classification2 using roberta

freenlpc/freenlpc.py

@@ -10,6 +10,7 @@ def __init__(self, api_keys: list, gpu: bool=False, lang: str="en") -> None:
         self.__lang = lang
         self.__task_model = {
             "classification": "bart-large-mnli-yahoo-answers",
+            "classification2": "xlm-roberta-large-xnli",
             "dialog_sum": "bart-large-samsum",
             "headline_gen": "t5-base-en-generate-headline",
             "entities_extraction": "en_core_web_lg",
@@ -66,6 +67,7 @@ def __init_api(self):
         api_key = self.__api_keys[self.activated_apikey]
 
         self.__models = {"classification": nlpcloud.Client(self.which_model("classification"), api_key, gpu=self.__gpu, lang=self.__lang),
+                         "classification2": nlpcloud.Client(self.which_model("classification2"), api_key, gpu=self.__gpu, lang=self.__lang),
                          "dialog_sum": nlpcloud.Client(self.which_model("dialog_sum"), api_key, gpu=self.__gpu, lang=self.__lang),
                          "headline_gen":nlpcloud.Client(self.which_model("headline_gen"), api_key, gpu=self.__gpu, lang=self.__lang),
                          "entities_extraction": nlpcloud.Client(self.which_model("entities_extraction"), api_key, gpu=self.__gpu, lang=self.__lang),
@@ -108,6 +110,35 @@ def classification(self, text: str, lables: list, multiclass: bool =True):
                 return {'scored_labels': ordered}
             except requests.exceptions.HTTPError:
                 self.__init_api()
+
+    def classification2(self, text: str):
+        """perform classification on a piece of text.
+
+        Args:
+            text (str): The block of text you want to analyze. 2,500 tokens maximum.
+            lables (list): A list of labels you want to use to classify your text. 25 labels maximum.
+            multiclass (bool, optional): Whether multiple labels should be applied to your text,
+                meaning that the model will calculate an independent score for each label. Defaults to True.
+
+
+        Returns:
+            dict: scored labels.
+        """
+        while True:
+            try:
+                sleep(1)
+                response = self.__models[self.classification2.__name__].classification(text)
+                result = []
+                for i in range(len(response['labels'])):
+                    info = {}
+                    info['label'] = response['labels'][i]
+                    info['score'] = response['scores'][i]
+                    result.append(info)
+
+                ordered = sorted(result, key=itemgetter('score'), reverse=True)
+                return {'scored_labels': ordered}
+            except requests.exceptions.HTTPError:
+                self.__init_api()
 
 
     def dialog_sum(self, dialog: str):

setup.py

@@ -1,6 +1,6 @@
 from setuptools import setup, find_packages
 
-VERSION = '0.1.3'
+VERSION = '0.1.4'
 DESCRIPTION = 'A wrapper for nlpcloud free-tier services with no requests per minute limits.'
 LONG_DESCRIPTION = """# freenlpc
 a wrapper for nlpcloud free-tier.