thbndi commited on
Commit
a72fda4
1 Parent(s): 4ca1b39

Update readme.md

Browse files
Files changed (1) hide show
  1. readme.md +7 -7
readme.md CHANGED
@@ -12,20 +12,20 @@ load_dataset('thbndi/Mimic4Dataset', task, mimic_path=mimic_data, config_path=co
12
 
13
  ## Arguments
14
 
15
- 1. `task` (string, default="Mortality"):
16
  - Description: Specifies the task you want to perform with the dataset.
17
  - Default: "Mortality"
18
  - Note: Possible Values : 'Phenotype', 'Length of Stay', 'Readmission', 'Mortality'
19
 
20
- 2. `mimic_path` (string):
21
  - Description: Complete path to the Mimic-IV raw data on user's machine.
22
  - Note: You need to provide the appropriate path where the Mimic-IV data is stored.
23
 
24
- 3. `config_path` (string) optionnal:
25
  - Description: Path to the configuration file for the cohort generation choices (more infos in '/config/readme.md').
26
  - Default: Configuration file provided in the 'config' folder.
27
 
28
- 4. `encoding` (string, default="concat"):
29
  - Description: Data encoding option for the features.
30
  - Options: "concat", "aggreg", "tensor", "raw", "text"
31
  - Default: "concat"
@@ -36,17 +36,17 @@ load_dataset('thbndi/Mimic4Dataset', task, mimic_path=mimic_data, config_path=co
36
  - "raw": Provide cohort from the pipeline without any encoding for custom data processing.
37
  - "text": Represents diagnoses as text suitable for BERT or other similar text-based models.
38
 
39
- 5. `generate_cohort` (bool, default=True):
40
  - Description: Determines whether to generate a new cohort from Mimic-IV data.
41
  - Default: True
42
  - Note: Set it to True to generate a cohort, or False to skip cohort generation.
43
 
44
- 6. `val_size`, 'test_size' (float):
45
  - Description: Proportion of the dataset used for validation during training.
46
  - Default: 0.1 for validation size and 0.2 for testing size.
47
  - Note: Can be set to 0.
48
 
49
- 7. `cache_dir` (string):
50
  - Description: Directory where the processed dataset will be cached.
51
  - Note: Providing a cache directory for each encoding type can avoid errors when changing the encoding type.
52
 
 
12
 
13
  ## Arguments
14
 
15
+ 1. `task` (string) :
16
  - Description: Specifies the task you want to perform with the dataset.
17
  - Default: "Mortality"
18
  - Note: Possible Values : 'Phenotype', 'Length of Stay', 'Readmission', 'Mortality'
19
 
20
+ 2. `mimic_path` (string) :
21
  - Description: Complete path to the Mimic-IV raw data on user's machine.
22
  - Note: You need to provide the appropriate path where the Mimic-IV data is stored.
23
 
24
+ 3. `config_path` (string) optionnal :
25
  - Description: Path to the configuration file for the cohort generation choices (more infos in '/config/readme.md').
26
  - Default: Configuration file provided in the 'config' folder.
27
 
28
+ 4. `encoding` (string) optionnal :
29
  - Description: Data encoding option for the features.
30
  - Options: "concat", "aggreg", "tensor", "raw", "text"
31
  - Default: "concat"
 
36
  - "raw": Provide cohort from the pipeline without any encoding for custom data processing.
37
  - "text": Represents diagnoses as text suitable for BERT or other similar text-based models.
38
 
39
+ 5. `generate_cohort` (bool) optionnal :
40
  - Description: Determines whether to generate a new cohort from Mimic-IV data.
41
  - Default: True
42
  - Note: Set it to True to generate a cohort, or False to skip cohort generation.
43
 
44
+ 6. `val_size`, 'test_size' (float) optionnal :
45
  - Description: Proportion of the dataset used for validation during training.
46
  - Default: 0.1 for validation size and 0.2 for testing size.
47
  - Note: Can be set to 0.
48
 
49
+ 7. `cache_dir` (string) optionnal :
50
  - Description: Directory where the processed dataset will be cached.
51
  - Note: Providing a cache directory for each encoding type can avoid errors when changing the encoding type.
52