File size: 7,949 Bytes
d955a63
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e0a3cde9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import openai\n",
    "from langchain_openai import ChatOpenAI, OpenAIEmbeddings\n",
    "import tiktoken\n",
    "from langchain.prompts.few_shot import FewShotPromptTemplate\n",
    "from langchain.prompts.prompt import PromptTemplate\n",
    "from operator import itemgetter\n",
    "from langchain.schema import StrOutputParser\n",
    "from langchain_core.output_parsers import StrOutputParser\n",
    "from langchain_core.runnables import RunnablePassthrough\n",
    "\n",
    "import langchain_community.embeddings.huggingface\n",
    "# help(langchain_community.embeddings.huggingface)\n",
    "from langchain_community.embeddings.huggingface import HuggingFaceBgeEmbeddings\n",
    "from langchain_community.vectorstores import FAISS\n",
    "\n",
    "from langchain.chains import LLMChain\n",
    "from langchain.chains.conversation.memory import ConversationBufferMemory, ConversationBufferWindowMemory, ConversationSummaryMemory, ConversationSummaryBufferMemory\n",
    "\n",
    "import os, dotenv\n",
    "from dotenv import load_dotenv\n",
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bf5aac26",
   "metadata": {},
   "outputs": [],
   "source": [
    "# embeddings = OpenAIEmbeddings(deployment=\"textembedding\", chunk_size = 16, api_key = os.environ[\"OPENAI_API_KEY\"])\n",
    "# index_name = \"SCLC\"\n",
    "\n",
    "model_name = \"bge-large-en-v1.5\"\n",
    "model_kwargs = {\"device\": \"cpu\"}\n",
    "# model_kwargs = {\"device\": \"cuda\"}\n",
    "encode_kwargs = {\"normalize_embeddings\": True}\n",
    "embeddings = HuggingFaceBgeEmbeddings(\n",
    "    # model_name=model_name, \n",
    "    model_kwargs = model_kwargs,\n",
    "    encode_kwargs = encode_kwargs)\n",
    "\n",
    "index_name = \"indexes/ChestPainRubrics\"\n",
    "\n",
    "# store = FAISS.load_local(index_name, embeddings)\n",
    "import db_firestore as db\n",
    "store = db.get_store(index_name, embeddings=embeddings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2d75b44a",
   "metadata": {},
   "outputs": [],
   "source": [
    "TEMPLATE = \"\"\"You are a teacher for medical students. Your task is to provide an overall assessment of a student's diagnosis, based on the \n",
    "You will be provided with the following information:\n",
    "1. The rubrics that the student should be judged based upon.\n",
    "2. The conversation history between the medical student and the patient.\n",
    "3. The final diagnosis that the student will make.\n",
    "\n",
    "Your grading should touch on every part of the rubrics, and grade the student holistically.\n",
    "Finally, provide an overall grade for the student.\n",
    "\n",
    "Here is the rubrics:\n",
    "{context}\n",
    "\n",
    "----------------------------------------------------------------\n",
    "You are to give a comprehensive judgement based on the student's diagnosis, with reference to the above rubrics.\n",
    "\n",
    "Here is the chat history, enclosed in triple backticks:\n",
    "```\n",
    "{chat_history}\n",
    "```\n",
    "----------------------------------------------------------------\n",
    "\n",
    "Student's final diagnosis:\n",
    "{question}\n",
    "----------------------------------------------------------------\n",
    "Your grade:\n",
    "\"\"\"\n",
    "\n",
    "prompt = PromptTemplate(\n",
    "    input_variables = [\"question\", \"context\"],\n",
    "    template = TEMPLATE\n",
    ")\n",
    "retriever = store.as_retriever(search_type=\"similarity\", search_kwargs={\"k\":2})\n",
    "def format_docs(docs):\n",
    "    return \"\\n--------------------\\n\".join(doc.page_content for doc in docs)\n",
    "\n",
    "\n",
    "llm = ChatOpenAI(model_name=\"gpt-3.5-turbo\", temperature=0)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd14b945",
   "metadata": {},
   "source": [
    "## some code to generate the fake history easier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e7fcc9e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[]\n"
     ]
    }
   ],
   "source": [
    "fake_history = []\n",
    "i = \" \"\n",
    "\n",
    "### Send no message to end it\n",
    "while i:\n",
    "    i = input()\n",
    "    if i:\n",
    "        fake_history.append(i)\n",
    "\n",
    "print(fake_history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "df09e0dc",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Sample history:\n",
    "\n",
    "fake_history = [\n",
    "    \"student: How are you mr tan?\",\n",
    "    \"patient: I am not feeling well...\",\n",
    "    \"student: oh, that's terrible! tell me more about what happened to you\",\n",
    "    \"patient: well I got this chest pain\",\n",
    "    \"student: I'm sorry to hear that. How long has your chest pain been?\",\n",
    "    \"patient: I can't remember\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "e6e21890",
   "metadata": {},
   "outputs": [],
   "source": [
    "## RESET MEMORY\n",
    "## To reset the memory (if it screws up), rerun this cell\n",
    "memory = ConversationSummaryBufferMemory(llm=llm, memory_key=\"chat_history\", input_key=\"question\" )\n",
    "\n",
    "chain = (\n",
    "    {\n",
    "        \"context\": retriever | format_docs, \n",
    "        \"question\": RunnablePassthrough(),\n",
    "        \"chat_history\": lambda x: '\\n'.join(fake_history)\n",
    "        } | \n",
    "    # prompt | \n",
    "    LLMChain(llm=llm, prompt=prompt, memory=memory, verbose=True) #| \n",
    "    # StrOutputParser()\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f443e8c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Based on the provided rubrics and the conversation history, I would assess the student's diagnosis as Grade E. \n",
      "\n",
      "The student's diagnosis of \"chest pain, but it will go away soon\" does not align with the symptoms and history provided by the patient. The patient's symptoms, such as acute tearing chest pain, radiation to the back, diaphoresis, syncope, and dysarthria, are indicative of a more serious condition, such as aortic dissection or acute coronary syndrome. The student did not consider these possibilities or explore them further.\n",
      "\n",
      "Additionally, the student did not adequately explore the patient's presenting complaints, associated symptoms, and relevant medical history. They did not elicit all the necessary information and did not rule out red flags. The student also did not present a comprehensive list of differentials with adequate justification.\n",
      "\n",
      "Overall, the student's diagnosis and assessment of the patient's condition are incomplete and do not meet the expected standards.\n"
     ]
    }
   ],
   "source": [
    "results = chain.invoke(\"I believe he has chest pain, but it will go away soon\")\n",
    "print(results.get(\"text\"))"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}