remove more adjoint plugin references

Author: yaugenst-flex

docs/install.rst

@@ -205,7 +205,6 @@ Tidy3D also offers plugins that require additional dependencies:
 
 - ``design``: Design space exploration and optimization.
 - ``pytorch``: A PyTorch wrapper for objective functions defined using autograd.
-- ``adjoint``: Adjoint optimization using JAX (deprecated in favor of native autograd support in Tidy3D)
 
 For example, to install the design plugin dependencies:
 

tidy3d/plugins/autograd/README.md

@@ -11,10 +11,9 @@ This functionality was previously available using the `adjoint` plugin, which us
 3. Because we inherited these classes from their `tidy3d` components, for technical reasons, we needed to separate the `jax`-traced fields from the regular fields.
    For example, `JaxSimulation.input_structures` and `.output_monitors` were needed.
 
-All of these limitations (among others) motivated us to come up with a new approach to automatic differentiation, which was introduced as an experimental feature in `2.7`.
-The `adjoint` plugin will continue to be supported indefinitely, but no new features will be developed for it.
-We also believe the new approach offers a far better user experience, so we encourage users to switch whenever is convenient.
-This guide will give some instructions on how to do so.
+All of these limitations (among others) motivated us to come up with a new approach to automatic differentiation, which was introduced as an experimental feature in `2.7` and is now the default.
+The previous JAX-based `adjoint` plugin has been removed in favor of this built-in workflow.
+This guide will give some instructions on how to migrate existing code.
 
 ## New implementation using `autograd`
 
@@ -89,18 +88,12 @@ Additionally, `autograd` has a `grad_with_aux` function that can be used to comp
 
 Otherwise, `jax` and `autograd` are very similar to each other in practice.
 
-### Migrating from `adjoint` plugin
+### Migrating from the deprecated `adjoint` plugin
 
 Converting code from the `adjoint` plugin to the native autograd support is straightforward.
 
-Instead of importing classes from the `tda` namespace, with names like `tda.Jax_`, we can just use regular `td.` classes.
-
-```py
-import tidy3d.plugins.adjoint as tda
-tda.JaxStructure(...)
-```
-
-becomes
+Instead of importing classes from the old `tidy3d.plugins.adjoint` namespace (for example `tda.JaxStructure`),
+you can use the regular `tidy3d` classes directly:
 
 ```py
 import tidy3d as td

tidy3d/plugins/invdes/README.md

@@ -4,7 +4,7 @@
 
 This describes the "Inverse Design" (`invdes`) plugin of Tidy3D.
 
-The goal of `invdes` is to provide a simpler interface for setting up most practical inverse design problems. It wraps the lower-level `adjoint` plugin of Tidy3D to perform the gradient calculations, but allows the user to focus on the important aspects of their design without getting into the details of `autograd`.
+The goal of `invdes` is to provide a simpler interface for setting up most practical inverse design problems. It leverages Tidy3D's native autograd-based adjoint infrastructure to perform the gradient calculations, while allowing the user to focus on the important aspects of their design without getting into low-level workflow details.
 
 In this notebook, we'll give a simple demo showing the inverse design of a 1 -> 3 splitter using the inverse design plugin.
 
@@ -97,7 +97,7 @@ Next we want to define the base `td.Simulation` that contains the static portion
 
 For this, we will make a bunch of regular `tidy3d` components (excluding the design region) and put them into a `td.Simulation`.
 
-> Note: we don't need to use `adjoint` components, such as `td.Simulation` in the `invdes` plugin. These components are created behind the scenes by the higher level wrappers that we define here.
+> Note: there is no need to use any special autograd-specific component classes in user code; the plugin builds the differentiable structures internally from the standard `td.*` objects defined here.
 
 ```py
 
@@ -285,7 +285,7 @@ design = tdi.InverseDesign(
     output_monitor_names=[mnt.name for mnt in monitors_out],
 )
 ```
-> Note: the `output_monitor_names` field is used to specify which monitors to use in the objective function. If they are not supplied, `invdes` will automatically include all compatible monitors, but at times these can raise warnings in the `adjoint` plugin, so it can be nice to specify.
+> Note: the `output_monitor_names` field is used to specify which monitors to use in the objective function. If they are not supplied, `invdes` will automatically include all compatible monitors, but at times these can raise warnings if the differentiable workflow cannot trace certain monitor types, so it can be nice to specify.
 
 The `InverseDesign` object can be exported to a `td.Simulation` given some parameters using the `to_simulation(params)` method.
 

tidy3d/plugins/invdes/design.py

@@ -128,10 +128,9 @@ class InverseDesign(AbstractInverseDesign):
         None,
         title="Output Monitor Names",
         description="Optional names of monitors whose data the differentiable output depends on."
-        "If this field is left ``None``, the plugin will try to add all compatible monitors to "
-        "``JaxSimulation.output_monitors``. While this will work, there may be warnings if the "
-        "monitors are not compatible with the ``adjoint`` plugin, for example if there are "
-        "``FieldMonitor`` instances with ``.colocate != False``.",
+        "If this field is left ``None``, the plugin will try to add all compatible monitors "
+        "automatically. While this will work, there may be warnings if certain monitor types are "
+        "not fully supported, for example ``FieldMonitor`` instances with ``.colocate != False``.",
     )
 
     _check_sim_pixel_size = check_pixel_size("simulation")
@@ -270,10 +269,9 @@ class InverseDesignMulti(AbstractInverseDesign):
         None,
         title="Output Monitor Names",
         description="Optional names of monitors whose data the differentiable output depends on."
-        "If this field is left ``None``, the plugin will try to add all compatible monitors to "
-        "``JaxSimulation.output_monitors``. While this will work, there may be warnings if the "
-        "monitors are not compatible with the ``adjoint`` plugin, for example if there are "
-        "``FieldMonitor`` instances with ``.colocate != False``.",
+        "If this field is left ``None``, the plugin will try to add all compatible monitors "
+        "automatically. While this will work, there may be warnings if certain monitor types are "
+        "not fully supported, for example ``FieldMonitor`` instances with ``.colocate != False``.",
     )
 
     _check_sim_pixel_size = check_pixel_size("simulations")

tidy3d/web/core/task_core.py

@@ -9,7 +9,6 @@
 from datetime import datetime
 from typing import Callable, Optional, Union
 
-import pydantic.v1 as pd
 from botocore.exceptions import ClientError
 from pydantic.v1 import Extra, Field, parse_obj_as
 
@@ -200,20 +199,6 @@ class SimulationTask(ResourceLifecycle, Submittable, extra=Extra.allow):
     #     description="List of parent task ids for the simulation, used internally only."
     # )
 
-    @pd.root_validator(pre=True)
-    def _error_if_jax_sim(cls, values):
-        """Raise error if user tries to submit simulation that's a JaxSimulation."""
-        sim = values.get("simulation")
-        if sim is None:
-            return values
-        if "JaxSimulation" in str(type(sim)):
-            raise ValueError(
-                "'JaxSimulation' not compatible with regular webapi functions. "
-                "Either convert it to Simulation with 'jax_sim.to_simulation()[0]' or use "
-                "the 'adjoint.run' function to run JaxSimulations."
-            )
-        return values
-
     @classmethod
     def create(
         cls,