Add base limitation

docs/toolchain/appendix/operators.md

@@ -38,7 +38,7 @@ Table below shows the list of operators supports base on ONNX operators.
 | LeakyRelu                     |  Y  |  Y  |  Y  |  Y  |  Y  |
 | LpNormalization               |  N  |  N  |  N  |  N  |  N  |
 | LRN                           |  N  |  N  |  N  |  N  |  N  |
-| LSTM                          |  N  |  Y  |  Y  |  Y  |  Y  |
+| LSTM                          |  N  |  N  |  N  |  N  |  Y  |
 | MatMul                        |  N  |  Y  |  Y  |  Y  |  Y  |
 | MaxPool<sup>8</sup>           |  Y  |  Y  |  Y  |  Y  |  Y  |
 | MaxRoiPool                    |  N  |  Y  |  Y  |  Y  |  Y  |

docs/toolchain/manual_3_onnx.md

@@ -6,7 +6,7 @@ Our toolchain utilities take ONNX files as inputs. The ONNX workflow is mainly a
 
 The conversion tools are different for different onnx versions. The environment is managed by miniconda. Current default environment is `onnx1.13`. It supports ONNX opset up to 18.
 
-There is also a `base` environment which is the default environment for the toolchain before v0.25.0. This `base` environment supports opset 8 to 12. It is not recommended to use this environment for new models. But if you have a model using those opsets, you can still use this environment. The following documents are all based on the `onnx1.13` environment. If you are using the `base` environment, please check the [ONNX Converters](appendix/converters.md) for the API usage.
+There is also a `base` environment which is the default environment for the toolchain before v0.25.0. This `base` environment supports opset 8 to 12. It is not recommended to use this environment for new models. But if you have a model using those opsets, you can still use this environment. The following documents are all based on the `onnx1.13` environment. If you are using the `base` environment, please check the [ONNX Converters](appendix/converters.md) for the API usage. **`base` environment DOES NOT support 730 toolchain.**
 
 ### 3.1.1. Convert models to ONNX