add db

app.py

@@ -1,22 +1,66 @@
 import gradio as gr
 from openai import OpenAI
+import os
+import chromadb
+from datasets import load_dataset
 
+ds = load_dataset("RTVS/SpotifyLyrics001")[:500]
+apikey=os.getenv('apikey')
 client = OpenAI(
   base_url="https://openrouter.ai/api/v1",
-  api_key="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",
+  api_key=f"{apikey}",
 )
 
+# Hàm chuẩn bị chạy trước Gradio
+def setup_chromadb():
+    global collection  # để dùng lại trong các hàm Gradio nếu cần
+    client = chromadb.Client()
+    collection_name = "my_collection"
+    
+    existing_collections = [col.name for col in client.list_collections()]
+    if collection_name not in existing_collections:
+        print(f"Creating collection: {collection_name}")
+        collection = client.create_collection(name=collection_name)     
+        ids = []
+        i = 0
+        for song,artist in zip(ds['train']['song'], ds['train']['artist']):
+            ids.append(f"Index: {i} - Name song:{song} - Artis:{artist}") 
+            i+=1
+        collection.add(
+            documents=ds['train']['text'],
+            ids=ids
+        )
+    else:
+        print(f"Collection {collection_name} already exists.")
+        collection = client.get_collection(name=collection_name)
+
+    return
+
+    
+setup_chromadb()
 
 def respond(
     message,
     history: list[tuple[str, str]],
 ):
     response = ""
-
+    musics = collection.query(
+    query_texts=[message], 
+    n_results=10 
+    )['ids'][0]
     response = client.chat.completions.create(
         extra_body={},
         model="openrouter/optimus-alpha",
         messages=[
+            {
+            "role": "system",
+            "content": [
+                {
+                "type": "text",
+                "text": f"Generate a recommendation for the song based on user and this list: {musics}"
+                },
+            ]
+            },
             {
             "role": "user",
             "content": [
@@ -35,19 +79,7 @@ def respond(
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """
 demo = gr.ChatInterface(
-    respond,
-    # additional_inputs=[
-    #     gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-    #     gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-    #     gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-    #     gr.Slider(
-    #         minimum=0.1,
-    #         maximum=1.0,
-    #         value=0.95,
-    #         step=0.05,
-    #         label="Top-p (nucleus sampling)",
-    #     ),
-    # ],
+    fn=respond,
 )
 
 

requirements.txt

@@ -1 +1,5 @@
-openai
+openai
+huggingface_hub
+datasets
+chromadb
+sentence_transformers

test_script/import_data_to_rag.py

@@ -0,0 +1,21 @@
+import chromadb
+
+chroma_client = chromadb.Client()
+
+collection = chroma_client.create_collection(name="SpotifyLyrics")
+
+collection.add(
+    documents=[
+        "This is a document about pineapple",
+        "This is a document about oranges",
+        "This is a query document about hawaii"
+    ],
+    ids=["id1", "id2", "id3"],
+)
+query_texts=["fruit"]
+
+results = collection.query(
+    query_texts=query_texts, 
+    n_results=2 
+)
+print(results)

test_script/summarize_text.py

@@ -0,0 +1,33 @@
+from datasets import load_dataset
+
+ds = load_dataset("RTVS/SpotifyLyrics001")
+from openai import OpenAI
+
+client = OpenAI(
+  base_url="https://openrouter.ai/api/v1",
+  api_key="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",
+)
+summarized_lyrics = []
+
+for index in range(len(ds['train']['text'])):
+    message = ds['train']['text'][index]
+    response = client.chat.completions.create(
+    extra_body={},
+    model="openrouter/optimus-alpha",
+    messages=[
+        {
+        "role": "user",
+        "content": [
+            {
+            "type": "text",
+            "text": f"summarize this lyrics of the song: {message}"
+            },
+        ]
+        }
+    ]
+    ).choices[0].message.content
+    print("ok")
+    summarized_lyrics.append(response)
+ds = ds.add_column("summarized-lyrics", summarized_lyrics)
+ds.push_to_hub("tqhuyen/Spotify-Lyrics-Summarized")
+