File size: 1,822 Bytes
c82994f
af3748c
 
 
 
 
 
 
 
 
c82994f
 
 
 
 
 
 
 
 
 
23e7d76
 
c82994f
 
 
 
a09bf1f
 
c82994f
a09bf1f
 
 
 
c82994f
 
 
 
 
 
 
af3748c
 
 
 
c82994f
af3748c
1e7b778
 
 
af3748c
 
 
 
 
 
e1855fa
af3748c
 
e1855fa
af3748c
 
 
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
---
language:
- ru
size_categories:
- 1K<n<10K
task_categories:
- question-answering
- text-classification
- text-generation
pretty_name: R
dataset_info:
  features:
  - name: topic
    dtype: string
  - name: user_question
    dtype: string
  - name: assistant_answer
    dtype: string
  - name: to_doctor
    dtype: string
  - name: __index_level_0__
    dtype: int64
  - name: prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 7652958.920651884
    num_examples: 3368
  - name: test
    num_bytes: 852096.0793481165
    num_examples: 375
  download_size: 3619976
  dataset_size: 8505055.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
tags:
- medical
- biology
- synthetic
---

# Russian-language dataset of 3743 patient conversations with a medical bot in QA manner. 
* The **training sample** includes 3014 conversations; 
* The **test sample** includes 335 conversations;

## Feature characteristics:
* *topic* - medical topic
* *user_question* - last user question
* *assistant_answer* - ai answer according the context and topic
* *to_doctor* - the specialty of the physician to whom the assistant referred the patient
* *prompt* - ready prompt for fincetuning instruct phi model (adapted for using with unsloth https://github.com/unslothai/unsloth?tab=readme-ov-file)

## Prompt format:
<|user|>USER QUESTION<|end|>\n<|assistant|>ASSISTANT ANSWER<|end|>

# Disclaimer
The data provided is solely for research purposes. It may include inaccuracies, potentially harmful content, or biases, and we disclaim any responsibility for these issues. Users should assess the data's accuracy and appropriateness before utilizing it. The opinions and views represented in the data do not necessarily align with those of the author.