remove XGBoostObjectiveJSON (#2508)

Author: sneaxiy

pkg/attribute/attribute.go

@@ -372,9 +372,6 @@ var extractSymbolOnce sync.Once
 // OptimizerParamsDocs stores parameters and documents of optimizers
 var OptimizerParamsDocs map[string]map[string]string
 
-// XGBoostObjectiveDocs stores options for xgboost objective
-var XGBoostObjectiveDocs map[string]string
-
 // ExtractSymbol extracts parameter documents of Python modules from doc strings
 func ExtractSymbol(module ...string) {
 	cmd := exec.Command("python", "-uc", fmt.Sprintf("__import__('symbol_extractor').print_param_doc('%s')", strings.Join(module, "', '")))
@@ -409,8 +406,5 @@ func init() {
 	if err := json.Unmarshal([]byte(OptimizerParameterJSON), &OptimizerParamsDocs); err != nil {
 		panic(err) // assertion
 	}
-	if err := json.Unmarshal([]byte(XGBoostObjectiveJSON), &XGBoostObjectiveDocs); err != nil {
-		panic(err)
-	}
 	removeUnnecessaryParams()
 }

pkg/attribute/xgboost_objective_params.go

@@ -13,23 +13,21 @@
 
 package attribute
 
-// XGBoostObjectiveJSON is xgboost objective param json extracted
+// XGBoostObjectiveDocs is xgboost objective parameter docs extracted
 // from https://xgboost.readthedocs.io/en/latest/parameter.html
-const XGBoostObjectiveJSON = `
-{
-    "binary:hinge": "hinge loss for binary classification. This makes predictions of 0 or 1, rather than producing probabilities.",
-    "binary:logistic": "logistic regression for binary classification, output probability",
-    "binary:logitraw": "logistic regression for binary classification, output score before logistic transformation",
-    "multi:softmax": "set XGBoost to do multiclass classification using the softmax objective, you also need to set num_class(number of classes)",
-    "multi:softprob": "same as softmax, but output a vector of ndata * nclass, which can be further reshaped to ndata * nclass matrix. The result contains predicted probability of each data point belonging to each class.",
-    "rank:map": "Use LambdaMART to perform list-wise ranking where Mean Average Precision (MAP) is maximized",
-    "rank:ndcg": "Use LambdaMART to perform list-wise ranking where Normalized Discounted Cumulative Gain (NDCG) is maximized",
-    "rank:pairwise": "Use LambdaMART to perform pairwise ranking where the pairwise loss is minimized",
-    "reg:gamma": "gamma regression with log-link. Output is a mean of gamma distribution. It might be useful, e.g., for modeling insurance claims severity, or for any outcome that might be gamma-distributed.",
-    "reg:logistic": "logistic regression",
-    "reg:squarederror": "regression with squared loss.",
-    "reg:squaredlogerror": "regression with squared log loss 1/2[log(pred+1)\u2212log(label+1)]^2",
-    "reg:tweedie": "Tweedie regression with log-link. It might be useful, e.g., for modeling total loss in insurance, or for any outcome that might be Tweedie-distributed.",
-    "survival:cox": "Cox regression for right censored survival time data (negative values are considered right censored). Note that predictions are returned on the hazard ratio scale (i.e., as HR = exp(marginal_prediction) in the proportional hazard function h(t) = h0(t) * HR)."
+var XGBoostObjectiveDocs = map[string]string{
+	"binary:hinge":        "hinge loss for binary classification. This makes predictions of 0 or 1, rather than producing probabilities.",
+	"binary:logistic":     "logistic regression for binary classification, output probability",
+	"binary:logitraw":     "logistic regression for binary classification, output score before logistic transformation",
+	"multi:softmax":       "set XGBoost to do multiclass classification using the softmax objective, you also need to set num_class(number of classes)",
+	"multi:softprob":      "same as softmax, but output a vector of ndata * nclass, which can be further reshaped to ndata * nclass matrix. The result contains predicted probability of each data point belonging to each class.",
+	"rank:map":            "Use LambdaMART to perform list-wise ranking where Mean Average Precision (MAP) is maximized",
+	"rank:ndcg":           "Use LambdaMART to perform list-wise ranking where Normalized Discounted Cumulative Gain (NDCG) is maximized",
+	"rank:pairwise":       "Use LambdaMART to perform pairwise ranking where the pairwise loss is minimized",
+	"reg:gamma":           "gamma regression with log-link. Output is a mean of gamma distribution. It might be useful, e.g., for modeling insurance claims severity, or for any outcome that might be gamma-distributed.",
+	"reg:logistic":        "logistic regression",
+	"reg:squarederror":    "regression with squared loss.",
+	"reg:squaredlogerror": "regression with squared log loss 1/2[log(pred+1)\u2212log(label+1)]^2",
+	"reg:tweedie":         "Tweedie regression with log-link. It might be useful, e.g., for modeling total loss in insurance, or for any outcome that might be Tweedie-distributed.",
+	"survival:cox":        "Cox regression for right censored survival time data (negative values are considered right censored). Note that predictions are returned on the hazard ratio scale (i.e., as HR = exp(marginal_prediction) in the proportional hazard function h(t) = h0(t) * HR).",
 }
-`