新增映射规则文档 (#7183)

* update docs

* update docs

* update P2POp

* fix docs

* fix bugs

* update docs

Author: Xuxuanang

docs/guides/model_convert/convert_from_pytorch/api_difference/distributed/torch.distributed.P2POp.md

@@ -0,0 +1,25 @@
+## [torch 参数更多]torch.distributed.P2POp
+
+### [torch.distributed.P2POp](https://pytorch.org/docs/stable/distributed.html#torch.distributed.P2POp)
+
+```python
+torch.distributed.P2POp(op, tensor, peer, group=None, tag=0)
+```
+
+### [paddle.distributed.P2POp](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/distributed/Overview_cn.html#paddle-distributed)
+
+```python
+paddle.distributed.P2POp(op, tensor, peer, group=None)
+```
+
+PyTorch 相比 Paddle 支持更多其他参数，具体如下：
+
+### 参数映射
+
+| PyTorch | PaddlePaddle    | 备注                                                              |
+| ------- | --------------- | ----------------------------------------------------------------- |
+| op  | op          | 表示操作类型。                                               |
+| tensor  | tensor          | 表示发送或接收的 Tensor。                                               |
+| peer     | peer             | 表示目标进程的 rank。                                                  |
+| group   | group           | 指定通信的进程组。   |
+| tag     | -               | 表示匹配接收标签，Paddle 无此参数，暂无转写方式。   |

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_ceil_.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_ceil_
+
+### [torch.\_foreach_ceil_](https://pytorch.org/docs/stable/generated/torch._foreach_ceil_.html#torch-foreach-ceil)
+
+```python
+torch._foreach_ceil_(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_ceil_(tensors)
+
+# Paddle 写法
+[paddle.assign(paddle.ceil(x), x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_erfc.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_erfc
+
+### [torch.\_foreach_erfc](https://pytorch.org/docs/stable/generated/torch._foreach_erfc.html#torch-foreach-erfc)
+
+```python
+torch._foreach_erfc(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_erfc(tensors)
+
+# Paddle 写法
+tuple([(1-paddle.erf(x)) for x in tensors])
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_erfc_.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_erfc_
+
+### [torch.\_foreach_erfc_](https://pytorch.org/docs/stable/generated/torch._foreach_erfc_.html#torch-foreach-erfc)
+
+```python
+torch._foreach_erfc_(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_erfc_(tensors)
+
+# Paddle 写法
+[paddle.assign(1-paddle.erf(x), x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_exp_.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_exp_
+
+### [torch.\_foreach_exp_](https://pytorch.org/docs/stable/generated/torch._foreach_exp_.html#torch._foreach_exp_)
+
+```python
+torch._foreach_exp_(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_exp_(tensors)
+
+# Paddle 写法
+[paddle.assign(paddle.exp(x), x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_floor_.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_floor_
+
+### [torch.\_foreach_floor_](https://pytorch.org/docs/stable/generated/torch._foreach_floor_.html#torch-foreach-floor)
+
+```python
+torch._foreach_floor_(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_floor_(tensors)
+
+# Paddle 写法
+[paddle.assign(paddle.floor(x), x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_log.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_log
+
+### [torch.\_foreach_log](https://pytorch.org/docs/stable/generated/torch._foreach_log.html#torch-foreach-log)
+
+```python
+torch._foreach_log(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_log(tensors)
+
+# Paddle 写法
+[paddle.log(x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/torch/torch._foreach_log_.md

@@ -0,0 +1,19 @@
+## [组合替代实现]torch.\_foreach_log_
+
+### [torch.\_foreach_log_](https://pytorch.org/docs/stable/generated/torch._foreach_log_.html#torch-foreach-log)
+
+```python
+torch._foreach_log_(self)
+```
+
+Paddle 无此 API，需要组合实现。
+
+### 转写示例
+
+```python
+# PyTorch 写法
+torch._foreach_log_(tensors)
+
+# Paddle 写法
+[paddle.log_(x) for x in tensors]
+```

docs/guides/model_convert/convert_from_pytorch/pytorch_api_mapping_cn.md

@@ -1042,22 +1042,20 @@
 | NOT-IMPLEMENTED-ITEM(`torch.jit.trace`, https://pytorch.org/docs/stable/generated/torch.jit.trace.html#torch-jit-trace, 可新增，但框架底层无相关设计，成本高) |
 | NOT-IMPLEMENTED-ITEM(`torch.jit.unused`, https://pytorch.org/docs/stable/generated/torch.jit.unused.html#torch-jit-unused, 可新增，但框架底层无相关设计，成本高) |
 | NOT-IMPLEMENTED-ITEM(`torch.utils.checkpoint.checkpoint_sequential`, https://pytorch.org/docs/stable/checkpoint.html#torch.utils.checkpoint.checkpoint_sequential, 可新增，但框架底层无相关设计，成本高) |
+| NOT-IMPLEMENTED-ITEM(`torch.nn.parameter.UninitializedBuffer`, https://pytorch.org/docs/stable/generated/torch.nn.parameter.UninitializedBuffer.html#torch.nn.parameter.UninitializedBuffer, 可新增，且框架底层有相关设计，成本低) |
+| NOT-IMPLEMENTED-ITEM(`torch.autograd.Function.jvp`, https://pytorch.org/docs/stable/generated/torch.autograd.Function.jvp.html#torch-autograd-function-jvp, 可新增，且框架底层有相关设计，成本低) |
+| NOT-IMPLEMENTED-ITEM(`torch.memory_format`, https://pytorch.org/docs/stable/tensor_attributes.html#torch.memory_format, 可新增，但框架底层无相关设计，成本高) |
+| NOT-IMPLEMENTED-ITEM(`torch.distributed.is_gloo_available`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.is_gloo_available, 可新增，且框架底层有相关设计，成本低) |
+| NOT-IMPLEMENTED-ITEM(`torch.distributed.get_group_rank`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.get_group_rank, 可新增，且框架底层有相关设计，成本低) |
+| NOT-IMPLEMENTED-ITEM(`torch.distributed.get_global_rank`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.get_process_group_ranks, 可新增，且框架底层有相关设计，成本低) |
+| NOT-IMPLEMENTED-ITEM(`torch.set_deterministic_debug_mode`, https://pytorch.org/docs/stable/generated/torch.set_deterministic_debug_mode.html#torch-set-deterministic-debug-mode, 可新增，但框架底层无相关设计，成本高) |
+| NOT-IMPLEMENTED-ITEM(`torch.get_deterministic_debug_mode`, https://pytorch.org/docs/stable/generated/torch.get_deterministic_debug_mode.html#torch-get-deterministic-debug-mode, 可新增，但框架底层无相关设计，成本高) |
 
 ## <span id="id26">映射关系开发中的 API 列表</span>
 
 | 序号 | Pytorch 最新 release | Paddle develop | 映射关系分类 | 备注 |
 | ----- | ----------- | ----------------- | ----------- | ------- |
 | IN-DEVELOPMENT-PATTERN(`torch.utils.tensorboard.writer.SummaryWriter`, https://pytorch.org/docs/stable/tensorboard.html#torch.utils.tensorboard.writer.SummaryWriter) |
-| IN-DEVELOPMENT-PATTERN(`torch.nn.parameter.UninitializedBuffer`, https://pytorch.org/docs/stable/generated/torch.nn.parameter.UninitializedBuffer.html#torch.nn.parameter.UninitializedBuffer) |
-| IN-DEVELOPMENT-PATTERN(`torch.autograd.Function.jvp`, https://pytorch.org/docs/stable/generated/torch.autograd.Function.jvp.html#torch-autograd-function-jvp) |
-| IN-DEVELOPMENT-PATTERN(`torch.memory_format`, https://pytorch.org/docs/stable/tensor_attributes.html#torch.memory_format) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_ceil_`, https://pytorch.org/docs/stable/generated/torch._foreach_ceil_.html#torch-foreach-ceil) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_erfc`, https://pytorch.org/docs/stable/generated/torch._foreach_erfc.html#torch-foreach-erfc) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_erfc_`, https://pytorch.org/docs/stable/generated/torch._foreach_erfc_.html#torch-foreach-erfc) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_exp_`, https://pytorch.org/docs/stable/generated/torch._foreach_exp_.html#torch-foreach-exp) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_floor_`, https://pytorch.org/docs/stable/generated/torch._foreach_floor_.html#torch-foreach-floor) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_log`, https://pytorch.org/docs/stable/generated/torch._foreach_log.html#torch-foreach-log) |
-| IN-DEVELOPMENT-PATTERN(`torch._foreach_log_`, https://pytorch.org/docs/stable/generated/torch._foreach_log_.html#torch-foreach-log) |
 | IN-DEVELOPMENT-PATTERN(`torch._foreach_log10`, https://pytorch.org/docs/stable/generated/torch._foreach_log10.html#torch-foreach-log10) |
 | IN-DEVELOPMENT-PATTERN(`torch._foreach_log10_`, https://pytorch.org/docs/stable/generated/torch._foreach_log10_.html#torch-foreach-log10) |
 | IN-DEVELOPMENT-PATTERN(`torch._foreach_log1p`, https://pytorch.org/docs/stable/generated/torch._foreach_log1p.html#torch-foreach-log1p) |
@@ -1094,10 +1092,6 @@
 | IN-DEVELOPMENT-PATTERN(`torch.autograd.graph.Node.register_prehook`, https://pytorch.org/docs/stable/generated/torch.autograd.graph.Node.register_prehook.html#torch-autograd-graph-node-register-prehook) |
 | IN-DEVELOPMENT-PATTERN(`torch.cuda.OutOfMemoryError`, https://pytorch.org/docs/stable/generated/torch.cuda.OutOfMemoryError.html#torch-cuda-outofmemoryerror) |
 | IN-DEVELOPMENT-PATTERN(`torch.backends.cpu.get_cpu_capability`, https://pytorch.org/docs/stable/backends.html#torch.backends.cpu.get_cpu_capability) |
-| IN-DEVELOPMENT-PATTERN(`torch.distributed.P2POp`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.P2POp) |
-| IN-DEVELOPMENT-PATTERN(`torch.distributed.is_gloo_available`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.is_gloo_available) |
-| IN-DEVELOPMENT-PATTERN(`torch.distributed.get_group_rank`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.get_group_rank) |
-| IN-DEVELOPMENT-PATTERN(`torch.distributed.get_global_rank`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.get_global_rank) |
 | IN-DEVELOPMENT-PATTERN(`torch.distributed.get_process_group_ranks`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.get_process_group_ranks) |
 | IN-DEVELOPMENT-PATTERN(`torch.distributed.batch_isend_irecv`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.batch_isend_irecv) |
 | IN-DEVELOPMENT-PATTERN(`torch.distributed.all_gather_into_tensor`, https://pytorch.org/docs/stable/distributed.html#torch.distributed.all_gather_into_tensor) |
@@ -1119,5 +1113,3 @@
 | IN-DEVELOPMENT-PATTERN(`torch.distributions.continuous_bernoulli.ContinuousBernoulli`, https://pytorch.org/docs/stable/distributions.html#torch.distributions.continuous_bernoulli.ContinuousBernoulli) |
 | IN-DEVELOPMENT-PATTERN(`torch.distributions.exponential.Exponential`, https://pytorch.org/docs/stable/distributions.html#torch.distributions.exponential.Exponential) |
 | IN-DEVELOPMENT-PATTERN(`torch.cuda.StreamContext`, https://pytorch.org/docs/stable/generated/torch.cuda.StreamContext.html#torch.cuda.StreamContext) |
-| IN-DEVELOPMENT-PATTERN(`torch.set_deterministic_debug_mode`, https://pytorch.org/docs/stable/generated/torch.set_deterministic_debug_mode.html#torch-set-deterministic-debug-mode) |
-| IN-DEVELOPMENT-PATTERN(`torch.get_deterministic_debug_mode`, https://pytorch.org/docs/stable/generated/torch.get_deterministic_debug_mode.html#torch-get-deterministic-debug-mode) |