Update app.py

app.py

@@ -6,9 +6,10 @@ import plotly.graph_objects as go
 from scipy.stats import norm, t
 from scipy.cluster.hierarchy import linkage, dendrogram, fcluster
 import plotly.figure_factory as ff
-#from vllm import LLM, SamplingParams
 from huggingface_hub import InferenceClient
 from sklearn.cluster import KMeans
+import json
+import math
 
 def sorting(df):
     df.index = list(map(float, df.index))
@@ -544,116 +545,29 @@ def upload_and_select_dataframe():
         st.sidebar.info("Please upload some files.")
         return None
 
+def sample_size_calculator(confidence_level, p, E):
+    Z = norm.ppf(1 - (1 - confidence_level) / 2)
     
-import json
-
-def categorize_responses_persian(initial_prompt: str, 
-                                dataframe: pd.DataFrame,
-                                id_column: str,
-                                text_column: str,
-                                api_key: str,
-                                max_retries: int = 3,
-                                delay: float = 1.5) -> pd.DataFrame:
+    n = (Z**2 * p * (1 - p)) / (E**2)
     
-    # اعتبارسنجی اولیه
-    def validate_api():
-        """بررسی سلامت API قبل از پردازش کل داده"""
-        test_payload = {
-            "model": "deepseek-chat",
-            "messages": [{"role": "user", "content": "سلام"}]
-        }
-        try:
-            response = requests.post(
-                "https://api.deepseek.com/v1/chat/completions",
-                headers={"Authorization": f"Bearer {api_key}"},
-                json=test_payload
-            )
-            response.raise_for_status()
-            return True
-        except Exception as e:
-            print(f"خطای احراز هویت API: {str(e)}")
-            print(f"پاسخ کامل API: {response.text if 'response' in locals() else ''}")
-            return False
-
-    if not validate_api():
-        raise ConnectionError("اتصال به API برقرار نشد. کلید API یا اتصال اینترنت را بررسی کنید.")
-
-    # پردازش اصلی
-    headers = {
-        "Authorization": f"Bearer {api_key}",
-        "Content-Type": "application/json; charset=utf-8"
-    }
-
-    def get_persian_category(answer: str) -> str:
-        messages = [
-            {
-                "role": "system",
-                "content": f"{initial_prompt}\n\nلطفا پاسخ را به یکی از دسته‌های تعیین شده اختصاص دهید. فقط نام دسته را بازگردانید."
-            },
-            {
-                "role": "user",
-                "content": answer
-            }
-        ]
-
-        payload = {
-            "model": "deepseek-chat",
-            "messages": messages,
-            "temperature": 0.3,
-            "max_tokens": 50
-        }
-
-        for attempt in range(max_retries):
-            try:
-                response = requests.post(
-                    "https://api.deepseek.com/v1/chat/completions",
-                    headers=headers,
-                    json=payload,
-                    timeout=10
-                )
-                
-                # دیباگ پاسخ خام
-                print(f"\nوضعیت HTTP: {response.status_code}")
-                print("هدرهای پاسخ:")
-                print(json.dumps(dict(response.headers), indent=2))
-                print("بدنه پاسخ:")
-                print(response.text)
-
-                response.raise_for_status()
-                
-                result = response.json()
-                return result['choices'][0]['message']['content'].strip()
-
-            except requests.exceptions.HTTPError as err:
-                print(f"\nخطای HTTP: {err}")
-                print(f"جزئیات خطا: {response.text}")
-                if response.status_code == 401:
-                    raise PermissionError("کلید API نامعتبر است")
-                time.sleep(2 ** attempt)  # Backoff نمایی
-
-            except Exception as e:
-                print(f"\nخطای عمومی: {str(e)}")
-                time.sleep(1)
+    n = math.ceil(n)
+    
+    return n
 
-        return "خطا: عدم دسترسی به API"
+#empty_col1, main_col, empty_col2 = st.columns([1.6, 2.8, 1.6])
 
-    # اجرا با نمایش پیشرفت
-    tqdm.pandas(desc="در حال پردازش پاسخ‌ها")
-    dataframe['category'] = dataframe[text_column].progress_apply(get_persian_category)
-    
-    return dataframe
-empty_col1, main_col, empty_col2 = st.columns([1.6, 2.8, 1.6])
+#with main_col:
+#    col1, col2 = st.columns([2, 3])
 
-with main_col:
-    col1, col2 = st.columns([2, 3])
+#    with col1:
+#        st.image("logo.png", width=400)
 
-    with col1:
-        st.image("logo.png", width=400)
+#    with col2:
+#        st.title("Insightzen")
 
-    with col2:
-        st.title("Chortke")
+#st.markdown('[Click to register a suggestion or comment](https://docs.google.com/forms/d/e/1FAIpQLScLyP7bBbqMfGdspjL7Ij64UZ6v2KjqjKNbm8gwEsgWsFs_Qg/viewform?usp=header)')
 
-st.markdown('[Click to register a suggestion or comment](https://docs.google.com/forms/d/e/1FAIpQLScLyP7bBbqMfGdspjL7Ij64UZ6v2KjqjKNbm8gwEsgWsFs_Qg/viewform?usp=header)')
+st.image("logo.png", width=600)
 
 df = upload_and_select_dataframe()
 
@@ -664,7 +578,7 @@ try:
     cols = edit_strings(df.columns)
     cols = sorted(list(set(cols)))
     
-    main_option = st.selectbox("Please select an option:", ["Tabulation", "Funnel Analysis", "Segmentation Analysis", "Hypothesis test", "Machine Learning", "Coding", "AI Chat"])
+    main_option = st.selectbox("Please select an option:", ["Tabulation", "Funnel Analysis", "Segmentation Analysis", "Hypothesis test", "Machine Learning", "Sample Size Calculator" ,"Coding", "AI Chat"])
     
     if main_option == "Tabulation":
         st.header("Tabulation Analysis")
@@ -1040,5 +954,16 @@ try:
         for chunk in stream:
             st.warning(chunk.choices[0].delta.content, end="")
 
+    elif main_option == "Sample Size Calculator":
+        st.header("Sample Size Calculator")
+        
+        confidence_level = int(st.text_input("Confidence levels:"))
+        p = int(st.text_input("Estimated probability of success:"))
+        E = int(st.text_input("Margin of error:"))
+
+        n = sample_size_calculator(confidence_level, p, E)
+
+        st.write(f"Sample size: {n}")
+
 except Exception as e:
         st.error(f"❌ Error: {e}")