python - deeppavlov意图dstc2分类输出不清楚(python)

Question

我正在关注这个文档: https://github.com/deepmipt/DeepPavlov/blob/master/docs/components/classifiers.rst#id53

我的代码如下:

import os
from deeppavlov import build_model, configs

os.environ["KERAS_BACKEND"] = "tensorflow"

CONFIG_PATH =  configs.classifiers.intents_dstc2_big
model = build_model(CONFIG_PATH, download=True)
print(model(["Hello"]))

我期待这样的输出:

"goals": {"pricerange": "cheap"},
"db_result": null,
"dialog-acts": [{"slots": [["pricerange", "cheap"]], "act": "inform"}]}

但是，我只得到一个像这样的数字数组:

[[0.004440320190042257, 0.0035526982974261045, 0.003814868861809373, 0.004386670421808958, 0.0026496422942727804, 0.004122086800634861, 0.004859328735619783, 0.005762884858995676, 0.006169301923364401, 0.9743947386741638, 0.005218957085162401, 0.004720163065940142, 0.006856555584818125, 0.0047727120108902454, 0.008368589915335178, 0.011183635331690311, 0.007578883320093155, 0.005414197687059641, 0.008248056285083294, 0.005105976946651936, 0.005934832151979208, 0.005890967790037394, 0.005130860488861799, 0.005532102193683386, 0.005490032024681568, 0.0046647703275084496, 0.004590084310621023, 0.004707065410912037]]

我应该如何正确显示或使用输出？

Answer 1

intents_dstc2_big模型不会为您提供正确的 DSTC2 输出，而是根据 act 识别话语的意图。和slot来自原始数据集的值。

例如，

"goals": {"food": "dontcare", "pricerange": "cheap", "area": "south"},
"db_result": null,
"dialog-acts": [{"slots": [], "act": "thankyou"}, {"slots": [], "act": "bye"}]}

此消息包含两个意图 (thankyou, bye)

为了获得意图方面的输出，您应该稍微更改配置。

import os
from deeppavlov import build_model, configs, train_model
from deeppavlov.core.common.file import read_json

os.environ["KERAS_BACKEND"] = "tensorflow"

model_config = read_json(configs.classifiers.intents_dstc2_big)
model_config['chainer']['out'] =  ['y_pred_labels']

model = build_model(model_config, download=True)
print(model(["thank you good bye"]))

您可以找到更多输出选项 in the configuration file .

请告诉我这是否足够有帮助。