def topic_word_counts(user_mc_key, topics_id, **kwargs):
# Return sampled word counts based on filters.
snapshots_id, timespans_id, foci_id, q = filters_from_args(request.args)
merged_args = {
'snapshots_id': snapshots_id,
'timespans_id': timespans_id,
'foci_id': foci_id,
'q': q,
'sample_size': WORD_COUNT_SAMPLE_SIZE,
'num_words': WORD_COUNT_UI_NUM_WORDS
}
merged_args.update(kwargs) # passed in args override anything pulled form the request.args
word_data = cached_topic_word_counts(user_mc_key, topics_id, **merged_args)
words = [w['term'] for w in word_data]
# and now add in word2vec model position data
if len(words) > 0:
google_word2vec_data = base_apicache.word2vec_google_2d(words)
for i in range(len(google_word2vec_data)):
word_data[i]['google_w2v_x'] = google_word2vec_data[i]['x']
word_data[i]['google_w2v_y'] = google_word2vec_data[i]['y']
topic_word2vec_data = _word2vec_topic_2d_results(topics_id, snapshots_id, words)
for i in range(len(topic_word2vec_data)):
word_data[i]['w2v_x'] = topic_word2vec_data[i]['x']
word_data[i]['w2v_y'] = topic_word2vec_data[i]['y']
return word_data
def query_wordcount(q,
fq,
ngram_size=1,
num_words=WORD_COUNT_UI_NUM_WORDS,
sample_size=WORD_COUNT_SAMPLE_SIZE):
word_data = base_apicache.word_count(q,
fq,
ngram_size=ngram_size,
num_words=num_words,
sample_size=sample_size)
# add in word2vec results
words = [w['term'] for w in word_data]
# and now add in word2vec model position data
if len(words) > 0:
google_word2vec_data = base_apicache.word2vec_google_2d(words)
for i in range(len(google_word2vec_data)):
word_data[i]['google_w2v_x'] = google_word2vec_data[i]['x']
word_data[i]['google_w2v_y'] = google_word2vec_data[i]['y']
return word_data
def _cached_word_count(user_mc_key,
q,
fq,
num_words,
sample_size=WORD_COUNT_SAMPLE_SIZE):
api_client = mc if user_mc_key is None else user_admin_mediacloud_client()
word_data = api_client.wordCount(q,
fq,
num_words=num_words,
sample_size=sample_size)
words = [w['term'] for w in word_data]
word2vec_data = base_apicache.word2vec_google_2d(words)
try:
for i in range(len(word2vec_data)):
word_data[i]['google_w2v_x'] = word2vec_data[i]['x']
word_data[i]['google_w2v_y'] = word2vec_data[i]['y']
except KeyError as e:
logger.warning("Didn't get valid data back from word2vec call")
logger.exception(e)
return word_data