Spaces:
Runtime error
Runtime error
File size: 61,114 Bytes
9922ae0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 18,
"id": "bbac0476",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"The gradio extension is already loaded. To reload it, use:\n",
" %reload_ext gradio\n"
]
}
],
"source": [
"%load_ext gradio"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a83f8fbf",
"metadata": {},
"outputs": [],
"source": [
"%%blocks\n",
"import gradio as gr\n",
"from llama_cpp import Llama\n",
"import llama_cpp\n",
"from langchain.callbacks.manager import CallbackManager\n",
"from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler\n",
"\n",
"\n",
"llm = Llama(\n",
" model_path=r\"C:\\Users\\user\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
" n_gpu_layers=100,\n",
" n_batch=512,\n",
" n_ctx=3000,\n",
" f16_kv=True,\n",
" callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
" verbose=False,\n",
")\n",
"\n",
"with gr.Blocks() as demo:\n",
" name=\"cora\"\n",
" gr.Markdown(f\"# Greetings {name}!\")\n",
" inp = gr.Textbox()\n",
" out = gr.Textbox()\n",
"\n",
" inp.change(fn=lambda x: x, inputs=inp, outputs=out)\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "c51b8778",
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"import gradio as gr\n",
"\n",
"# 定義處理函數\n",
"def process_user_input(message):\n",
" return message\n",
"\n",
"# 定義主函數\n",
"def main_pipeline(message, history):\n",
" # 呼叫處理函數\n",
" response = process_user_input(message)\n",
" # 將輸出加入歷史訊息\n",
" return response\n",
"\n",
"# 創建 Gradio 介面\n",
"chat_interface = gr.Interface(fn=main_pipeline,inputs=\"text\",outputs=\"text\",live=True)\n",
"\n",
"# 啟動應用程式\n",
"if __name__ == \"__main__\":\n",
" \n",
" chat_interface.launch()\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3ff36b6c-1b1d-4703-96f7-65642fae5722",
"metadata": {},
"outputs": [],
"source": [
"import gradio as gr\n",
"\n",
"def process_user_input(message):\n",
" return message\n",
"\n",
"def main_pipeline(message, history):\n",
" response = process_user_input(message)\n",
" return response\n",
"\n",
"chat_interface = gr.ChatInterface(main_pipeline, type=\"messages\")\n",
"\n",
"if __name__ == \"__main__\":\n",
" chat_interface.launch()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "ee5766db-3500-4082-a634-2b1cdad5859b",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "4d1d8fe7",
"metadata": {},
"outputs": [],
"source": [
"import gradio as gr\n",
"from llama_cpp import Llama\n",
"from langchain_community.llms import LlamaCpp\n",
"from langchain.prompts import PromptTemplate\n",
"import llama_cpp\n",
"from langchain.callbacks.manager import CallbackManager\n",
"from sentence_transformers import SentenceTransformer\n",
"from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler\n",
"import numpy as np\n",
"import pandas as pd\n",
"import re\n",
"import os\n",
"from sklearn.metrics.pairwise import cosine_similarity\n",
"\n",
"# 定義輔助函式\n",
"def process_user_input(message):\n",
" return message\n",
"\n",
"# # 假設 PromptTemplate 和 invoke_with_temperature 已正確定義\n",
"# user_mental_state4 = PromptTemplate(\n",
"# input_variables=[\"input\"],\n",
"# template=\"\"\"...\"\"\"\n",
"# )\n",
" \n",
"# df_user = pd.DataFrame(columns=[\"輸入內容\", \"形容詞1\", \"形容詞2\", \"形容詞3\", \"角色1\", \"角色2\", \"角色3\"])\n",
"# prompt_value1 = user_mental_state4.invoke({\"input\": message})\n",
"# string = invoke_with_temperature(prompt_value1)\n",
"# adjectives = [adj.strip() for adj in re.split('[,、，]', string)]\n",
"# index = len(df_user)\n",
"# df_user.loc[index, '輸入內容'] = message\n",
"# if len(adjectives) == 3:\n",
"# df_user.loc[index, '形容詞1'] = adjectives[0]\n",
"# df_user.loc[index, '形容詞2'] = adjectives[1]\n",
"# df_user.loc[index, '形容詞3'] = adjectives[2]\n",
"# df_user.to_excel(\"user_gradio系統.xlsx\")\n",
"# return message\n",
"\n",
"# 主邏輯\n",
"def main_pipeline(message, history):\n",
" df_user = process_user_input(message)\n",
" return df_user\n",
"\n",
"demo=gr.ChatInterface(main_pipeline)\n",
"\n",
"# 主程式進入點\n",
"if __name__ == \"__main__\":\n",
" \n",
" demo.launch()\n",
" \n",
"# import gradio as gr\n",
"\n",
"# # 定義處理函數\n",
"# def process_user_input(message):\n",
"# return message\n",
"\n",
"# # 定義主函數\n",
"# def main_pipeline(message, history):\n",
"# # 呼叫處理函數\n",
"# response = process_user_input(message)\n",
"# # 將輸出加入歷史訊息\n",
"# return response\n",
"\n",
"# # 創建 Gradio 介面\n",
"# chat_interface = gr.ChatInterface(main_pipeline)\n",
"\n",
"# # 啟動應用程式\n",
"# if __name__ == \"__main__\":\n",
"# chat_interface.launch()\n",
" "
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "4df2a74d",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"WARNING:tensorflow:From C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\tf_keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\n",
"\n",
"Running on local URL: http://127.0.0.1:7864\n",
"\n",
"To create a public link, set `share=True` in `launch()`.\n"
]
},
{
"data": {
"text/html": [
"<div><iframe src=\"http://127.0.0.1:7864/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\analytics.py:106: UserWarning: IMPORTANT: You are using gradio version 4.44.0, however version 4.44.1 is available, please upgrade. \n",
"--------\n",
" warnings.warn(\n",
"Traceback (most recent call last):\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\queueing.py\", line 536, in process_events\n",
" response = await route_utils.call_process_api(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\route_utils.py\", line 322, in call_process_api\n",
" output = await app.get_blocks().process_api(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\blocks.py\", line 1935, in process_api\n",
" result = await self.call_function(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\blocks.py\", line 1518, in call_function\n",
" prediction = await fn(*processed_input)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\utils.py\", line 793, in async_wrapper\n",
" response = await f(*args, **kwargs)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\chat_interface.py\", line 623, in _submit_fn\n",
" response = await anyio.to_thread.run_sync(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\to_thread.py\", line 56, in run_sync\n",
" return await get_async_backend().run_sync_in_worker_thread(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 2441, in run_sync_in_worker_thread\n",
" return await future\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 943, in run\n",
" result = context.run(func, *args)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Temp\\ipykernel_17468\\2638884024.py\", line 229, in main_pipeline\n",
" df_filter=filter(sorted_df)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Temp\\ipykernel_17468\\2638884024.py\", line 162, in filter\n",
" p=len(df_user)-1\n",
"NameError: name 'df_user' is not defined\n"
]
}
],
"source": [
"import gradio as gr\n",
"from llama_cpp import Llama\n",
"from langchain_community.llms import LlamaCpp\n",
"from langchain.prompts import PromptTemplate\n",
"import llama_cpp\n",
"from langchain.callbacks.manager import CallbackManager\n",
"from sentence_transformers import SentenceTransformer\n",
"from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler\n",
"import numpy as np\n",
"import pandas as pd\n",
"import re\n",
"import os\n",
"from sklearn.metrics.pairwise import cosine_similarity\n",
"\n",
"model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-mpnet-base-v2',device='cpu')\n",
"\n",
"# llm = LlamaCpp(\n",
"# model_path=r\"C:\\Users\\Cora\\.cache\\lm-studio\\models\\YC-Chen\\Breeze-7B-Instruct-v1_0-GGUF\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
"# n_gpu_layers=100,\n",
"# n_batch=512,\n",
"# n_ctx=3000,\n",
"# f16_kv=True,\n",
"# callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
"# verbose=False,\n",
"# )\n",
"\n",
"llm = LlamaCpp(\n",
" model_path=r\"C:\\Users\\user\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
" n_gpu_layers=100,\n",
" n_batch=512,\n",
" n_ctx=3000,\n",
" f16_kv=True,\n",
" callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
" verbose=False,\n",
")\n",
"\n",
"embedd_bk=pd.read_pickle(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞_677.pkl\")\n",
"df_bk=pd.read_excel(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞.xlsx\")\n",
"\n",
"def invoke_with_temperature(prompt, temperature=0.4):\n",
" return llm.invoke(prompt, temperature=temperature)\n",
"\n",
"def process_user_input(message):\n",
" user_mental_state4= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的人的心理困擾<</SYS>> \n",
" 請根據{input}描述三個最有可能心理困擾，輸出只包含三個心理困擾，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
" user_character= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的大學生，在生活中的多重角色身分<</SYS>> \n",
" 請你根據談話內容{input}，客觀的判斷說話的大學生，在談話內容中的角色，以及他生活中其他角色的身分，提供三個最有可能的角色身分名詞，\n",
" 輸出只包含三個身分名詞，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
"\n",
" df_user=pd.DataFrame(columns=[\"輸入內容\",\"形容詞1\", \"形容詞2\", \"形容詞3\", \"角色1\", \"角色2\", \"角色3\"])\n",
" #df_user_record=pd.read_excel(r\"C:\\Users\\Cora\\推薦系統實作\\gradio系統歷史紀錄.xlsx\")\n",
" \n",
"\n",
" prompt_value1=user_mental_state4.invoke({\"input\":message})\n",
" string=invoke_with_temperature(prompt_value1)\n",
" #print(\"\\n\")\n",
"\n",
" # 將字符串分割為名詞\n",
" adjectives = [adj.strip() for adj in re.split('[,、，]', string)]\n",
" \n",
" index=len(df_user)\n",
" df_user.loc[index, '輸入內容'] = message\n",
"\n",
" # 確保形容詞數量符合欄位數量\n",
" if len(adjectives) == 3:\n",
" df_user.loc[index, '形容詞1'] = adjectives[0]\n",
" df_user.loc[index, '形容詞2'] = adjectives[1]\n",
" df_user.loc[index, '形容詞3'] = adjectives[2]\n",
" df_user.to_excel(\"user_gradio系統.xlsx\")\n",
" return df_user\n",
"\n",
"def embedd_df_user(df_user):\n",
" \n",
" columns_to_encode=df_user.loc[:,[\"形容詞1\", \"形容詞2\", \"形容詞3\"]]\n",
"\n",
" # 初始化一個空的 DataFrame，用來存儲向量化結果\n",
" embedd_user=df_user[[\"輸入內容\"]]\n",
" #user_em= user_em.assign(形容詞1=None, 形容詞2=None, 形容詞3=None,角色1=None,角色2=None,角色3=None)\n",
" embedd_user= embedd_user.assign(形容詞1=None, 形容詞2=None, 形容詞3=None)\n",
" \n",
"\n",
" # 遍歷每一個單元格，將結果存入新的 DataFrame 中\n",
" i=len(df_user)-1\n",
" for col in columns_to_encode:\n",
" #print(i,col)\n",
" # 將每個單元格的內容進行向量化\n",
" embedd_user.at[i, col] = model.encode(df_user.at[i, col]) \n",
" \n",
" embedd_user.to_pickle(r\"C:\\Users\\user\\推薦系統實作\\user_gradio系統.pkl\")\n",
" \n",
" return embedd_user\n",
"\n",
"def top_n_books_by_average(df, n=3):\n",
" \n",
" # 根据 `average` 列降序排序\n",
" sorted_df = df.sort_values(by='average', ascending=False)\n",
" \n",
" # 选择前 N 行\n",
" top_n_df = sorted_df.head(n)\n",
" \n",
" # 提取书名列\n",
" top_books = top_n_df['書名'].tolist()\n",
" \n",
" return top_books,sorted_df\n",
"\n",
"def similarity(embedd_user,embedd_bk,df_bk):\n",
" df_similarity= pd.DataFrame(df_bk[['書名',\"內容簡介\",\"URL\",\"形容詞1\", \"形容詞2\", \"形容詞3\", '角色1', '角色2', '角色3']])\n",
" df_similarity['average'] = np.nan\n",
" #for p in range(len(embedd_user)): \n",
" index=len(embedd_user)-1 \n",
" for k in range(len(embedd_bk)):\n",
" list=[]\n",
" for i in range(1,4):\n",
" for j in range(3,6):\n",
" vec1=embedd_user.iloc[index,i]#i是第i個形容詞，數字是第幾個是使用者輸入\n",
" vec2=embedd_bk.iloc[k,j]\n",
" similarity = cosine_similarity([vec1], [vec2])\n",
" list.append(similarity[0][0])\n",
" # 计算总和\n",
" total_sum = sum(list)\n",
" # 计算数量\n",
" count = len(list)\n",
" # 计算平均值\n",
" average = total_sum / count\n",
" df_similarity.loc[k,'average']=average\n",
"\n",
" top_books,sorted_df = top_n_books_by_average(df_similarity)\n",
" return sorted_df \n",
"\n",
"def filter(sorted_df):\n",
" filter_prompt4 = PromptTemplate(\n",
" input_variables=[\"mental_issue\", \"user_identity\",\" book\",\"book_reader\", \"book_description\"],\n",
" template=\"\"\"[INST]<<SYS>>你是專業的心理諮商師和書籍推薦專家，擅長根據使用者的心理問題、身份特質，以及書名、書籍針對的主題和適合的讀者，判斷書籍是否適合推薦給使用者。\n",
"\n",
" 你的目的是幫助讀者找到可以緩解心理問題的書籍。請注意：\n",
" 1. 若書籍針對的問題與使用者的心理問題有關聯，即使書籍適合的讀者群與使用者身份沒有直接關聯，應偏向推薦。\n",
" 2. 若使用者身份的需求與書籍針對的問題有潛在關聯，應偏向推薦。\n",
" 3. 若書籍適合的讀者與使用者身份特質有任何關聯，應傾向推薦。\n",
" 4. 若書名跟使用者的心理問題或身分特質有任何關聯，應偏向推薦<</SYS>>\n",
"\n",
" 使用者提供的資訊如下：\n",
" 使用者身份是「{user_identity}」，其心理問題是「{mental_issue}」。書名是{book}，書籍適合的讀者群為「{book_reader}」，書籍針對的問題是「{book_description}」。\n",
"\n",
" 請根據以上資訊判斷這本書是否適合推薦給該使用者。\n",
" 僅輸出「是」或「否」，輸出後即停止。[/INST]\"\"\"\n",
" )\n",
" df_filter=sorted_df.iloc[:20,:]\n",
" df_filter = df_filter.reset_index(drop=True)\n",
" df_filter=df_filter.assign(推薦=None)\n",
" #df_similarity= pd.DataFrame(df_bk[['書名',\"內容簡介\",\"URL\",\"形容詞1\", \"形容詞2\", \"形容詞3\", '角色1', '角色2', '角色3']])\n",
" #df_similarity['average'] = np.nan\n",
"\n",
" \n",
" p=len(df_user)-1\n",
" for k in range(len(df_filter)): \n",
" word=df_user[\"輸入內容\"].iloc[p]\n",
" #book_reader = df_filter[\"角色1\"].iloc[p] + \"or\" + df_filter[\"角色2\"].iloc[p] + \"or\" + df_filter[\"角色3\"].iloc[p]\n",
" book=df_filter[\"書名\"].iloc[k] \n",
" book_reader = df_filter[\"角色1\"].iloc[k] \n",
" user_identity = df_user[\"角色1\"].iloc[p]\n",
" mental_issue=df_user[\"形容詞1\"].iloc[p]+\"、\"+df_user[\"形容詞2\"].iloc[p]+\"、\"+df_user[\"形容詞3\"].iloc[p]\n",
" book_description=df_filter[\"形容詞1\"].iloc[k]+\"、\"+df_filter[\"形容詞2\"].iloc[k]+\"、\"+df_filter[\"形容詞3\"].iloc[k]\n",
" print(book_reader)\n",
" print(user_identity)\n",
" #output = filter_prompt1.invoke({\"user_identity\": user_identity, \"book_reader\": book_reader})\n",
" output = filter_prompt4.invoke({\"mental_issue\":mental_issue,\"user_identity\": user_identity, \"book\":book,\"book_description\":book_description,\"book_reader\": book_reader})\n",
" string2=invoke_with_temperature(output)\n",
" df_filter.loc[k, '推薦'] =string2\n",
" df_recommend=df_filter[df_filter[\"推薦\"].str.strip() == \"是\"]\n",
" \n",
" return df_recommend\n",
"def output_content(df_recommend):\n",
" content_prompt = PromptTemplate(\n",
" input_variables=[\"content\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>你是一個有同理心的心理師，\n",
" 請根據{content}，用平易近人且不官方的語氣，先介紹這本書的內容，總共約50-70字[/INST]\"\"\"\n",
" )\n",
"\n",
" a=0\n",
" title=df_recommend.loc[a,\"書名\"]\n",
" #URL=sorted_df.iloc[a,1]\n",
" #content=sorted_df.iloc[a,2]\n",
" \n",
"# prompt_value2=content_prompt.invoke({\"content\":content})\n",
"# summary=invoke_with_temperature(prompt_value2)\n",
"# recommend_prompt = PromptTemplate(\n",
"# input_variables=[\"title\",\"URL\",\"summary\"],\n",
"# template=\"\"\"<<SYS>>\n",
"# 你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>\n",
"# [INST] \n",
"# 請根據{title}{URL}{summary}產出訊息，開頭不要有空格，並在最後面加入一句或兩句鼓勵的話\n",
"# 格式為：根據您的狀態，這裡提供一本書供您參考\\n\n",
"# 書名：{title}\\n\n",
"# 本書介紹:{summary}\\n\n",
"# 購書網址：{URL}\\n\n",
"# 希望對您有所幫助\n",
"# [/INST]\"\"\"\n",
"# )\n",
"# prompt_value1=recommend_prompt.invoke({\"title\":title,\"URL\":URL,\"summary\":summary})\n",
" \n",
" recommend_prompt = PromptTemplate(\n",
" input_variables=[\"title\"],\n",
" template=\"\"\"<<SYS>>\n",
" 你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>\n",
" [INST] \n",
" 請根據{title}產出訊息，開頭不要有空格，並在最後面加入一句或兩句鼓勵的話\n",
" 格式為：根據您的狀態，這裡提供一本書供您參考\\n\n",
" 書名：{title}\\n\n",
" 希望對您有所幫助\n",
" [/INST]\"\"\"\n",
" )\n",
" prompt_value1=recommend_prompt.invoke({\"title\":title})\n",
" output=invoke_with_temperature(prompt_value1,temperature=0.4)\n",
" return output \n",
" \n",
"def main_pipeline(message,history):\n",
" \n",
" df_user=process_user_input(message)\n",
" embedd_user=embedd_df_user(df_user)\n",
" sorted_df=similarity(embedd_user,embedd_bk,df_bk)\n",
" df_filter=filter(sorted_df)\n",
" final=output_content(df_filter)\n",
" return final \n",
" \n",
"\n",
"# def recommend(message,history):\n",
"# result=main_pipeline(message)\n",
"# return result\n",
"\n",
"demo=gr.ChatInterface(main_pipeline,type=\"messages\")\n",
"\n",
"# with gr.Blocks() as demo:\n",
"# gr.Markdown(\"Start typing below and then click **Run** to see the output.\")\n",
"# with gr.Row():\n",
"# inp = gr.Textbox(placeholder=\"What is your name?\")\n",
"# out = gr.Textbox()\n",
"# btn = gr.Button(\"Run\")\n",
"# btn.click(fn=recommend, inputs=inp, outputs=out)\n",
"if __name__ == \"__main__\":\n",
" demo.launch()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "487c853d",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Running on local URL: http://127.0.0.1:7866\n",
"\n",
"To create a public link, set `share=True` in `launch()`.\n"
]
},
{
"data": {
"text/html": [
"<div><iframe src=\"http://127.0.0.1:7866/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\analytics.py:106: UserWarning: IMPORTANT: You are using gradio version 4.44.0, however version 4.44.1 is available, please upgrade. \n",
"--------\n",
" warnings.warn(\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
" 情緒控制困難,壓力負荷過高,人際衝突"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Traceback (most recent call last):\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\queueing.py\", line 536, in process_events\n",
" response = await route_utils.call_process_api(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\route_utils.py\", line 322, in call_process_api\n",
" output = await app.get_blocks().process_api(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\blocks.py\", line 1935, in process_api\n",
" result = await self.call_function(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\blocks.py\", line 1518, in call_function\n",
" prediction = await fn(*processed_input)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\utils.py\", line 793, in async_wrapper\n",
" response = await f(*args, **kwargs)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\chat_interface.py\", line 623, in _submit_fn\n",
" response = await anyio.to_thread.run_sync(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\to_thread.py\", line 56, in run_sync\n",
" return await get_async_backend().run_sync_in_worker_thread(\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 2441, in run_sync_in_worker_thread\n",
" return await future\n",
" File \"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\anyio\\_backends\\_asyncio.py\", line 943, in run\n",
" result = context.run(func, *args)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Temp\\ipykernel_17468\\1758736236.py\", line 78, in main_pipeline\n",
" df_filter=filter(sorted_df)\n",
" File \"C:\\Users\\user\\AppData\\Local\\Temp\\ipykernel_17468\\2638884024.py\", line 162, in filter\n",
" p=len(df_user)-1\n",
"NameError: name 'df_user' is not defined\n"
]
}
],
"source": [
"import gradio as gr\n",
"from llama_cpp import Llama\n",
"from langchain_community.llms import LlamaCpp\n",
"from langchain.prompts import PromptTemplate\n",
"import llama_cpp\n",
"from langchain.callbacks.manager import CallbackManager\n",
"from sentence_transformers import SentenceTransformer\n",
"from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler\n",
"import numpy as np\n",
"import pandas as pd\n",
"import re\n",
"import os\n",
"from sklearn.metrics.pairwise import cosine_similarity\n",
"\n",
"model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-mpnet-base-v2',device='cpu')\n",
"\n",
"llm = LlamaCpp(\n",
" model_path=r\"C:\\Users\\user\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
" n_gpu_layers=100,\n",
" n_batch=512,\n",
" n_ctx=3000,\n",
" f16_kv=True,\n",
" callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
" verbose=False,\n",
")\n",
"\n",
"embedd_bk=pd.read_pickle(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞_677.pkl\")\n",
"df_bk=pd.read_excel(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞.xlsx\")\n",
"\n",
"def invoke_with_temperature(prompt, temperature=0.4):\n",
" return llm.invoke(prompt, temperature=temperature)\n",
"\n",
"def process_user_input(message):\n",
" \n",
" user_mental_state4= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的人的心理困擾<</SYS>> \n",
" 請根據{input}描述三個最有可能心理困擾，輸出只包含三個心理困擾，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
" user_character= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的大學生，在生活中的多重角色身分<</SYS>> \n",
" 請你根據談話內容{input}，客觀的判斷說話的大學生，在談話內容中的角色，以及他生活中其他角色的身分，提供三個最有可能的角色身分名詞，\n",
" 輸出只包含三個身分名詞，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
"\n",
" df_user=pd.DataFrame(columns=[\"輸入內容\",\"形容詞1\", \"形容詞2\", \"形容詞3\", \"角色1\", \"角色2\", \"角色3\"])\n",
" #df_user_record=pd.read_excel(r\"C:\\Users\\Cora\\推薦系統實作\\gradio系統歷史紀錄.xlsx\")\n",
" \n",
"\n",
" prompt_value1=user_mental_state4.invoke({\"input\":message})\n",
" string=invoke_with_temperature(prompt_value1)\n",
" #print(\"\\n\")\n",
"\n",
" # 將字符串分割為名詞\n",
" adjectives = [adj.strip() for adj in re.split('[,、，]', string)]\n",
" \n",
" index=len(df_user)\n",
" df_user.loc[index, '輸入內容'] = message\n",
"\n",
" # 確保形容詞數量符合欄位數量\n",
" if len(adjectives) == 3:\n",
" df_user.loc[index, '形容詞1'] = adjectives[0]\n",
" df_user.loc[index, '形容詞2'] = adjectives[1]\n",
" df_user.loc[index, '形容詞3'] = adjectives[2]\n",
" df_user.to_excel(\"user_gradio系統.xlsx\")\n",
" return df_user\n",
"\n",
" \n",
" \n",
"def main_pipeline(message,history):\n",
" \n",
" df_user=process_user_input(message)\n",
" embedd_user=embedd_df_user(df_user)\n",
" sorted_df=similarity(embedd_user,embedd_bk,df_bk)\n",
" df_filter=filter(sorted_df)\n",
" final=output_content(df_filter)\n",
" return final \n",
" \n",
"\n",
"\n",
"demo=gr.ChatInterface(main_pipeline,type=\"messages\")\n",
"\n",
"\n",
"if __name__ == \"__main__\":\n",
" demo.launch()"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "b3cadc4a-6f63-4038-bcfb-ef419ad5394a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Running on local URL: http://127.0.0.1:7873\n",
"\n",
"To create a public link, set `share=True` in `launch()`.\n"
]
},
{
"data": {
"text/html": [
"<div><iframe src=\"http://127.0.0.1:7873/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\user\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\gradio\\analytics.py:106: UserWarning: IMPORTANT: You are using gradio version 4.44.0, however version 4.44.1 is available, please upgrade. \n",
"--------\n",
" warnings.warn(\n"
]
}
],
"source": [
"import gradio as gr\n",
"from llama_cpp import Llama\n",
"from langchain_community.llms import LlamaCpp\n",
"from langchain.prompts import PromptTemplate\n",
"import llama_cpp\n",
"from langchain.callbacks.manager import CallbackManager\n",
"from sentence_transformers import SentenceTransformer\n",
"from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler\n",
"import numpy as np\n",
"import pandas as pd\n",
"import re\n",
"import os\n",
"from sklearn.metrics.pairwise import cosine_similarity\n",
"\n",
"model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-mpnet-base-v2',device='cpu')\n",
"\n",
"title=\"書籍推薦平台\"\n",
"\n",
"# llm = LlamaCpp(\n",
"# model_path=r\"C:\\Users\\Cora\\.cache\\lm-studio\\models\\YC-Chen\\Breeze-7B-Instruct-v1_0-GGUF\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
"# n_gpu_layers=100,\n",
"# n_batch=512,\n",
"# n_ctx=3000,\n",
"# f16_kv=True,\n",
"# callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
"# verbose=False,\n",
"# )\n",
"\n",
"llm = LlamaCpp(\n",
" model_path=r\"C:\\Users\\user\\breeze-7b-instruct-v1_0-q4_k_m.gguf\",\n",
" n_gpu_layers=100,\n",
" n_batch=512,\n",
" n_ctx=3000,\n",
" f16_kv=True,\n",
" callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]),\n",
" verbose=False,\n",
")\n",
"\n",
"embedd_bk=pd.read_pickle(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞_677.pkl\")\n",
"df_bk=pd.read_excel(r\"C:\\Users\\user\\推薦系統實作\\bk_description1_角色形容詞.xlsx\")\n",
"\n",
"def invoke_with_temperature(prompt, temperature=0.4):\n",
" return llm.invoke(prompt, temperature=temperature)\n",
"\n",
"def process_user_input(message):\n",
" user_mental_state4= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的人的心理困擾<</SYS>> \n",
" 請根據{input}描述三個最有可能心理困擾，輸出只包含三個心理困擾，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
" user_character= PromptTemplate(\n",
" input_variables=[\"input\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一位具有同理心的專業心理諮商師，沒有性別歧視，你可以客觀的根據談話內容的描述，判斷說話的大學生，在生活中的多重角色身分<</SYS>> \n",
" 請你根據談話內容{input}，客觀的判斷說話的大學生，在談話內容中的角色，以及他生活中其他角色的身分，提供三個最有可能的角色身分名詞，\n",
" 輸出只包含三個身分名詞，回答格式只採用CSV格式，分隔符號使用逗號，參考以下範例：名詞1,名詞2,名詞3。[/INST]\"\"\"\n",
" )\n",
" \n",
"\n",
" df_user=pd.DataFrame(columns=[\"輸入內容\",\"形容詞1\", \"形容詞2\", \"形容詞3\", \"角色1\", \"角色2\", \"角色3\"])\n",
" #df_user_record=pd.read_excel(r\"C:\\Users\\Cora\\推薦系統實作\\gradio系統歷史紀錄.xlsx\")\n",
" \n",
"\n",
" prompt_value1=user_mental_state4.invoke({\"input\":message})\n",
" string=invoke_with_temperature(prompt_value1)\n",
" #print(\"\\n\")\n",
"\n",
" # 將字符串分割為名詞\n",
" adjectives = [adj.strip() for adj in re.split('[,、，]', string)]\n",
" \n",
" index=len(df_user)\n",
" df_user.loc[index, '輸入內容'] = message\n",
"\n",
" # 確保形容詞數量符合欄位數量\n",
" if len(adjectives) == 3:\n",
" df_user.loc[index, '形容詞1'] = adjectives[0]\n",
" df_user.loc[index, '形容詞2'] = adjectives[1]\n",
" df_user.loc[index, '形容詞3'] = adjectives[2]\n",
" df_user.to_excel(\"user_gradio系統.xlsx\")\n",
" return df_user\n",
" #return message\n",
"\n",
"def embedd_df_user(df_user):\n",
" \n",
" columns_to_encode=df_user.loc[:,[\"形容詞1\", \"形容詞2\", \"形容詞3\"]]\n",
"\n",
" # 初始化一個空的 DataFrame，用來存儲向量化結果\n",
" embedd_user=df_user[[\"輸入內容\"]]\n",
" #user_em= user_em.assign(形容詞1=None, 形容詞2=None, 形容詞3=None,角色1=None,角色2=None,角色3=None)\n",
" embedd_user= embedd_user.assign(形容詞1=None, 形容詞2=None, 形容詞3=None)\n",
" \n",
"\n",
" # 遍歷每一個單元格，將結果存入新的 DataFrame 中\n",
" i=len(df_user)-1\n",
" for col in columns_to_encode:\n",
" #print(i,col)\n",
" # 將每個單元格的內容進行向量化\n",
" embedd_user.at[i, col] = model.encode(df_user.at[i, col]) \n",
" \n",
" embedd_user.to_pickle(r\"C:\\Users\\user\\推薦系統實作\\user_gradio系統.pkl\")\n",
" \n",
" return embedd_user\n",
" #word=\"happy\"\n",
" #return word\n",
"\n",
"def top_n_books_by_average(df, n=3):\n",
" \n",
" # 根据 `average` 列降序排序\n",
" sorted_df = df.sort_values(by='average', ascending=False)\n",
" \n",
" # 选择前 N 行\n",
" top_n_df = sorted_df.head(n)\n",
" \n",
" # 提取书名列\n",
" top_books = top_n_df['書名'].tolist()\n",
" \n",
" return top_books,sorted_df\n",
"\n",
"def similarity(embedd_user,embedd_bk,df_bk):\n",
" df_similarity= pd.DataFrame(df_bk[['書名',\"內容簡介\",\"URL\",\"形容詞1\", \"形容詞2\", \"形容詞3\", '角色1', '角色2', '角色3']])\n",
" df_similarity['average'] = np.nan\n",
" #for p in range(len(embedd_user)): \n",
" index=len(embedd_user)-1 \n",
" for k in range(len(embedd_bk)):\n",
" list=[]\n",
" for i in range(1,4):\n",
" for j in range(3,6):\n",
" vec1=embedd_user.iloc[index,i]#i是第i個形容詞，數字是第幾個是使用者輸入\n",
" vec2=embedd_bk.iloc[k,j]\n",
" similarity = cosine_similarity([vec1], [vec2])\n",
" list.append(similarity[0][0])\n",
" # 计算总和\n",
" total_sum = sum(list)\n",
" # 计算数量\n",
" count = len(list)\n",
" # 计算平均值\n",
" average = total_sum / count\n",
" df_similarity.loc[k,'average']=average\n",
"\n",
" top_books,sorted_df = top_n_books_by_average(df_similarity)\n",
" return sorted_df \n",
"\n",
"def filter(sorted_df,df_user):\n",
" filter_prompt4 = PromptTemplate(\n",
" input_variables=[\"mental_issue\", \"user_identity\",\" book\",\"book_reader\", \"book_description\"],\n",
" template=\"\"\"[INST]<<SYS>>你是專業的心理諮商師和書籍推薦專家，擅長根據使用者的心理問題、身份特質，以及書名、書籍針對的主題和適合的讀者，判斷書籍是否適合推薦給使用者。\n",
"\n",
" 你的目的是幫助讀者找到可以緩解心理問題的書籍。請注意：\n",
" 1. 若書籍針對的問題與使用者的心理問題有關聯，即使書籍適合的讀者群與使用者身份沒有直接關聯，應偏向推薦。\n",
" 2. 若使用者身份的需求與書籍針對的問題有潛在關聯，應偏向推薦。\n",
" 3. 若書籍適合的讀者與使用者身份特質有任何關聯，應傾向推薦。\n",
" 4. 若書名跟使用者的心理問題或身分特質有任何關聯，應偏向推薦<</SYS>>\n",
"\n",
" 使用者提供的資訊如下：\n",
" 使用者身份是「{user_identity}」，其心理問題是「{mental_issue}」。書名是{book}，書籍適合的讀者群為「{book_reader}」，書籍針對的問題是「{book_description}」。\n",
"\n",
" 請根據以上資訊判斷這本書是否適合推薦給該使用者。\n",
" 僅輸出「是」或「否」，輸出後即停止。[/INST]\"\"\"\n",
" )\n",
" df_filter=sorted_df.iloc[:20,:]\n",
" df_filter = df_filter.reset_index(drop=True)\n",
" df_filter=df_filter.assign(推薦=None)\n",
" #df_similarity= pd.DataFrame(df_bk[['書名',\"內容簡介\",\"URL\",\"形容詞1\", \"形容詞2\", \"形容詞3\", '角色1', '角色2', '角色3']])\n",
" #df_similarity['average'] = np.nan\n",
"\n",
" \n",
" p=len(df_user)-1\n",
" for k in range(len(df_filter)): \n",
" word=df_user[\"輸入內容\"].iloc[p]\n",
" #book_reader = df_filter[\"角色1\"].iloc[p] + \"or\" + df_filter[\"角色2\"].iloc[p] + \"or\" + df_filter[\"角色3\"].iloc[p]\n",
" book=df_filter[\"書名\"].iloc[k] \n",
" book_reader = df_filter[\"角色1\"].iloc[k] \n",
" user_identity = df_user[\"角色1\"].iloc[p]\n",
" mental_issue=df_user[\"形容詞1\"].iloc[p]+\"、\"+df_user[\"形容詞2\"].iloc[p]+\"、\"+df_user[\"形容詞3\"].iloc[p]\n",
" book_description=df_filter[\"形容詞1\"].iloc[k]+\"、\"+df_filter[\"形容詞2\"].iloc[k]+\"、\"+df_filter[\"形容詞3\"].iloc[k]\n",
" print(book_reader)\n",
" print(user_identity)\n",
" #output = filter_prompt1.invoke({\"user_identity\": user_identity, \"book_reader\": book_reader})\n",
" output = filter_prompt4.invoke({\"mental_issue\":mental_issue,\"user_identity\": user_identity, \"book\":book,\"book_description\":book_description,\"book_reader\": book_reader})\n",
" string2=invoke_with_temperature(output)\n",
" df_filter.loc[k, '推薦'] =string2\n",
" df_recommend=df_filter[df_filter[\"推薦\"].str.strip() == \"是\"]\n",
" \n",
" return df_recommend\n",
" \n",
"def output_content(df_recommend):\n",
" content_prompt = PromptTemplate(\n",
" input_variables=[\"content\"],\n",
" template=\"\"\"[INST]<<SYS>>你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>你是一個有同理心的心理師，\n",
" 請根據{content}，用平易近人且不官方的語氣，先介紹這本書的內容，總共約50-70字[/INST]\"\"\"\n",
" )\n",
"\n",
" a=0\n",
" title=df_recommend.iloc[a,0]#不用loc，因為filter的時候index沒有重新歸零\n",
" #URL=sorted_df.iloc[a,1]\n",
" #content=sorted_df.iloc[a,2]\n",
" \n",
"# prompt_value2=content_prompt.invoke({\"content\":content})\n",
"# summary=invoke_with_temperature(prompt_value2)\n",
"# recommend_prompt = PromptTemplate(\n",
"# input_variables=[\"title\",\"URL\",\"summary\"],\n",
"# template=\"\"\"<<SYS>>\n",
"# 你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>\n",
"# [INST] \n",
"# 請根據{title}{URL}{summary}產出訊息，開頭不要有空格，並在最後面加入一句或兩句鼓勵的話\n",
"# 格式為：根據您的狀態，這裡提供一本書供您參考\\n\n",
"# 書名：{title}\\n\n",
"# 本書介紹:{summary}\\n\n",
"# 購書網址：{URL}\\n\n",
"# 希望對您有所幫助\n",
"# [/INST]\"\"\"\n",
"# )\n",
"# prompt_value1=recommend_prompt.invoke({\"title\":title,\"URL\":URL,\"summary\":summary})\n",
" \n",
" recommend_prompt = PromptTemplate(\n",
" input_variables=[\"title\"],\n",
" template=\"\"\"<<SYS>>\n",
" 你是一個有同理心的心理師，負責推薦相關書籍給使用者<</SYS>>\n",
" [INST] \n",
" 請根據{title}產出訊息，開頭不要有空格，並在最後面加入一句或兩句鼓勵的話\n",
" 格式為：根據您的狀態，這裡提供一本書供您參考\\n\n",
" 書名：{title}\\n\n",
" 希望對您有所幫助\n",
" [/INST]\"\"\"\n",
" )\n",
" prompt_value1=recommend_prompt.invoke({\"title\":title})\n",
" output=invoke_with_temperature(prompt_value1,temperature=0.4)\n",
" return output \n",
" \n",
"def main_pipeline(message,history):\n",
" \n",
" df_user=process_user_input(message)\n",
" embedd_user=embedd_df_user(df_user)\n",
" sorted_df=similarity(embedd_user,embedd_bk,df_bk)\n",
" df_filter=filter(sorted_df,df_user)\n",
" final=output_content(df_filter)\n",
" return final\n",
" #return embedd_user\n",
" \n",
" \n",
" \n",
"\n",
"# def recommend(message,history):\n",
"# result=main_pipeline(message)\n",
"# return result\n",
"\n",
"demo=gr.ChatInterface(main_pipeline,type=\"messages\")\n",
"\n",
"\n",
"if __name__ == \"__main__\":\n",
" demo.launch()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "aaad2fb2-a8d8-46e4-b9d4-c276f5ef0cb0",
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"pip install tf-keras"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "cc344ace-81f1-45f2-ad4a-9cca508aa053",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>書名</th>\n",
" <th>內容簡介</th>\n",
" <th>URL</th>\n",
" <th>形容詞1</th>\n",
" <th>形容詞2</th>\n",
" <th>形容詞3</th>\n",
" <th>角色1</th>\n",
" <th>角色2</th>\n",
" <th>角色3</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>這僅有一次的人生, 我不想說抱歉</td>\n",
" <td>你走太快了，容易迷路，要等靈魂跟上來，才能走更遠的路。那些你想要做成的事情，你做成了的事情，...</td>\n",
" <td>https://www.eslite.com/product/100120106326824...</td>\n",
" <td>自我期許</td>\n",
" <td>自我反思</td>\n",
" <td>人生目標</td>\n",
" <td>自我提升</td>\n",
" <td>人生感悟</td>\n",
" <td>內心成長</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>我只是想分手而已: 親密殺人, 被深愛的男人殺死的女人們</td>\n",
" <td>親密殺人不是約會暴力是整個社會必須全力阻止的連續殺人！只是想跟他分手的我，為何最後卻送了命？...</td>\n",
" <td>https://www.eslite.com/product/100120106326824...</td>\n",
" <td>心理創傷</td>\n",
" <td>暴力受暴經驗</td>\n",
" <td>感情困境</td>\n",
" <td>法律系學生</td>\n",
" <td>社會工作者</td>\n",
" <td>性別平權運動者</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>輕鬆思考法: 培養靈活觀點的150個啟示</td>\n",
" <td>本書特色150則啟示點醒沉浮於忙碌生活的現代人！篇幅短小、內容精闢，1分鐘打開新思維！擁有不...</td>\n",
" <td>https://www.eslite.com/product/100121372526824...</td>\n",
" <td>焦慮</td>\n",
" <td>孤獨</td>\n",
" <td>成長</td>\n",
" <td>職場人</td>\n",
" <td>旅行愛好者</td>\n",
" <td>追求自我成長的人</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>因為人類思維太僵化, 所以需要創新心理學: 心態革命, 大腦中的髮夾彎, 掀起你的思路風暴</td>\n",
" <td>大腦中的髮夾彎，掀起你的思路風暴！從理性到感性，不同思考方式將會開啟新的視角、新的世界！逆向...</td>\n",
" <td>https://www.eslite.com/product/100122024826824...</td>\n",
" <td>焦慮</td>\n",
" <td>壓力</td>\n",
" <td>抑鬱</td>\n",
" <td>好奇心旺盛的思考愛好者</td>\n",
" <td>希望提高日常解決問題技巧的人</td>\n",
" <td>渴望提升創新思維能力的人</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>人生心理學</td>\n",
" <td>找出人生發展的路向從來都不容易，無論你是臨近畢業的大專生，或是已在職場上打滾了好些年正在瓶頸...</td>\n",
" <td>https://www.eslite.com/product/100121238026824...</td>\n",
" <td>生涯規劃</td>\n",
" <td>自我覺察</td>\n",
" <td>人生意義</td>\n",
" <td>大學生</td>\n",
" <td>職場工作者</td>\n",
" <td>心理學研究者</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>672</th>\n",
" <td>你想要的一切, 宇宙早已為你預備</td>\n",
" <td>宇宙會把最好的獻給你。如果你願意放下自我限制的信念，全然信任這股奇妙的力量，豐盛的人生自然會...</td>\n",
" <td>https://www.eslite.com/product/100120176426824...</td>\n",
" <td>困惑</td>\n",
" <td>壓力</td>\n",
" <td>不滿</td>\n",
" <td>焦慮症患者</td>\n",
" <td>心理負荷過重者</td>\n",
" <td>靈性追求者</td>\n",
" </tr>\n",
" <tr>\n",
" <th>673</th>\n",
" <td>妄想的力量: 迷信、儀式感與過度樂觀的非理性心理學</td>\n",
" <td>妄想雖然可恥但是有用！亞馬遜讀者五星強推！最熱愛怪力亂神的美國心理學暢銷作家帶你重新認識幻想...</td>\n",
" <td>https://www.eslite.com/product/100120106326824...</td>\n",
" <td>妄想</td>\n",
" <td>心理假象</td>\n",
" <td>樂觀</td>\n",
" <td>心理學家</td>\n",
" <td>精神病患者家屬</td>\n",
" <td>普通大眾</td>\n",
" </tr>\n",
" <tr>\n",
" <th>674</th>\n",
" <td>悲傷復原力: 一位心理學專家, 也是位失去愛女的母親, 透過復原力心理學, 走過分離崩解的悲傷</td>\n",
" <td>面對至親至愛的離去，如果悲傷難免，我們可以做些什麼，度過這場巨大風暴？紐約時報、華爾街日報，...</td>\n",
" <td>https://www.eslite.com/product/100121380726824...</td>\n",
" <td>喪親之痛</td>\n",
" <td>悲傷</td>\n",
" <td>復原力</td>\n",
" <td>親人失去</td>\n",
" <td>悲傷修復</td>\n",
" <td>心理健康</td>\n",
" </tr>\n",
" <tr>\n",
" <th>675</th>\n",
" <td>淬鍊幸福, 剛剛好的回憶練習 (限量贈暖心陪伴藏書卡)</td>\n",
" <td>為什麼自己會突然情緒崩潰？從什麼時候開始，變得越來越少話？每當回憶起某件事時，就會止不住的落...</td>\n",
" <td>https://www.eslite.com/product/100120303926824...</td>\n",
" <td>創傷後壓力症候群</td>\n",
" <td>自我懷疑</td>\n",
" <td>內心傷痛</td>\n",
" <td>創傷癒後者</td>\n",
" <td>單親父母</td>\n",
" <td>成長經歷過困難的讀者</td>\n",
" </tr>\n",
" <tr>\n",
" <th>676</th>\n",
" <td>不是為了爭吵才跟你在一起: 如何在溝通中改善親密關係</td>\n",
" <td>為什麼開始親密無間的兩個人，會在關係中越走越遠、越來越疏離？外人對你們羨慕不已，但其實是假性...</td>\n",
" <td>https://www.eslite.com/product/100120106326824...</td>\n",
" <td>焦慮</td>\n",
" <td>疏離</td>\n",
" <td>衝突</td>\n",
" <td>兩性關係人士</td>\n",
" <td>婚姻治療師或專家</td>\n",
" <td>伴侶或夫妻</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>677 rows × 9 columns</p>\n",
"</div>"
],
"text/plain": [
" 書名 \\\n",
"0 這僅有一次的人生, 我不想說抱歉 \n",
"1 我只是想分手而已: 親密殺人, 被深愛的男人殺死的女人們 \n",
"2 輕鬆思考法: 培養靈活觀點的150個啟示 \n",
"3 因為人類思維太僵化, 所以需要創新心理學: 心態革命, 大腦中的髮夾彎, 掀起你的思路風暴 \n",
"4 人生心理學 \n",
".. ... \n",
"672 你想要的一切, 宇宙早已為你預備 \n",
"673 妄想的力量: 迷信、儀式感與過度樂觀的非理性心理學 \n",
"674 悲傷復原力: 一位心理學專家, 也是位失去愛女的母親, 透過復原力心理學, 走過分離崩解的悲傷 \n",
"675 淬鍊幸福, 剛剛好的回憶練習 (限量贈暖心陪伴藏書卡) \n",
"676 不是為了爭吵才跟你在一起: 如何在溝通中改善親密關係 \n",
"\n",
" 內容簡介 \\\n",
"0 你走太快了，容易迷路，要等靈魂跟上來，才能走更遠的路。那些你想要做成的事情，你做成了的事情，... \n",
"1 親密殺人不是約會暴力是整個社會必須全力阻止的連續殺人！只是想跟他分手的我，為何最後卻送了命？... \n",
"2 本書特色150則啟示點醒沉浮於忙碌生活的現代人！篇幅短小、內容精闢，1分鐘打開新思維！擁有不... \n",
"3 大腦中的髮夾彎，掀起你的思路風暴！從理性到感性，不同思考方式將會開啟新的視角、新的世界！逆向... \n",
"4 找出人生發展的路向從來都不容易，無論你是臨近畢業的大專生，或是已在職場上打滾了好些年正在瓶頸... \n",
".. ... \n",
"672 宇宙會把最好的獻給你。如果你願意放下自我限制的信念，全然信任這股奇妙的力量，豐盛的人生自然會... \n",
"673 妄想雖然可恥但是有用！亞馬遜讀者五星強推！最熱愛怪力亂神的美國心理學暢銷作家帶你重新認識幻想... \n",
"674 面對至親至愛的離去，如果悲傷難免，我們可以做些什麼，度過這場巨大風暴？紐約時報、華爾街日報，... \n",
"675 為什麼自己會突然情緒崩潰？從什麼時候開始，變得越來越少話？每當回憶起某件事時，就會止不住的落... \n",
"676 為什麼開始親密無間的兩個人，會在關係中越走越遠、越來越疏離？外人對你們羨慕不已，但其實是假性... \n",
"\n",
" URL 形容詞1 形容詞2 \\\n",
"0 https://www.eslite.com/product/100120106326824... 自我期許 自我反思 \n",
"1 https://www.eslite.com/product/100120106326824... 心理創傷 暴力受暴經驗 \n",
"2 https://www.eslite.com/product/100121372526824... 焦慮 孤獨 \n",
"3 https://www.eslite.com/product/100122024826824... 焦慮 壓力 \n",
"4 https://www.eslite.com/product/100121238026824... 生涯規劃 自我覺察 \n",
".. ... ... ... \n",
"672 https://www.eslite.com/product/100120176426824... 困惑 壓力 \n",
"673 https://www.eslite.com/product/100120106326824... 妄想 心理假象 \n",
"674 https://www.eslite.com/product/100121380726824... 喪親之痛 悲傷 \n",
"675 https://www.eslite.com/product/100120303926824... 創傷後壓力症候群 自我懷疑 \n",
"676 https://www.eslite.com/product/100120106326824... 焦慮 疏離 \n",
"\n",
" 形容詞3 角色1 角色2 角色3 \n",
"0 人生目標 自我提升 人生感悟 內心成長 \n",
"1 感情困境 法律系學生 社會工作者 性別平權運動者 \n",
"2 成長 職場人 旅行愛好者 追求自我成長的人 \n",
"3 抑鬱 好奇心旺盛的思考愛好者 希望提高日常解決問題技巧的人 渴望提升創新思維能力的人 \n",
"4 人生意義 大學生 職場工作者 心理學研究者 \n",
".. ... ... ... ... \n",
"672 不滿 焦慮症患者 心理負荷過重者 靈性追求者 \n",
"673 樂觀 心理學家 精神病患者家屬 普通大眾 \n",
"674 復原力 親人失去 悲傷修復 心理健康 \n",
"675 內心傷痛 創傷癒後者 單親父母 成長經歷過困難的讀者 \n",
"676 衝突 兩性關係人士 婚姻治療師或專家 伴侶或夫妻 \n",
"\n",
"[677 rows x 9 columns]"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_bk"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "b88bb194-7064-4dcb-8907-b392d8f1e82f",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.8"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
|