Update app.py

app.py

@@ -1,3 +1,8 @@
+import os
+import openai
+api_key = os.environ.get('OPENAI_API_KEY')
+openai.api_key = api_key
+
 import numpy as np
 from PIL import Image
 import streamlit as st
@@ -8,11 +13,6 @@ model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-cap
 feature_extractor = ViTFeatureExtractor.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
 tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
 
-# Load the pre-trained model and tokenizer
-model_name = "gpt2"
-tokenizer_1 = GPT2Tokenizer.from_pretrained(model_name)
-model_2 = GPT2LMHeadModel.from_pretrained(model_name)
-
 def generate_captions(image):
     image = Image.open(image).convert("RGB")
     generated_caption = tokenizer.decode(model.generate(feature_extractor(image, return_tensors="pt").pixel_values.to("cpu"))[0])
@@ -21,21 +21,27 @@ def generate_captions(image):
     generated_caption = sentence.replace(text_to_remove, "")
     return generated_caption
 
-# Define the Streamlit app
-def generate_paragraph(prompt):
-    # Tokenize the prompt
-    input_ids = tokenizer_1.encode(prompt, return_tensors="pt")
+def generate_paragraph(caption):
+    prompt = "Generate a paragraph based on the following caption: " + caption
 
-    # Generate the paragraph
-    output = model_2.generate(input_ids, max_length=200, num_return_sequences=1, no_repeat_ngram_size=2, early_stopping=True)
+    # Make the API call to GPT-3
+    response = openai.Completion.create(
+        engine='text-davinci-003',  # Specify the GPT-3 model
+        prompt=prompt,
+        max_tokens=200,  # Adjust the desired length of the generated text
+        n = 1,  # Set the number of completions to generate
+        stop=None,  # Specify an optional stop sequence
+        temperature=0.7  # Adjust the temperature for randomness (between 0 and 1)
+    )
 
-    # Decode the generated output into text
-    paragraph = tokenizer_1.decode(output[0], skip_special_tokens=True)
-    return paragraph
+    # Extract the generated text from the API response
+    generated_text = response.choices[0].text.strip()
     
+    return generated_text
+
 # create the Streamlit app
 def app():
-    st.title('Image from your Side, Trending Hashtags from our Side')
+    st.title('Image from your Side, Detailed description from my site')
 
     st.write('Upload an image to see what we have in store.')