Update app.py

app.py

@@ -1,16 +1,23 @@
 # import sentencepiece
 
 
-# import streamlit as st
-# from transformers import pipeline
+import streamlit as st
+from transformers import pipeline
+
+sentiment_analysis = pipeline("sentiment-analysis")
+translation = pipeline("translation_en_to_ar", model="anibahug/marian-finetuned-kde4-en-to-ar")
+
+text = st.text_input("Enter some text")
 
-# sentiment_analysis = pipeline("sentiment-analysis")
+if text:
+    result = sentiment_analysis(text)
+    st.json(result)
+
+if text:
+    result = translation(text)[0]
+    st.write(f"Translated text: {result['translation_text']}")
 
-# text = st.text_input("Enter some text")
 
-# if text:
-#     result = sentiment_analysis(text)
-#     st.json(result)
 
 
 
@@ -38,157 +45,4 @@
 #     confidence = results[0]['score']
     
 #     st.write(f"Sentiment: {sentiment}")
-#     st.write(f"Confidence: {confidence:.2f}")
-
-
-
-import torch
-from transformers import AutoModelForDiffusion, DiffusionConfig
-
-# Specify the model parameters
-model_config = DiffusionConfig(
-    image_size=64,
-    num_channels=3,
-    num_steps=1000,
-    diffusion_steps=100,
-    noise_schedule="linear",
-    learning_rate=0.0001,
-)
-
-# Load the model
-model = AutoModelForDiffusion.from_pretrained("runwayml/stable-diffusion-v1-5", config=model_config)
-
-# Generate an image
-image = model.generate(text="This is a test")
-
-# Save the image
-torch.save(image, "image.png")
-
-
-
-
-# # Import streamlit and huggingface libraries
-# import streamlit as st
-# from transformers import pipeline
-
-# # Create a text to image pipeline
-# text_to_image = pipeline("text-to-image", model="runwayml/stable-diffusion-v1-5")
-
-# # Create a text input widget
-# text = st.text_input("Enter some text")
-
-# # Generate an image and display the result
-# if text:
-#     result = text_to_image(text)[0]
-#     st.image(result["image"])
-
-
-
-
-# import streamlit as st
-# from transformers import pipeline
-
-
-
-
-# translation = pipeline("translation_en_to_ar", model="anibahug/marian-finetuned-kde4-en-to-ar")
-
-# if text:
-#     result = translation(text)[0]
-#     st.write(f"Translated text: {result['translation_text']}")
-
-
-
-
-
-
-# text_to_speech = pipeline("Text-to-Speech")
-
-# if text:
-#     result = text_to_speech(text)[0]
-#     st.audio(result["audio"])
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-# import streamlit as st
-# from transformers import pipeline, MarianTokenizer, MarianForCausalLM
-
-# # Set up Streamlit layout
-# st.title("Text-to-Speech App")
-# st.write("Enter text below:")
-
-# text_input = st.text_area("Input Text")
-
-# # Load the text-to-speech model
-# model_name = "facebook/wav2vec2-base-960h"  # Replace with your desired TTS model
-# tokenizer = MarianTokenizer.from_pretrained(model_name)
-# model = MarianForCausalLM.from_pretrained(model_name)
-
-# # Generate audio
-# if st.button("Generate Audio"):
-#     if text_input:
-#         input_ids = tokenizer.encode(text_input, return_tensors="pt")
-#         with st.spinner("Generating audio..."):
-#             output = model.generate(input_ids)
-#             st.audio(output[0].numpy(), format="audio/wav")
-
-# # Display credits and instructions
-# st.markdown("Powered by Hugging Face Spaces API and Streamlit.")
-
-
-
-
-
-
-
-
-
-
-
-
-# import streamlit as st
-# from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
-
-# # Load the text-to-speech model from Huggingface Spaces
-# model_id = "facebook/bart-base-tts"
-# model = AutoModelForSeq2SeqLM.from_pretrained(model_id)
-# tokenizer = AutoTokenizer.from_pretrained(model_id)
-
-# # Create a text input field
-# text = st.text_input("Enter some text:")
-
-# # If the user enters some text, generate the speech
-# if text:
-#     # Tokenize the text
-#     inputs = tokenizer(text=text, return_tensors="pt")
-
-#     # Generate the speech
-#     generated_speech = model.generate(**inputs)
-
-#     # Play the speech
-#     st.audio(generated_speech, format="wav")
-
-
+#     st.write(f"Confidence: {confidence:.2f}")