def text_processing():
title = request.form.get('d_title', None)
body = request.form.get('d_body', None)
keywords = None
categories = None
datasets = None
dataset_ids = dict([(dataset['name'], index) for index, dataset in enumerate(iotools.load_datasets())])
if title or body:
keywords = generate_dataset_keywords_dict({"name": title, "long_desc": body, "short_desc": ""})
keywords = sorted(keywords['all'])
similarity_dict = get_dataset_compatibility(keywords)
categories = []
for key, val in sorted(similarity_dict.items(), key=lambda x: x[1], reverse=True):
if val > 0:
row = {
"name": key,
"similarity": "%4.1f%%" % (val * 100,)
}
categories.append(row)
related_datasets = get_related_datasets(keywords)
datasets = []
for name, val in sorted(related_datasets[:20], key=lambda x: x[1], reverse=True):
if val == 0:
continue
row = iotools.load_dataset(name)
row['similarity'] = "%4.1f%%" % (val * 100,)
datasets.append(row)
return render_template("text_processing.html", d_title=title, d_body=body, keywords=keywords,
categories=categories, related_datasets=datasets, dataset_ids=dataset_ids)
def get_related_datasets_html():
keywords = request.form['keywords'].split('\n')
new_style = 'old_style' not in request.form
related_datasets = get_related_datasets(keywords, new_style)
datasets = []
for name, val in related_datasets[:20]:
row = iotools.load_dataset(name)
row['similarity'] = "%3.0f%%" % (val * 100,)
datasets.append(row)
return render_template('related-datasets.html', datasets=datasets)
def get_all_conflicts():
datasets = {}
for name, conflictList in all_conflicts().items():
row = iotools.load_dataset(name)
row['conflicts'] = conflictList
category = row['category']
if category not in datasets:
datasets[category] = []
datasets[category].append(row)
return render_template('all-datasets.html', dataset_dict=datasets)