QA example refactoring. (#36)

* QA example refactoring.
- Notebook updated as to match new version.

* README.md updated.
- QA notebook badge added.
- Badges introduced.

* Dependency badge (source) updated.

* Updates from review.

* Updates from review and qa pipeline refactoring.

* Updates from review and qa pipeline refactoring.

* Notebook output cells cleared.

* Bug fix on QA Processor & requirements.txt added for QA example.

* Version update.

* Notebook and example project readmes updated.
- Notebook utils added.

* debug_scripts ignored.

* README updated.
- Notebook utils updated.

* Code formatting.

* README.md updated.

* Experiment config fetch added.

* Notebook updated.
- Prepare data is updated.

.gitignore

@@ -42,4 +42,7 @@ coverage.xml
 *dataset_infos.json
 
 # mypy
-.mypy_cache
+.mypy_cache
+
+# Misc
+debug_scripts/

README.md

@@ -1,4 +1,16 @@
-# Trapper (Transformers wRAPPER)
+<h1 align="center">Trapper (Transformers wRAPPER)</h1>
+
+<p align="center">
+<a href="https://pypi.org/project/trapper"><img src="https://img.shields.io/pypi/pyversions/trapper" alt="Python versions"></a>
+<a href="https://pypi.org/project/trapper"><img src="https://img.shields.io/pypi/v/trapper?color=blue" alt="PyPI version"></a>
+<a href="https://github.com/obss/trapper/releases/latest"><img alt="Latest Release" src="https://img.shields.io/github/release-date/obss/trapper"></a>
+<a href="https://colab.research.google.com/github/obss/trapper/blob/main/examples/question_answering/question_answering.ipynb"><img alt="Open in Colab" src="https://colab.research.google.com/assets/colab-badge.svg"></a>
+<br>
+<a href="https://github.com/obss/trapper/actions"><img alt="Build status" src="https://github.com/obss/trapper/actions/workflows/ci.yml/badge.svg"></a>
+<a href="https://libraries.io/pypi/trapper"><img alt="Dependencies" src="https://img.shields.io/librariesio/release/pypi/trapper"></a>
+<a href="https://github.com/psf/black"><img alt="Code style: black" src="https://img.shields.io/badge/code%20style-black-000000.svg"></a>
+<a href="https://github.com/obss/trapper/blob/main/LICENSE"><img alt="License: MIT" src="https://img.shields.io/github/license/obss/trapper"></a>
+</p>
 
 Trapper is an NLP library that aims to make it easier to train transformer based
 models on downstream tasks. It wraps the HuggingFace's `transformers` library to
@@ -300,15 +312,15 @@ thanks to configuration file based experiments.
 ### Training a POS Tagging Model on CONLL2003
 
 Since the transformers library lacks a direct support for POS tagging, we added an
-example project that trains a transformer model on `CONLL2003` POS tagging dataset
-and perform inference using it. You can find it in `examples/pos_tagging`. It is a
+[example project](./examples/pos_tagging) that trains a transformer model on `CONLL2003` POS tagging dataset
+and perform inference using it. It is a
 self-contained project including its own requirements file, therefore you can copy
 the folder into another directory to use as a template for your own project. Please
 follow its `README.md` to get started.
 
 ### Training a Question Answering Model on SQuAD Dataset
 
-You can use the notebook in `examples/question_answering/question_answering. ipynb`
+You can use the notebook in the [Example QA Project](./examples/question_answering) `examples/question_answering/question_answering.ipynb`
 to follow the steps while training a transformer model on SQuAD v1.
 
 ## Installation

examples/pos_tagging/README.md

@@ -4,8 +4,7 @@ This project show how to train a transformer model from on CONLL2003 dataset
 from `HuggingFace datasets`. You can explore the dataset
 from [its page](https://huggingface.co/datasets/conll2003). This project is intended
 to serve as a demo for using trapper as a library to train and evaluate a
-transformer model on a custom task/dataset as well as perform inference using it. We
-start by creating a fresh python environment and install the dependencies.
+transformer model on a custom task/dataset as well as perform inference using it. To see an example of supported task, see [Question answering example](../question_answering). We start by creating a fresh python environment and install the dependencies.
 
 ## Environment Creation and Dependency Installation
 

examples/pos_tagging/requirements.txt

@@ -1 +1 @@
-trapper==0.0.3
+trapper==0.0.4

examples/question_answering/README.md

@@ -0,0 +1,6 @@
+## Question Answering Demo
+
+<a href="https://colab.research.google.com/github/obss/jury/blob/main/examples/question_answering/question_answering.ipynb"><img alt="Open in Colab" src="https://colab.research.google.com/assets/colab-badge.svg"></a>
+
+
+This notebook serves as an example for demonstrating training and inference using `trapper`. Question-answering task is supported by `trapper` already, and thus in this notebook we only give a basic [configuration file](./experiment.jsonnet) and let the trapper take care of the rest. For implementation of a desired task using trapper, see [Pos tagging example](../pos_tagging). 

examples/question_answering/experiments/question-answering/experiment.jsonnet → examples/question_answering/experiment.jsonnet

@@ -30,10 +30,10 @@ local result_dir = std.extVar("OUTPUT_PATH");
         "type": "default",
         "output_dir": checkpoint_dir,
         "result_dir": result_dir,
-        "num_train_epochs": 2,
+        "num_train_epochs": 10,
         "per_device_train_batch_size": 2,
         "gradient_accumulation_steps": 12,
-        "per_device_eval_batch_size": 4,
+        "per_device_eval_batch_size": 2,
         "logging_dir": checkpoint_dir + "/logs",
         "no_cuda": false,
         "logging_steps": 500,