basic questions generation

.env.example

@@ -1 +1,2 @@
-LASTFM_API_KEY=<your_lastfm_api_key>
+LASTFM_API_KEY=<your_lastfm_api_key>
+GROQ_API_KEY=<your_groq_api_key>

requirements.txt

@@ -1,4 +1,6 @@
 fastapi[standard]
 requests
 pandas
-ruff
+groq
+scikit-learn
+ruff

src/app/api/ai.py

@@ -0,0 +1,26 @@
+import os
+from dotenv import load_dotenv
+from groq import Groq
+
+load_dotenv()
+
+GROQ_API_KEY = os.getenv("GROQ_API_KEY")
+print(GROQ_API_KEY)
+client = Groq(
+  api_key=GROQ_API_KEY,
+)
+
+
+def ai(df):
+  chat_completion = client.chat.completions.create(
+    messages=[
+      {
+        "role": "user",
+        "content": "Generate 3 quiz questions about the data. Only ask questions about tracks with a cluster of 1 (don't mention cluster in the questions)"
+        + df.to_json(),
+      }
+    ],
+    model="llama-3.3-70b-versatile",
+  )
+
+  print(chat_completion.choices[0].message.content)

src/app/api/combine.py

@@ -1,10 +1,12 @@
+from ai import ai
 from fetch import fetch
-from model import preprocess
+from ml import ml
 
 
 def combine(username, period):
   metadata = {"username": username, "period": period}
   tracks = fetch(username, period)
-  preprocess(tracks)
+  df = ml(tracks)
+  ai(df)
   return_data = {"metadata": metadata, "tracks": tracks}
   return return_data

src/app/api/fetch.py

@@ -18,7 +18,7 @@ def get_top(username, period):
   clean = []
   counter = 0
   for track in raw:
-    if counter >= 5:
+    if counter >= 20:
       break
     clean.append(
       {
@@ -41,7 +41,7 @@ def get_info(data):
         API_URL, API_KEY, track["artist"], track["name"]
       )
     )
-    if counter >= 5:
+    if counter >= 20:
       break
     temp = response.json()
 

src/app/api/ml.py

@@ -0,0 +1,17 @@
+import pandas as pd
+from sklearn.cluster import KMeans
+import numpy as np
+
+
+def ml(data):
+  df = cluster(data)
+  return df
+
+
+def cluster(data):
+  df = pd.DataFrame(data)
+  play_counts = np.array(df["playcount"]).reshape(-1, 1)
+  kmeans = KMeans(n_clusters=3, random_state=42).fit(play_counts)
+  df["cluster"] = kmeans.labels_
+  print(df)
+  return df