diff --git a/src/operator/contrib/bilinear_resize-inl.h b/src/operator/contrib/bilinear_resize-inl.h
index 0db9494748a0..6032c881ed46 100644
--- a/src/operator/contrib/bilinear_resize-inl.h
+++ b/src/operator/contrib/bilinear_resize-inl.h
@@ -61,10 +61,10 @@ struct BilinearSampleParam : public dmlc::Parameter<BilinearSampleParam> {
   int mode;
   bool align_corners;
   DMLC_DECLARE_PARAMETER(BilinearSampleParam) {
-    DMLC_DECLARE_FIELD(height).set_default(1).set_range(1, 10000)
+    DMLC_DECLARE_FIELD(height).set_default(1).set_lower_bound(1)
     .describe("output height (required, but ignored if scale_height is defined or mode is not "
               "\"size\")");
-    DMLC_DECLARE_FIELD(width).set_default(1).set_range(1, 10000)
+    DMLC_DECLARE_FIELD(width).set_default(1).set_lower_bound(1)
     .describe("output width (required, but ignored if scale_width is defined or mode is not "
               "\"size\")");
     DMLC_DECLARE_FIELD(scale_height).set_default(dmlc::optional<float>())
diff --git a/src/operator/contrib/deformable_convolution-inl.h b/src/operator/contrib/deformable_convolution-inl.h
index ca453e3da8aa..5e0e86bb3a54 100644
--- a/src/operator/contrib/deformable_convolution-inl.h
+++ b/src/operator/contrib/deformable_convolution-inl.h
@@ -75,13 +75,13 @@ struct DeformableConvolutionParam : public dmlc::Parameter<DeformableConvolution
       .describe("Convolution dilate: (h, w) or (d, h, w). Defaults to 1 for each dimension.");
     DMLC_DECLARE_FIELD(pad).set_default(mxnet::TShape(0, -1))
       .describe("Zero pad for convolution: (h, w) or (d, h, w). Defaults to no padding.");
-    DMLC_DECLARE_FIELD(num_filter).set_range(1, 100000)
+    DMLC_DECLARE_FIELD(num_filter).set_lower_bound(1)
       .describe("Convolution filter(channel) number");
     DMLC_DECLARE_FIELD(num_group).set_default(1)
       .describe("Number of group partitions.");
     DMLC_DECLARE_FIELD(num_deformable_group).set_default(1)
       .describe("Number of deformable group partitions.");
-    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_lower_bound(0)
       .describe("Maximum temperal workspace allowed for convolution (MB).");
     DMLC_DECLARE_FIELD(no_bias).set_default(false)
       .describe("Whether to disable bias parameter.");
diff --git a/src/operator/contrib/modulated_deformable_convolution-inl.h b/src/operator/contrib/modulated_deformable_convolution-inl.h
index ace11b04834b..6ec5d8861e01 100644
--- a/src/operator/contrib/modulated_deformable_convolution-inl.h
+++ b/src/operator/contrib/modulated_deformable_convolution-inl.h
@@ -77,13 +77,13 @@ struct ModulatedDeformableConvolutionParam
       .describe("Convolution dilate: (h, w) or (d, h, w). Defaults to 1 for each dimension.");
     DMLC_DECLARE_FIELD(pad).set_default(mxnet::TShape(0, -1))
       .describe("Zero pad for convolution: (h, w) or (d, h, w). Defaults to no padding.");
-    DMLC_DECLARE_FIELD(num_filter).set_range(1, 100000)
+    DMLC_DECLARE_FIELD(num_filter).set_lower_bound(1)
       .describe("Convolution filter(channel) number");
     DMLC_DECLARE_FIELD(num_group).set_default(1)
       .describe("Number of group partitions.");
     DMLC_DECLARE_FIELD(num_deformable_group).set_default(1)
       .describe("Number of deformable group partitions.");
-    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_lower_bound(0)
       .describe("Maximum temperal workspace allowed for convolution (MB).");
     DMLC_DECLARE_FIELD(no_bias).set_default(false)
       .describe("Whether to disable bias parameter.");
diff --git a/src/operator/convolution_v1-inl.h b/src/operator/convolution_v1-inl.h
index 9b6ab96ae2ac..0b9981737634 100644
--- a/src/operator/convolution_v1-inl.h
+++ b/src/operator/convolution_v1-inl.h
@@ -70,12 +70,12 @@ struct ConvolutionV1Param : public dmlc::Parameter<ConvolutionV1Param> {
     .describe("convolution dilate: (h, w) or (d, h, w)");
     DMLC_DECLARE_FIELD(pad).set_default(mxnet::TShape(0, 0))
     .describe("pad for convolution: (h, w) or (d, h, w)");
-    DMLC_DECLARE_FIELD(num_filter).set_range(1, 100000)
+    DMLC_DECLARE_FIELD(num_filter).set_lower_bound(1)
     .describe("convolution filter(channel) number");
     DMLC_DECLARE_FIELD(num_group).set_default(1)
     .describe("Number of group partitions. Equivalent to slicing input into num_group\n    "
               "partitions, apply convolution on each, then concatenate the results");
-    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_lower_bound(0)
     .describe("Maximum temporary workspace allowed for convolution (MB)."
               "This parameter determines the effective batch size of the convolution "
               "kernel, which may be smaller than the given batch size. "
diff --git a/src/operator/nn/convolution-inl.h b/src/operator/nn/convolution-inl.h
index 7d5f7c7d5757..87c82c370d1e 100644
--- a/src/operator/nn/convolution-inl.h
+++ b/src/operator/nn/convolution-inl.h
@@ -75,11 +75,11 @@ struct ConvolutionParam : public dmlc::Parameter<ConvolutionParam> {
     .describe("Convolution dilate: (w,), (h, w) or (d, h, w). Defaults to 1 for each dimension.");
     DMLC_DECLARE_FIELD(pad).set_default(mxnet::TShape(0, 0))
     .describe("Zero pad for convolution: (w,), (h, w) or (d, h, w). Defaults to no padding.");
-    DMLC_DECLARE_FIELD(num_filter).set_range(1, 100000)
+    DMLC_DECLARE_FIELD(num_filter).set_lower_bound(1)
     .describe("Convolution filter(channel) number");
     DMLC_DECLARE_FIELD(num_group).set_default(1)
     .describe("Number of group partitions.");
-    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(1024).set_lower_bound(0)
     .describe("Maximum temporary workspace allowed (MB) in convolution."
               "This parameter has two usages. When CUDNN is not used, it determines the "
               "effective batch size of the convolution kernel. When CUDNN is used, it controls "
diff --git a/src/operator/nn/deconvolution-inl.h b/src/operator/nn/deconvolution-inl.h
index 58f9be702396..0aac39a2202b 100644
--- a/src/operator/nn/deconvolution-inl.h
+++ b/src/operator/nn/deconvolution-inl.h
@@ -85,11 +85,11 @@ struct DeconvolutionParam : public dmlc::Parameter<DeconvolutionParam> {
                   "`adj` will be ignored and computed accordingly.");
     DMLC_DECLARE_FIELD(target_shape).set_default(mxnet::TShape(0, 0))
         .describe("Shape of the output tensor: (w,), (h, w) or (d, h, w).");
-    DMLC_DECLARE_FIELD(num_filter).set_range(1, 100000)
+    DMLC_DECLARE_FIELD(num_filter).set_lower_bound(1)
         .describe("Number of output filters.");
     DMLC_DECLARE_FIELD(num_group).set_default(1)
         .describe("Number of groups partition.");
-    DMLC_DECLARE_FIELD(workspace).set_default(512).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(512).set_lower_bound(0)
         .describe("Maximum temporary workspace allowed (MB) in deconvolution."
                   "This parameter has two usages. When CUDNN is not used, it determines the "
                   "effective batch size of the deconvolution kernel. When CUDNN is used, "
diff --git a/src/operator/nn/upsampling-inl.h b/src/operator/nn/upsampling-inl.h
index 8219e3e9bd8d..daf74b71e410 100644
--- a/src/operator/nn/upsampling-inl.h
+++ b/src/operator/nn/upsampling-inl.h
@@ -56,7 +56,7 @@ struct UpSamplingParam : public dmlc::Parameter<UpSamplingParam> {
   uint64_t workspace;
   DMLC_DECLARE_PARAMETER(UpSamplingParam) {
     DMLC_DECLARE_FIELD(scale)
-    .set_range(1, 1000)
+    .set_lower_bound(1)
     .describe("Up sampling scale");
     DMLC_DECLARE_FIELD(num_filter)
     .describe("Input filter. Only used by bilinear sample_type."
@@ -79,7 +79,7 @@ struct UpSamplingParam : public dmlc::Parameter<UpSamplingParam> {
     "upsampling, this can be 1-N; the size of output will be"
     "(scale*h_0,scale*w_0) and all other inputs will be upsampled to the"
     "same size. For bilinear upsampling this must be 2; 1 input and 1 weight.");
-    DMLC_DECLARE_FIELD(workspace).set_default(512).set_range(0, 8192)
+    DMLC_DECLARE_FIELD(workspace).set_default(512).set_lower_bound(0)
     .describe("Tmp workspace for deconvolution (MB)");
   }
 };  // struct UpSamplingParam