servertester.py

import requests

speaker_ids = ["A", "B", "A", "B", "A", "B"]
utterances = ["I don't know, ", "Where did you go?", "What?", " Where did you go?", "I went to University.", "Uh-huh."]
utt_ids = ["1", "2", " 3", "4", "5", "6"]
emotions = ["neutral", "surprise", "surprise", "angry", "frustration", "neutral"]

text = "$$$$".join(speaker_ids) + "?????" + "$$$$".join(utterances) + "?????" + "$$$$".join(
    utt_ids) + "?????" + "$$$$".join(emotions)

link = "http://0.0.0.0:4004/"
link = "http://eb22add1.eu.ngrok.io/"
results = requests.post(link + 'predict_das', json={"text": text}).json()['result']

result_items = results.split('?????')
f_kappa_score_text = result_items[0]
k_alpha_score_text = result_items[1]
overall_data_assessment = result_items[2]

speaker_id, utt_id, utterance, emotion = [], [], [], []
eda1, eda2, eda3, eda4, eda5, EDA = [], [], [], [], [], []
all_match, con_match, match = [], [], []

swda_elmo_non_con_out, swda_elmo_non_con_out_confs = [], []
swda_elmo_mean_non_con_out, swda_elmo_mean_non_con_out_confs = [], []
swda_elmo_con_out, swda_elmo_con_out_confs = [], []
swda_elmo_mean_con_out, swda_elmo_mean_con_out_confs = [], []
swda_elmo_top_con_out, swda_elmo_top_con_out_confs = [], []

for item in result_items[3:]:
    elements = item.split('$$$$')
    speaker_id.append(elements[0]), utt_id.append(elements[1]), utterance.append(elements[2]), emotion.append(
        elements[3])
    swda_elmo_non_con_out.append(elements[4]), swda_elmo_non_con_out_confs.append(float(elements[5]))
    swda_elmo_mean_non_con_out.append(elements[6]), swda_elmo_mean_non_con_out_confs.append(float(elements[7]))
    swda_elmo_con_out.append(elements[8]), swda_elmo_con_out_confs.append(float(elements[9]))
    swda_elmo_mean_con_out.append(elements[10]), swda_elmo_mean_con_out_confs.append(float(elements[11]))
    swda_elmo_top_con_out.append(elements[12]), swda_elmo_top_con_out_confs.append(float(elements[13]))

print(f_kappa_score_text, k_alpha_score_text, overall_data_assessment)

print(speaker_id, utt_id, utterance, emotion, swda_elmo_non_con_out, swda_elmo_non_con_out_confs,
      swda_elmo_mean_non_con_out, swda_elmo_mean_non_con_out_confs,
      swda_elmo_con_out, swda_elmo_con_out_confs,
      swda_elmo_mean_con_out, swda_elmo_mean_con_out_confs,
      swda_elmo_top_con_out, swda_elmo_top_con_out_confs)