Delete pages/demo_type_text.py

pages/demo_type_text.py

@@ -1,205 +0,0 @@
-import streamlit as st
-import pandas as pd
-from io import StringIO
-import json
-import torch
-from transformers import pipeline # AutoTokenizer, AutoModelForCausalLM, AutoModelForTokenClassification
-from sentence_transformers import SentenceTransformer, util 
-
-import os
-os.getenv("HF_TOKEN") 
-
-#for k, v in st.session_state.items():
-#    st.session_state[k] = v 
-
-#st.title("📘Map internal description to SBS codes V2.0")
-#st.subheader("Select specific Chapter for quicker results")
-#df_chapters = pd.read_csv("SBS_V2_0/Chapter_Index_Rows.csv") 
-
-#startrowindex_list = df_chapters["from_row_index"].tolist()
-#endrowindex_list = df_chapters["to_row_index"].tolist()
-#allchapters_rows_list = []
-#for s, e in zip(startrowindex_list, endrowindex_list):
-#    eachchapter_rows_list = list(range(s,e))
-#    allchapters_rows_list.append(eachchapter_rows_list)
-#f_chapters['range_of_rows'] = allchapters_rows_list 
-
-def dataframe_with_selections(df_chapters: pd.DataFrame, init_value: bool = False) -> pd.DataFrame:
-    df_with_selections = df_chapters.copy()
-    df_with_selections.insert(0, "Select", init_value)
-
-    # Get dataframe row-selections from user with st.data_editor
-    edited_df = st.data_editor(
-        df_with_selections,
-        hide_index=True,
-        column_config={"Select": st.column_config.CheckboxColumn(required=True)},
-        disabled=df_chapters.columns,
-    )
-
-    # Filter the dataframe using the temporary column, then drop the column
-    selected_rows = edited_df[edited_df.Select]
-    return selected_rows.drop('Select', axis=1)
-
-#if "selected_chapters" not in st.session_state:
-#    st.session_state['selected_chapters'] = []
-#    st.session_state['selected_rows'] = []
-#selected_chapters_list = st.session_state.selected_chapters
-#if "selected_rows" not in st.session_state: 
-#    st.session_state['selected_rows'] = []
-#selected_rows_list = st.session_state.selected_rows
-
-#selected_chapters = dataframe_with_selections(df_chapters)
-#st.write("Your selection:")
-#st.write(selected_chapters) 
-#selected_rows = dataframe_with_selections(df_chapters)
-#st.write("Your selection:")
-#st.write(selected_rows)
-
-#selected_chapters_list = selected_chapters.iloc[:,0].tolist() 
-#st.write("SELECTED CHAPTERS: ", selected_chapters_list) 
-#selected_rows_list = selected_chapters.iloc[:,6].tolist() 
-#st.write("SELECTED ROWS: ", selected_rows_list) 
-
-#if selected_chapters is not None:
-#    st.session_state.selected_chapters = selected_chapters_list
-#    st.session_state.selected_rows = selected_rows_list 
-
-
-
-    
-#selected_chapters_floatlist = list(st.session_state.items())[0][1]
-#selected_chapters_intlist = [int(i) for i in selected_chapters_floatlist]
-#st.write("SELECTED CHAPTERS: ", selected_chapters_intlist)
-#for item in st.session_state.items():
-#    st.write("IIIIIIIII: ", item)
-#selected_rows_list = list(st.session_state.items())[1][1]
-#st.write("SELECTED ROWS: ", selected_rows_list)
-
-def get_device_map() -> str:
-    return 'cuda' if torch.cuda.is_available() else 'cpu'
-device = get_device_map()  # 'cpu'
-
-def on_click():
-    st.session_state.user_input = ""
-    
-#@st.cache
-def convert_df(df:pd.DataFrame):
-     return df.to_csv(index=False).encode('utf-8')
-
-#@st.cache
-def convert_json(df:pd.DataFrame):
-    result = df.to_json(orient="index")
-    parsed = json.loads(result)
-    json_string = json.dumps(parsed)
-    #st.json(json_string, expanded=True)
-    return json_string
-
-INTdesc_input = st.text_input("Type internal description", key="user_input") 
-
-createSBScodes, right_column = st.columns(2)
-createSBScodes_clicked = createSBScodes.button("Map to SBS codes", key="user_createSBScodes")
-right_column.button("Reset", on_click=on_click)
-
-numMAPPINGS_input = 5
-#numMAPPINGS_input = st.text_input("Type number of mappings", key="user_input_numMAPPINGS")
-#st.button("Clear text", on_click=on_click)
-
-@st.cache_resource
-def load_model():
-    model = SentenceTransformer('all-MiniLM-L6-v2') # fastest
-    #model = SentenceTransformer('all-mpnet-base-v2') # best performance
-    #model = SentenceTransformers('all-distilroberta-v1')
-    #model = SentenceTransformer('sentence-transformers/msmarco-bert-base-dot-v5') 
-    #model = SentenceTransformer('clips/mfaq')
-    return model
-model = load_model() 
-
-INTdesc_embedding = model.encode(INTdesc_input)
-
-# Semantic search, Compute cosine similarity between all pairs of SBS descriptions
-
-#df_allchaps = pd.read_csv("SBS_V2_0/Chapter_Index_Rows.csv", usecols=["Chapter", "from_row_index", "to_row_index"])
-#st.dataframe(df_allchaps)
-#df_selectedchaps = df.loc[df['City'] == 'Chicago']
-
-#dict_allchaps = df_allchaps.to_dict(orient='index')
-#st.write("ALL CHAPTERS: ", dict_allchaps)
-#for chapter in dict_allchaps.get("Chapter"): 
-#    st.write(chapter)
-
-selected_rows_list = []
-
-#if len(selected_rows_list) == 0: 
-#    st.warning("Please select at least one chapter") 
-#    selected_rows_list = [0, 10080]
-    #st.write("SELECTED ROWS: ", selected_rows_list)
-#df_SBS = pd.read_csv("SBS_V2_0/Code_Table.csv", header=0, skip_blank_lines=False, skiprows = lambda x: x not in selected_rows_list)
-
-#df_SBS = pd.read_csv("SBS_V2_0/Code_Table.csv", index_col="SBS_Code", usecols=["Long_Description"]) # na_values=['NA']
-#df_SBS = pd.read_csv("SBS_V2_0/Code_Table.csv", usecols=["SBS_Code_Hyphenated","Long_Description"]) 
-from_row_index = 0 # Imaging services chapter start, adjust as needed
-to_row_index = 10080 # Imaging services chapter end, adjust as needed
-nrows = to_row_index - from_row_index + 1
-skiprows = list(range(1,from_row_index - 1))
-df_SBS = pd.read_csv("SBS_V2_0/Code_Table.csv", header=0, skip_blank_lines=False, skiprows=skiprows, nrows=nrows)
-st.write(df_SBS.head(5))
-
-SBScorpus = df_SBS['Long_Description'].values.tolist()
-SBScorpus_embeddings = model.encode(SBScorpus)
-
-#my_model_results = pipeline("ner", model= "checkpoint-92")
-HF_model_results = util.semantic_search(INTdesc_embedding, SBScorpus_embeddings)
-HF_model_results_sorted = sorted(HF_model_results, key=lambda x: x[1], reverse=True)
-HF_model_results_displayed = HF_model_results_sorted[0:numMAPPINGS_input]
-
-@st.cache_resource
-def load_pipe():
-    pipe = pipeline("text-generation", model="meta-llama/Llama-3.2-1B-Instruct", device_map=device,) # device_map="auto", torch_dtype=torch.bfloat16 
-    #pipe = pipeline("text-generation", model="Qwen/Qwen2-1.5B-Instruct", device_map=device,) # device_map="auto", torch_dtype="auto"
-    return pipe 
-pipe = load_pipe()
-
-dictA = {"Score": [], "SBS Code": [], "SBS Description V2.0": []}
-dfALL = pd.DataFrame.from_dict(dictA)
-
-if INTdesc_input is not None and createSBScodes_clicked == True: 
-    for i, result in enumerate(HF_model_results_displayed):
-        dictA.update({"Score": "%.4f" % result[0]["score"], "SBS Code": df_SBS.loc[df_SBS["Long_Description"] == SBScorpus[result[0]["corpus_id"]],"SBS_Code_Hyphenated"].values[0], "SBS Description V2.0": SBScorpus[result[0]["corpus_id"]]})
-        dfALL = pd.concat([dfALL, pd.DataFrame([dictA])], ignore_index=True)
-        dictA.update({"Score": "%.4f" % result[1]["score"], "SBS Code": df_SBS.loc[df_SBS["Long_Description"] == SBScorpus[result[1]["corpus_id"]],"SBS_Code_Hyphenated"].values[0], "SBS Description V2.0": SBScorpus[result[1]["corpus_id"]]})
-        dfALL = pd.concat([dfALL, pd.DataFrame([dictA])], ignore_index=True)
-        dictA.update({"Score": "%.4f" % result[2]["score"], "SBS Code": df_SBS.loc[df_SBS["Long_Description"] == SBScorpus[result[2]["corpus_id"]],"SBS_Code_Hyphenated"].values[0], "SBS Description V2.0": SBScorpus[result[2]["corpus_id"]]})
-        dfALL = pd.concat([dfALL, pd.DataFrame([dictA])], ignore_index=True)
-        dictA.update({"Score": "%.4f" % result[3]["score"], "SBS Code": df_SBS.loc[df_SBS["Long_Description"] == SBScorpus[result[3]["corpus_id"]],"SBS_Code_Hyphenated"].values[0], "SBS Description V2.0": SBScorpus[result[3]["corpus_id"]]})
-        dfALL = pd.concat([dfALL, pd.DataFrame([dictA])], ignore_index=True)
-        dictA.update({"Score": "%.4f" % result[4]["score"], "SBS Code": df_SBS.loc[df_SBS["Long_Description"] == SBScorpus[result[4]["corpus_id"]],"SBS_Code_Hyphenated"].values[0], "SBS Description V2.0": SBScorpus[result[4]["corpus_id"]]})
-        dfALL = pd.concat([dfALL, pd.DataFrame([dictA])], ignore_index=True)
-        
-    st.dataframe(data=dfALL, hide_index=True)
-
-    question = "Which one, if any, of the following Saudi Billing System descriptions A, B, C, D, or E corresponds best to " + INTdesc_input +"? " 
-    shortlist = [SBScorpus[result[0]["corpus_id"]], SBScorpus[result[1]["corpus_id"]], SBScorpus[result[2]["corpus_id"]], SBScorpus[result[3]["corpus_id"]], SBScorpus[result[4]["corpus_id"]]] 
-    prompt = question + " " +"A: "+ shortlist[0] + " " +"B: " + shortlist[1] + " " + "C: " + shortlist[2] + " " + "D: " + shortlist[3] + " " + "E: " + shortlist[4]
-    st.write(prompt)
-    
-    messages = [
-    {"role": "system", "content": "You are a knowledgable AI assistant who always answers truthfully and precisely!"},
-    {"role": "user", "content": prompt},
-    ]
-    outputs = pipe(
-        messages,
-        max_new_tokens=256,
-    )
-    st.write(outputs[0]["generated_text"][-1]["content"])
-    
-    bs, b1, b2, b3, bLast = st.columns([0.75, 1.5, 1.5, 1.5, 0.75])
-    with b1:
-        #csvbutton = download_button(results, "results.csv", "📥 Download .csv")
-        csvbutton = st.download_button(label="📥 Download .csv", data=convert_df(dfALL), file_name= "results.csv", mime='text/csv', key='csv_b')
-    with b2:
-        #textbutton = download_button(results, "results.txt", "📥 Download .txt")
-        textbutton = st.download_button(label="📥 Download .txt", data=convert_df(dfALL), file_name= "results.text", mime='text/plain',  key='text_b')
-    with b3:
-        #jsonbutton = download_button(results, "results.json", "📥 Download .json")
-        jsonbutton = st.download_button(label="📥 Download .json", data=convert_json(dfALL), file_name= "results.json", mime='application/json',  key='json_b')
-