[Docs] 增加第二类api difference

docs/guides/model_convert/convert_from_pytorch/api_difference/args_name_diff/torch.nn.functional.glu.md

@@ -1,12 +1,12 @@
 ## [ 仅参数名不一致 ]torch.nn.functional.glu
 ### [torch.nn.functional.glu](https://pytorch.org/docs/stable/generated/torch.nn.functional.glu.html?highlight=glu#torch.nn.functional.glu)
 ```python
-torch.nn.functional.glu(input, dim=- 1)
+torch.nn.functional.glu(input, dim=-1)
 ```
 
 ### [paddle.nn.functional.glu](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/nn/functional/glu_cn.html)
 ```python
-paddle.nn.functional.glu(x, axis=- 1, name=None)
+paddle.nn.functional.glu(x, axis=-1, name=None)
 ```
 
 两者功能一致，仅参数名不一致，具体如下：

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.bernoulli.md

@@ -9,7 +9,7 @@ torch.Tensor.bernoulli(generator=None)
 ### [paddle.bernoulli](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/bernoulli_cn.html#paddle/bernoulli_cn#cn-api-paddle-bernoulli)
 
 ```python
-paddle.bernoulli(x, name=None)
+paddle.bernoulli(x, p=None, name=None)
 ```
 
 两者功能一致，但调用方式不一致，具体如下：

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.Tensor.sparse_mask.md

@@ -0,0 +1,25 @@
+## [ 仅 API 调用方式不一致 ]torch.Tensor.sparse_mask
+
+### [torch.Tensor.sparse_mask](https://pytorch.org/docs/stable/tensors.html#torch.Tensor.sparse_mask)
+
+```python
+torch.Tensor.sparse_mask(mask)
+```
+
+### [paddle.sparse.mask_as](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/sparse/mask_as_cn.html#paddle/sparse/mask_as_cn#cn-api-paddle-sparse-mask_as)
+
+```python
+paddle.sparse.mask_as(x, mask, name=None)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+out = x.sparse_mask(coo)
+
+# Paddle 写法
+out = paddle.sparse.mask_as(x, coo)
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.distributions.distribution.Distribution.log_prob.md

@@ -0,0 +1,27 @@
+## [ 仅 API 调用方式不一致 ]torch.distributions.distribution.Distribution.log_prob
+
+### [torch.distributions.distribution.Distribution.log_prob](https://pytorch.org/docs/stable/generated/torch.distributions.distribution.Distribution.html#torch.distributions.distribution.Distribution.log_prob)
+
+```python
+torch.distributions.distribution.Distribution.log_prob(value)
+```
+
+### [paddle.distribution.Distribution.log_prob](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/distribution/Distribution/log_prob_cn.html#paddle/distribution/Distribution/log_prob_cn#cn-api-paddle-distribution-Distribution-log_prob)
+
+```python
+paddle.distribution.Distribution.log_prob(value)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+uniform = torch.distributions.Uniform(0.0, 1.0)
+result = uniform.log_prob(torch.tensor(0.3))
+
+# Paddle 写法
+uniform = paddle.distribution.Uniform(0.0, 1.0)
+result = uniform.log_prob(paddle.to_tensor(0.3))
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.max.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.max
+
+### [torch.max](https://pytorch.org/docs/stable/generated/torch.max.html)
+
+```python
+torch.max(*args, **kwargs)
+```
+
+### [paddle.compat.max](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/max_cn.html#paddle/compat/max_cn#cn-api-paddle-compat-max)
+
+```python
+paddle.compat.max(*args, **kwargs)
+```
+
+torch.max 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.max(x)
+
+# Paddle 写法
+result = paddle.compat.max(x)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.median.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.median
+
+### [torch.median](https://pytorch.org/docs/stable/generated/torch.median.html)
+
+```python
+torch.median(*args, **kwargs)
+```
+
+### [paddle.compat.median](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/median_cn.html#paddle/compat/median_cn#cn-api-paddle-compat-median)
+
+```python
+paddle.compat.median(*args, **kwargs)
+```
+
+torch.median 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.median(input)
+
+# Paddle 写法
+result = paddle.compat.median(input)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.min.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.min
+
+### [torch.min](https://pytorch.org/docs/stable/generated/torch.min.html)
+
+```python
+torch.min(*args, **kwargs)
+```
+
+### [paddle.compat.min](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/min_cn.html#paddle/compat/min_cn#cn-api-paddle-compat-min)
+
+```python
+paddle.compat.min(*args, **kwargs)
+```
+
+torch.min 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.min(x)
+
+# Paddle 写法
+result = paddle.compat.min(x)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nanmedian.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.nanmedian
+
+### [torch.nanmedian](https://pytorch.org/docs/stable/generated/torch.nanmedian.html)
+
+```python
+torch.nanmedian(*args, **kwargs)
+```
+
+### [paddle.compat.nanmedian](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/nanmedian_cn.html#paddle/compat/nanmedian_cn#cn-api-paddle-compat-nanmedian)
+
+```python
+paddle.compat.nanmedian(*args, **kwargs)
+```
+
+torch.namedian 有多种签名与用法，均只需修改 torch 前缀为 paddle.compat，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.nanmedian(input)
+
+# Paddle 写法
+result = paddle.compat.nanmedian(input)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.Unfold.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.nn.Unfold
+
+### [torch.nn.Unfold](https://pytorch.org/docs/stable/generated/torch.nn.Unfold.html#torch.nn.Unfold)
+
+```python
+torch.nn.Unfold(kernel_size, dilation=1, padding=0, stride=1)
+```
+
+### [paddle.compat.nn.Unfold](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/nn/Unfold_cn.html#paddle/compat/nn/Unfold_cn#cn-api-paddle-compat-nn-Unfold)
+
+```python
+paddle.compat.nn.Unfold(kernel_size, dilation=1, padding=0, stride=1)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+unfold = torch.nn.Unfold(kernel_size=(2, 2))
+
+# Paddle 写法
+unfold = paddle.compat.nn.Unfold(kernel_size=(2, 2))
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.pad.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.nn.functional.pad
+
+### [torch.nn.functional.pad](https://pytorch.org/docs/stable/nn.functional.html#torch.nn.functional.pad)
+
+```python
+torch.nn.functional.pad(input, pad, mode="constant", value=None)
+```
+
+### [paddle.compat.nn.functional.pad](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/nn/functional/pad_cn.html#paddle/compat/nn/pad_cn#cn-api-paddle-compat-nn-pad)
+
+```python
+paddle.compat.nn.functional.pad(input, pad, mode="constant", value=0.0)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.nn.functional.pad(x, [0, 0, 0, 0, 0, 1, 2, 3], value=1)
+
+# Paddle 写法
+result = paddle.compat.nn.functional.pad(x, [0, 0, 0, 0, 0, 1, 2, 3], value=1)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.nn.functional.softmax.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.nn.functional.softmax
+
+### [torch.nn.functional.softmax](https://pytorch.org/docs/stable/nn.functional.html#torch.nn.functional.softmax)
+
+```python
+torch.nn.functional.softmax(input, dim=None, _stacklevel=3, dtype=None)
+```
+
+### [paddle.compat.nn.functional.softmax](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/nn/functional/softmax_cn.html#paddle/compat/nn/softmax_cn#cn-api-paddle-compat-nn-softmax)
+
+```python
+paddle.compat.nn.functional.softmax(input, dim=None, _stacklevel=3, dtype=None, out=None)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.nn.functional.softmax(x, -1)
+
+# Paddle 写法
+result = paddle.compat.nn.functional.softmax(x, -1)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.optim.Optimizer.add_param_group.md

@@ -0,0 +1,36 @@
+## [ 仅 API 调用方式不一致 ]torch.optim.Optimizer.add_param_group
+
+### [torch.optim.Optimizer.add_param_group](https://pytorch.org/docs/stable/generated/torch.optim.Optimizer.html#torch.optim.Optimizer.add_param_group)
+
+```python
+torch.optim.Optimizer.add_param_group(param_group)
+```
+
+### [paddle.optimizer.Optimizer._add_param_group](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/Optimizer/_add_param_group_cn.html#paddle/optimizer/Optimizer/_add_param_group_cn#cn-api-paddle-optimizer-Optimizer-_add_param_group)
+
+```python
+paddle.optimizer.Optimizer._add_param_group(param_group)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+optimizer = torch.optim.SGD(pg1, lr=0.1, momentum=0.9, weight_decay=0.0005)
+optimizer.add_param_group({
+    'params': pg2,
+    'lr': 0.1 * 2,
+    'weight_decay': 0.0
+})
+
+# Paddle 写法
+optimizer = paddle.optimizer.SGD(learning_rate=0.1, parameters=pg1, weight_decay=0.0005)
+optimizer._add_param_group({
+    'params': pg2,
+    'learning_rate': 0.1 * 2,
+    'weight_decay': 0.0
+})
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.optim.Optimizer.load_state_dict.md

@@ -0,0 +1,30 @@
+## [ 仅 API 调用方式不一致 ]torch.optim.Optimizer.load_state_dict
+
+### [torch.optim.Optimizer.load_state_dict](https://pytorch.org/docs/stable/generated/torch.optim.Optimizer.html#torch.optim.Optimizer.load_state_dict)
+
+```python
+torch.optim.Optimizer.load_state_dict(state_dict)
+```
+
+### [paddle.optimizer.Optimizer.load_state_dict](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/optimizer/Optimizer/load_state_dict_cn.html#paddle/optimizer/Optimizer/load_state_dict_cn#cn-api-paddle-optimizer-Optimizer-load_state_dict)
+
+```python
+paddle.optimizer.Optimizer.load_state_dict(state_dict)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+optim = torch.optim.SGD([theta], lr=1.0)
+result = optim.state_dict()
+optim.load_state_dict(result)
+
+# Paddle 写法
+optim = paddle.optimizer.SGD(learning_rate=1.0, parameters=[theta])
+result = optim.state_dict()
+optim.load_state_dict(result)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.sort.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.sort
+
+### [torch.sort](https://pytorch.org/docs/stable/generated/torch.sort.html)
+
+```python
+torch.sort(input, dim=-1, descending=False, *, stable=False, out=None)
+```
+
+### [paddle.compat.sort](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/sort_cn.html#paddle/compat/sort_cn#cn-api-paddle-compat-sort)
+
+```python
+paddle.compat.sort(input, dim=-1, descending=False, stable=False, out=None)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.sort(a)
+
+# Paddle 写法
+result = paddle.compat.sort(a)
+
+```

docs/guides/model_convert/convert_from_pytorch/api_difference/invok_only_diff/torch.split.md

@@ -0,0 +1,26 @@
+## [ 仅 API 调用方式不一致 ]torch.split
+
+### [torch.split](https://pytorch.org/docs/stable/generated/torch.split.html)
+
+```python
+torch.split(tensor, split_size_or_sections, dim=0)
+```
+
+### [paddle.compat.split](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/api/paddle/compat/split_cn.html#paddle/compat/split_cn#cn-api-paddle-compat-split)
+
+```python
+paddle.compat.split(tensor, split_size_or_sections, dim=0)
+```
+
+两者功能一致，但调用方式不一致，具体如下：
+
+### 转写示例
+
+```python
+# PyTorch 写法
+result = torch.split(a, 2)
+
+# Paddle 写法
+result = paddle.compat.split(a, 2)
+
+```