_detection/exporter_main_v2.py give me error "ImportError: cannot import name 'builder'" HOT 1
- Trying to Load Keras Model Returns ListIndex Error HOT 4
- several errors (4) on movinet streaming_model_training_and_inference notebook when simply ran through on kaggle and colab HOT 7
- Issue with link to MaskConver implementation HOT 1
- How to add additional class (QR Codes) for coco dataset or any suggested dataset for the class I have mentioned HOT 4
- Keras 3 compatibility! HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from models.