cap

interfaces/cap.py

@@ -104,7 +104,7 @@ def predict(text, model_id, tokenizer_id):
 
     gr.Info("Tokenizing")
     inputs = tokenizer(text,
-                       max_length=4,
+                       max_length=256,
                        truncation=True,
                        padding="do_not_pad",
                        return_tensors="pt").to(device)

utils.py

@@ -7,15 +7,22 @@ from interfaces.manifesto import languages as languages_manifesto
 from interfaces.manifesto import languages as languages_manifesto
 """
 
+from interfaces.cap import languages as languages_cap
+from interfaces.cap import domains as domains_cap
+
 from interfaces.cap import build_huggingface_path as hf_cap_path
 from interfaces.manifesto import build_huggingface_path as hf_manifesto_path
 from interfaces.sentiment import build_huggingface_path as hf_sentiment_path
 from interfaces.emotion import build_huggingface_path as hf_emotion_path
 
-
 HF_TOKEN = os.environ["hf_read"]
 
+# should be a temporary solution
 models = [hf_manifesto_path(""), hf_sentiment_path(""), hf_emotion_path("")]
+for languages in languages_cap:
+    for domains in domains_cap:
+        models.append(hf_cap_path(language, domain))
+
 tokenizers = ["xlm-roberta-large"]
 
 def download_hf_models():