Update README.md
Browse files
README.md
CHANGED
@@ -43,7 +43,7 @@ with open('./generation-configs/meta2diff.json', 'r') as f:
|
|
43 |
|
44 |
# prepare sample
|
45 |
config_sample = {"inputs": config_data, "mode": "meta2diff", "parameters": {
|
46 |
-
"temperature": 0.
|
47 |
"top_p": 0.2,
|
48 |
"top_k": 3550,
|
49 |
"n_next_tokens": 50
|
@@ -109,8 +109,7 @@ You can use the following instructions (one or several at a time):
|
|
109 |
|
110 |
1. disease2diff2disease - generate signature for disease
|
111 |
2. compound2diff2compound - generate signature for compound
|
112 |
-
3. age_group2diff2age_group - generate signature for age
|
113 |
-
4. age_individ2diff2age_individ - generate signature based on age value
|
114 |
|
115 |
|
116 |
### Other meta-data (`inputs.` in config)
|
@@ -164,20 +163,20 @@ Here we asked model to generate signature for Human, male, in tissue - whole blo
|
|
164 |
|
165 |
_Example 2_
|
166 |
|
167 |
-
You want to generate signature for healthy Human, male, 40 years, in tissue - whole blood.
|
168 |
```json
|
169 |
{
|
170 |
"inputs": {
|
171 |
-
"instruction": "disease2diff2disease",
|
172 |
"tissue": ["whole blood"],
|
173 |
-
"age":
|
174 |
"cell": "",
|
175 |
"efo": "",
|
176 |
"datatype": "",
|
177 |
"drug": "",
|
178 |
"dose": "",
|
179 |
"time": "",
|
180 |
-
"case": "",
|
181 |
"control": "",
|
182 |
"dataset_type": "expression",
|
183 |
"gender": "m",
|
@@ -196,7 +195,7 @@ You want to generate signature for healthy Human, male, 40 years, in tissue - wh
|
|
196 |
|
197 |
```
|
198 |
Note, here we used ```disease2diff2disease``` instruction, but we expected to generate signatures for healthy human, that's why we'd set ```efo``` to empty string "".
|
199 |
-
Alternatively, we can add one more instruction to example 2 - ```"instruction": ["disease2diff2disease", "
|
200 |
|
201 |
---
|
202 |
|
|
|
43 |
|
44 |
# prepare sample
|
45 |
config_sample = {"inputs": config_data, "mode": "meta2diff", "parameters": {
|
46 |
+
"temperature": 0.8,
|
47 |
"top_p": 0.2,
|
48 |
"top_k": 3550,
|
49 |
"n_next_tokens": 50
|
|
|
109 |
|
110 |
1. disease2diff2disease - generate signature for disease
|
111 |
2. compound2diff2compound - generate signature for compound
|
112 |
+
3. age_group2diff2age_group - generate signature for age group / predict age group based on signature
|
|
|
113 |
|
114 |
|
115 |
### Other meta-data (`inputs.` in config)
|
|
|
163 |
|
164 |
_Example 2_
|
165 |
|
166 |
+
You want to generate signature for healthy Human, male, 40-50 years, in tissue - whole blood.
|
167 |
```json
|
168 |
{
|
169 |
"inputs": {
|
170 |
+
"instruction": ["disease2diff2disease", "age_group2diff2age_group"],
|
171 |
"tissue": ["whole blood"],
|
172 |
+
"age": "",
|
173 |
"cell": "",
|
174 |
"efo": "",
|
175 |
"datatype": "",
|
176 |
"drug": "",
|
177 |
"dose": "",
|
178 |
"time": "",
|
179 |
+
"case": "40.0-50.0",
|
180 |
"control": "",
|
181 |
"dataset_type": "expression",
|
182 |
"gender": "m",
|
|
|
195 |
|
196 |
```
|
197 |
Note, here we used ```disease2diff2disease``` instruction, but we expected to generate signatures for healthy human, that's why we'd set ```efo``` to empty string "".
|
198 |
+
Alternatively, we can add one more instruction to example 2 - ```"instruction": ["disease2diff2disease", "age_group2diff2age_group"]```
|
199 |
|
200 |
---
|
201 |
|