Update README.md
Browse files
README.md
CHANGED
@@ -94,11 +94,13 @@ E.g: To predict the Tc value Nd2Fe14B1 magnet composition, the features are Nd=2
|
|
94 |
|
95 |
**Application & Limitations**
|
96 |
|
|
|
|
|
97 |
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.
|
98 |
|
99 |
**Model pipeline**
|
100 |
|
101 |
-
The voting regressor to predict the Tc combines the following four base models and equal weight is assigned to each base
|
102 |
|
103 |
1. Extra tree regressor (ET)
|
104 |
2. Extreme gradient boosting (XGB)
|
@@ -117,9 +119,8 @@ Tc_predictor = load('Magnet_Tc_predictor.joblib') # trained model
|
|
117 |
config = json.load(open('config.json')) # config file
|
118 |
features = config['features'] # feature extraction
|
119 |
|
120 |
-
#data = pd.read_excel("data.xlsx") # read test file with new compositions
|
121 |
data = data[features]
|
122 |
-
#data.columns = ["feat_" + str(col) for col in data.columns]
|
123 |
|
124 |
Predicted_value = Tc_predictor.predict(data) # predict Tc values
|
125 |
print("Predicted Tc value is: {0:.2f}'.format(predictions)")
|
|
|
94 |
|
95 |
**Application & Limitations**
|
96 |
|
97 |
+
Input feature as the chemical composition of the test sample should match the sequence of the features described in the config file.
|
98 |
+
|
99 |
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.
|
100 |
|
101 |
**Model pipeline**
|
102 |
|
103 |
+
The voting regressor to predict the Tc combines the following four base models and equal weight is assigned to each base model.
|
104 |
|
105 |
1. Extra tree regressor (ET)
|
106 |
2. Extreme gradient boosting (XGB)
|
|
|
119 |
config = json.load(open('config.json')) # config file
|
120 |
features = config['features'] # feature extraction
|
121 |
|
122 |
+
#data = pd.read_excel("data.xlsx") # read test file with new compositions
|
123 |
data = data[features]
|
|
|
124 |
|
125 |
Predicted_value = Tc_predictor.predict(data) # predict Tc values
|
126 |
print("Predicted Tc value is: {0:.2f}'.format(predictions)")
|