Graph-level-operation elements are placed on the left-top of the page. Currently, there are four buttons: `Preview`, `Reset`, `Download` and `Add node`. They can do:
Graph-level-operation elements are placed on the left-top of the page. Currently, there are four buttons: `Preview`, `Reset`, `Download` and `Add node`. They can do:
- `Refresh`: Refresh the model graph to preview editing effects.
- `Refresh`: Refresh the model graph to preview editing effects.
> In this version, the model graph is refreshed automatically as soon as an editting operation is invoked. So this button can be used much fewer than ealier versions.
> In this version, the model graph is refreshed automatically as soon as an editing operation is invoked. So this button can be used much fewer than earlier versions.
- `Reset`: Reset the whole model graph to its initial state;
- `Reset`: Reset the whole model graph to its initial state;
- `Download`: Save the modified model into disk.
- `Download`: Save the modified model into disk.
- `Add node`: Add a new node into the model.
- `Add node`: Add a new node into the model.
@ -69,17 +61,15 @@ Node-level-operation elements are all in the sidebar, which can be invoked by cl
Let's take a closer look.
Let's take a closer look.
## Delete/recover nodes
## Delete/recover nodes
There are two modes for deleting node: `Delete With Children` and `Delete Single Node`. `Delete Single Node` only deletes the clicked node, while `Delete With Children` also deletes all the node rooted on the clicked node, which is convenient and natural if we want to delete a long path of nodes.
There are two modes for deleting node: `Delete With Children` and `Delete Single Node`. `Delete Single Node` only deletes the clicked node, while `Delete With Children` also deletes all the node rooted on the clicked node, which is convenient and natural if we want to delete a long path of nodes.
> The implementation of `Delete With Children` is based on the backtracking algorithm.
> The implementation of `Delete With Children` is based on the backtracking algorithm.
For previewing, The deleted nodes are in grey mode at first. If a node is deleted by mistake, `Recover Node` button can help us recover it back to graph. Click `Enter` button to take the deleting operation into effect, then the updated graph will show on the page automatically.
For previewing, The deleted nodes are in grey mode at first. If a node is deleted by mistake, `Recover Node` button can help us recover it back to graph. Click `Enter` button to take the deleting operation into effect, then the updated graph will show on the page automatically.
The following figure shows a typical deleting process.
The following figure shows a typical deleting process:
@ -95,13 +85,15 @@ For example, Now we want remove the preprocess operators (`Sub->Mul->Sub->Trans
> Note: To link node $A$ (`data_0` in the above example) to node $B$ (the 1st `Conv` in the above example), **it is suggested to edit the input of node $B$ to the output of node `A`, rather than edit the output of node $A$ to the input of node `B`.** Because the input of $B$ can also be other node's output and unexpected result will happen.
> Note: To link node $A$ (`data_0` in the above example) to node $B$ (the 1st `Conv` in the above example), **it is suggested to edit the input of node $B$ to the output of node `A`, rather than edit the output of node $A$ to the input of node `B`.** Because the input of $B$ can also be other node's output and unexpected result will happen.
Sometimes we want to add new nodes into the exsited model. `onnx-modifier` supports this feature experimentally now.
Sometimes we want to add new nodes into the exsited model. `onnx-modifier` supports this feature experimentally now.
@ -112,9 +104,11 @@ Note there is an `Add node` button, following with a selector elements on the to
2. Click the new node and edit it in the invoked siderbar. What we need to fill are the node Attributes (`undefined` by default) and its Inputs/Outputs (which decide where the node will be inserted in the graph).
2. Click the new node and edit it in the invoked siderbar. What we need to fill are the node Attributes (`undefined` by default) and its Inputs/Outputs (which decide where the node will be inserted in the graph).
1. :warning: Currently, adding nodes with initializer (such as weight parameters) are not supported (such as `Conv`, `BatchNormalization`). Adding nodes without initializer are tested and work as expected in my tested case (such as `Flatten`, `ArgMax`, `Concat`).
1. :warning: Currently, adding nodes with initializer (such as weight parameters) is not supported (such as `Conv`, `BatchNormalization`). Adding nodes without initializer are tested and work as expected in my tested case (such as `Flatten`, `ArgMax`, `Concat`).
2. Click the selector and type the first letter for the new node type (`f` for `Flatten` node for example), we can be quickly navigated to the node type.
2. Click the selector and type the first letter for the new node type (`f` for `Flatten` node for example), we can be quickly navigated to the node type.
ZhangGe6
4 years ago
