showcase_app to spanish

showcase_app.py

@@ -15,14 +15,14 @@ from filters import (
     remove_urls,
 )
 
-st.sidebar.markdown("## Models loaded")
+st.sidebar.markdown("## Modelos cargados")
 
 
-st.title("Sentiment Analys for Spanish Tweets!")
+st.title("Clasificador de sentimientos para Tweets en Español")
 st.write(
-    "I use the Hugging Face Transformers library to clasify the sentiment \
-    of tweets passed as input as postive, neutral or negative. \
-    This app is built using [Streamlit](https://docs.streamlit.io/en/stable/getting_started.html)."
+    "Utilizo la biblioteca Huggin Face Transformers para clasificar el sentimiento \
+    de tweets pasados como entrada para positivos, neutrales o negativos . \
+    Esta aplicación está construida usando [Streamlit](https://docs.streamlit.io/en/stable/getting_started.html)."
 )
 
 models = [
@@ -34,12 +34,12 @@ models = [
     "francisco-perez-sorrosal/dccuchile-distilbert-base-spanish-uncased-finetuned-with-spanish-tweets-clf-cleaned-ds",
 ]
 
-load_all_models = st.checkbox("Load all models?")
+load_all_models = st.checkbox("¿Cargar todos los modelos?")
 
 if "pipelines" not in st.session_state:
     st.session_state.pipelines = []
     for model in models:
-        with st.spinner(f"Loading model {model}"):
+        with st.spinner(f"Cargando modelo {model}"):
             pipe = pipeline(
                 "text-classification",
                 model=AutoModelForSequenceClassification.from_pretrained(model),
@@ -47,8 +47,8 @@ if "pipelines" not in st.session_state:
                 return_all_scores=True,
             )
         st.sidebar.subheader(pipe.model.config.name_or_path)
-        st.sidebar.write(f"Tokenizer:\n{pipe.tokenizer}")
-        st.sidebar.write(f"Model:\n{pipe.model.config}")
+        st.sidebar.write(f"Tokenizador:\n{pipe.tokenizer}")
+        st.sidebar.write(f"Modelo:\n{pipe.model.config}")
         st.session_state.pipelines.append(pipe)
         if not load_all_models:
             break
@@ -63,62 +63,62 @@ def update_model(
     if not load_all_models:
         if local_model_id:
             model_id = local_model_id
-        st.text(f"Loading model {model_id}")
+        st.text(f"Cargando modelo {model_id}")
         pipe = pipeline(
-            "text-classification",
+            "clasificación-de-texto",
             model=AutoModelForSequenceClassification.from_pretrained(model_id),
             tokenizer=AutoTokenizer.from_pretrained(model_id),
             return_all_scores=True,
         )
         st.sidebar.subheader(pipe.model.config.name_or_path)
-        st.sidebar.write(f"Tokenizer:\n{pipe.tokenizer}")
-        st.sidebar.write(f"Model:\n{pipe.model.config}")
+        st.sidebar.write(f"Tokenizador:\n{pipe.tokenizer}")
+        st.sidebar.write(f"Modelo:\n{pipe.model.config}")
         st.session_state.pipelines.append(pipe)
     else:
         for model in models:
-            with st.spinner(f"Loading model {model}"):
+            with st.spinner(f"Cargando modelo {model}"):
                 pipe = pipeline(
-                    "text-classification",
+                    "clasificación-de-texto",
                     model=AutoModelForSequenceClassification.from_pretrained(model),
                     tokenizer=AutoTokenizer.from_pretrained(model),
                     return_all_scores=True,
                 )
                 st.sidebar.subheader(pipe.model.config.name_or_path)
-                st.sidebar.write(f"Tokenizer:\n{pipe.tokenizer}")
-                st.sidebar.write(f"Model:\n{pipe.model.config}")
+                st.sidebar.write(f"Tokenizador:\n{pipe.tokenizer}")
+                st.sidebar.write(f"Modelo:\n{pipe.model.config}")
             st.session_state.pipelines.append(pipe)
     st.session_state.last_updated = datetime.datetime.now().time()
 
 
-model_id = st.selectbox(f"Choose a model {load_all_models}", models)
-local_model_id = st.text_input(f"Choose a model from local disk")
+model_id = st.selectbox(f"Elige un modelo {load_all_models}", models)
+local_model_id = st.text_input(f"Elige un modelo del disco local")
 st.button(
-    "Load model/s",
+    "Cargando modelo/s",
     on_click=update_model,
     args=(
         local_model_id,
         model_id,
     ),
 )
-st.write("Last Updated = ", st.session_state.last_updated)
+st.write("Ultima actualización = ", st.session_state.last_updated)
 
 
-form = st.form(key="sentiment-form")
-tweet_text = form.text_area("Enter your tweet text")
-clean_input = form.checkbox("Clean input text?")
-ask_chatgpt = form.checkbox("Ask ChatGPT too?")
+form = st.form(key="sentimient")
+tweet_text = form.text_area("Introduce tu texto")
+clean_input = form.checkbox("¿Limpiar el texto?")
+ask_chatgpt = form.checkbox("¿Preguntarle a ChatGPT?")
 openai_key = form.text_input("OpenAI Key")
-submit = form.form_submit_button("Submit")
+submit = form.form_submit_button("Cargar")
 
 if submit:
     if clean_input:
         tweet_text = remove_twitter_handles({"text": tweet_text})["text"]
         tweet_text = remove_urls({"text": tweet_text})["text"]
         tweet_text = clean_html({"text": tweet_text})["text"]
-    st.write(f"Sending this tweet content to the model: {tweet_text}")
+    st.write(f"Enviando este texto al modelo: {tweet_text}")
 
     for classifier in st.session_state.pipelines:
-        st.subheader(f"Model\n{classifier.model.config.name_or_path}")
+        st.subheader(f"Modelo\n{classifier.model.config.name_or_path}")
         result = classifier(tweet_text)
         st.json(result, expanded=False)
         predictions = result[0]
@@ -130,11 +130,11 @@ if submit:
                 score = p["score"]
 
         if label == "P":
-            st.success(f"{label} sentiment (score: {score})")
+            st.success(f"{label} sentimiento (puntuación: {score})")
         elif label == "NEU":
-            st.warning(f"{label} sentiment (score: {score})")
+            st.warning(f"{label} sentimiento (puntuación: {score})")
         else:
-            st.error(f"{label} sentiment (score: {score})")
+            st.error(f"{label} sentimiento (puntuación: {score})")
 
     if ask_chatgpt:
         openai.api_key = openai_key