Update README.md
Browse files
README.md
CHANGED
@@ -9,17 +9,35 @@ tags:
|
|
9 |
**Model Description**
|
10 |
|
11 |
The magnet Curie temperature (Tc [K]) predictor model has been trained using a supervised learning approach on a specific set of magnet classes having 14:2:1 phases.
|
|
|
|
|
|
|
12 |
It predicts the Tc value using the chemical composition as a feature.
|
13 |
-
E.g: To predict the Tc value Nd2Fe14B1 magnet composition, the features are Nd=2, Fe=14, and B=1.
|
14 |
|
|
|
15 |
|
16 |
**Application & Limitations**
|
17 |
|
18 |
-
The trained model is valid for 14:2:1 phases only, which are stoichiometric compositions.
|
19 |
|
20 |
**Model Plot**
|
21 |
|
22 |
**How to use the trained model for inference**
|
23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
|
|
|
|
25 |
|
|
|
|
9 |
**Model Description**
|
10 |
|
11 |
The magnet Curie temperature (Tc [K]) predictor model has been trained using a supervised learning approach on a specific set of magnet classes having 14:2:1 phases.
|
12 |
+
The dataset to train the Tc prediction model is a distinct literature source.
|
13 |
+
Further, the Tc values for various 14:2:1 magnet phases at room temperature are considered for dataset creation.
|
14 |
+
|
15 |
It predicts the Tc value using the chemical composition as a feature.
|
|
|
16 |
|
17 |
+
E.g: To predict the Tc value Nd2Fe14B1 magnet composition, the features are Nd=2, Fe=14, and B=1.
|
18 |
|
19 |
**Application & Limitations**
|
20 |
|
21 |
+
The trained model is valid for 14:2:1 phases only, which are stoichiometric compositions and the predicted Tc value is in Kelvin and at room temperature.
|
22 |
|
23 |
**Model Plot**
|
24 |
|
25 |
**How to use the trained model for inference**
|
26 |
|
27 |
+
```python
|
28 |
+
import json
|
29 |
+
from joblib
|
30 |
+
import pandas as pd
|
31 |
+
|
32 |
+
Tc_predictor = load('Magnet_Tc_predictor.joblib') # trained model
|
33 |
+
config = json.load(open('config.json')) # config file
|
34 |
+
features = config['features'] # feature extraction
|
35 |
+
|
36 |
+
#data = pd.read_excel("data.xlsx") # read test file with new compositions
|
37 |
+
data = data[features]
|
38 |
+
#data.columns = ["feat_" + str(col) for col in data.columns]
|
39 |
|
40 |
+
Predicted_value = Tc_predictor.predict(data) # predict Tc values
|
41 |
+
print("Predicted Tc value is: {0:.2f}'.format(predictions)")
|
42 |
|
43 |
+
```
|