diff --git a/source/onnx.js b/source/onnx.js
index f859e47acc..681259f298 100644
--- a/source/onnx.js
+++ b/source/onnx.js
@@ -703,8 +703,7 @@ onnx.Attribute = class {
             }
             if (metadata.type === 'DataType') {
                 this._type = metadata.type;
-                const value = this._value ? parseInt(this._value.toString(), 10) : this._value;
-                this._value = Number.isInteger(value) ? context.createDataType(value) : value;
+                this._value = context.createDataType(this._value);
             }
         }
     }
@@ -821,6 +820,9 @@ onnx.Tensor = class {
             this._location = context.createLocation(tensor.data_location);
             if (tensor.data_location === onnx.DataLocation.DEFAULT) {
                 switch (tensor.data_type) {
+                    case onnx.DataType.UNDEFINED: {
+                        break;
+                    }
                     case onnx.DataType.FLOAT16:
                         if (tensor.int32_data && tensor.int32_data.length > 0) {
                             const buffer = new Uint8Array(tensor.int32_data.length << 1);
@@ -1321,7 +1323,7 @@ onnx.GraphContext = class {
     constructor(context, nodes) {
         this._context = context;
         this._dataTypes = new Map(Object.entries(onnx.DataType).map((entry) => [ entry[1], entry[0].toLowerCase() ]));
-        this._dataTypes.set(onnx.DataType.UNDEFINED, 'UNDEFINED');
+        this._dataTypes.set(onnx.DataType.UNDEFINED, 'undefined');
         this._dataTypes.set(onnx.DataType.BOOL, 'boolean');
         this._dataTypes.set(onnx.DataType.FLOAT, 'float32');
         this._dataTypes.set(onnx.DataType.DOUBLE, 'float64');
@@ -1474,7 +1476,21 @@ onnx.GraphContext = class {
     }
 
     createDataType(value) {
-        return this._dataTypes.has(value) ? this._dataTypes.get(value) : this._dataTypes.get(onnx.DataType.UNDEFINED);
+        if (!Number.isInteger(value)) {
+            if (value && value.toNumber) {
+                value = value.toNumber();
+            }
+            else if (value && typeof value === 'string' && onnx.DataType[value.toUpperCase()] !== undefined) {
+                value = onnx.DataType[value.toUpperCase()];
+            }
+            else {
+                throw new onnx.Error("Unsupported data type '" + JSON.stringify(value) + "'.");
+            }
+        }
+        if (this._dataTypes.has(value)) {
+            return this._dataTypes.get(value);
+        }
+        throw new onnx.Error("Unsupported data type '" + JSON.stringify(value) + "'.");
     }
 
     createLocation(value) {
diff --git a/test/models.json b/test/models.json
index 0dd0a46bdb..4a006b4b9e 100644
--- a/test/models.json
+++ b/test/models.json
@@ -3238,14 +3238,6 @@
     "format":   "ONNX v3",
     "link":     "https://github.com/onnx/models/tree/main/models/face_recognition/ArcFace"
   },
-  {
-    "type":     "onnx",
-    "target":   "bidaf-9.onnx.zip",
-    "source":   "https://github.com/lutzroeder/netron/files/6572387/bidaf-9.onnx.zip",
-    "format":   "ONNX v4",
-    "producer": "CNTK 2.7",
-    "link":     "https://github.com/lutzroeder/netron/issues/6"
-  },
   {
     "type":     "onnx",
     "target":   "bert-base-uncased.onnx.zip",
@@ -3263,10 +3255,10 @@
   },
   {
     "type":     "onnx",
-    "target":   "mnist_bfloat16.onnx",
-    "source":   "https://github.com/lutzroeder/netron/files/8556399/mnist_bfloat16.onnx.zip[mnist_bfloat16.onnx]",
+    "target":   "bidaf-9.onnx.zip",
+    "source":   "https://github.com/lutzroeder/netron/files/6572387/bidaf-9.onnx.zip",
     "format":   "ONNX v4",
-    "producer": "pytorch 1.12.0",
+    "producer": "CNTK 2.7",
     "link":     "https://github.com/lutzroeder/netron/issues/6"
   },
   {
@@ -3507,6 +3499,14 @@
     "format":   "ONNX v3",
     "link":     "https://github.com/Microsoft/Windows-Machine-Learning/tree/master/Samples/MNIST/Tutorial/cs/Assets"
   },
+  {
+    "type":     "onnx",
+    "target":   "mnist_bfloat16.onnx",
+    "source":   "https://github.com/lutzroeder/netron/files/8556399/mnist_bfloat16.onnx.zip[mnist_bfloat16.onnx]",
+    "format":   "ONNX v4",
+    "producer": "pytorch 1.12.0",
+    "link":     "https://github.com/lutzroeder/netron/issues/6"
+  },
   {
     "type":     "onnx",
     "target":   "mlnet_encoder.onnx",