diff --git a/docs/pytorch_project_convertor/API_docs/ops/README.md b/docs/pytorch_project_convertor/API_docs/ops/README.md index f75ab519f9a6e54bbb3fc7f0a6ddff65b0638a53..1de45de1e8ff8daa2c877aba4821375755151ae6 100644 --- a/docs/pytorch_project_convertor/API_docs/ops/README.md +++ b/docs/pytorch_project_convertor/API_docs/ops/README.md @@ -90,5 +90,56 @@ | 85 | [torch.argmin](https://pytorch.org/docs/stable/generated/torch.argmin.html?highlight=argmin#torch.argmin) | [paddle.argmin](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/argmin_cn.html#argmin) | 功能一致，参数名不一致 | | 86 | [torch.max](https://pytorch.org/docs/stable/generated/torch.max.html?highlight=max#torch.max) | [paddle.max](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/max_cn.html#max) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | | 87 | [torch.min](https://pytorch.org/docs/stable/generated/torch.min.html?highlight=min#torch.min) | [paddle.min](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/min_cn.html#min) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 88 | [torch.square](https://pytorch.org/docs/stable/generated/torch.square.html?highlight=square#torch.square) | [paddle.square](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/square_cn.html#square) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 89 | [torch.sub](https://pytorch.org/docs/stable/generated/torch.sub.html?highlight=sub#torch.sub) | [paddle.subtract](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/subtract_cn.html#subtract) | [差异对比](./torch.sub.md) | +| 90 | [torch.subtract](https://pytorch.org/docs/stable/generated/torch.subtract.html?highlight=subtract#torch.subtract) | [paddle.subtract](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/subtract_cn.html#subtract) | [差异对比](./torch.subtract.md) | +| 91 | [torch.tanh](https://pytorch.org/docs/stable/generated/torch.tanh.html?highlight=tanh#torch.tanh) | [paddle.tanh](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/tanh_cn.html#tanh) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 92 | [torch.true_divide](https://pytorch.org/docs/stable/generated/torch.true_divide.html?highlight=true_divide#torch.true_divide) | [paddle.divide](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/divide_cn.html#divide) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 93 | [torch.dist](https://pytorch.org/docs/stable/generated/torch.dist.html?highlight=dist) | [paddle.dist](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/dist_cn.html#dist) | 功能一致，参数名不一致 | +| 94 | [torch.logsumexp](https://pytorch.org/docs/stable/generated/torch.logsumexp.html?highlight=logsumexp#torch.logsumexp) | [paddle.logsumexp](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/logsumexp_cn.html#logsumexp) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 95 | [torch.mean](https://pytorch.org/docs/stable/generated/torch.mean.html?highlight=mean) | [paddle.mean](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/mean_cn.html#mean) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 96 | [torch.median](https://pytorch.org/docs/stable/generated/torch.median.html?highlight=median#torch.median) | [paddle.median](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/median_cn.html#median) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 97 | [torch.norm](https://pytorch.org/docs/stable/generated/torch.norm.html?highlight=norm#torch.norm) | [paddle.norm](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/norm_cn.html#norm) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor以及`dtype`参数代表输出Tensor类型 | +| 98 | [torch.prod](https://pytorch.org/docs/stable/generated/torch.prod.html?highlight=prod#torch.prod) | [paddle.prod](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/prod_cn.html#prod) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 99 | [torch.std](https://pytorch.org/docs/stable/generated/torch.std.html?highlight=std#torch.std) | [paddle.std](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/std_cn.html#std) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 100 | [torch.std_mean](https://pytorch.org/docs/stable/generated/torch.std_mean.html?highlight=std_mean#torch.std_mean) | 无对应实现 | [组合实现](./torch.std_mean.md) | +| 101 | [torch.sum](https://pytorch.org/docs/stable/generated/torch.sum.html?highlight=sum#torch.sum) | [paddle.sum](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/sum_cn.html#sum) | 功能一致，参数名不一致 | +| 102 | [torch.unique](https://pytorch.org/docs/stable/generated/torch.unique.html?highlight=unique#torch.unique) | [paddle.unique](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/unique_cn.html#unique) | [差异对比](./torch.unique.md) | +| 103 | [torch.var](https://pytorch.org/docs/stable/generated/torch.var.html?highlight=var#torch.var) | [paddle.var](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/var_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 104 | [torch.var_mean](https://pytorch.org/docs/stable/generated/torch.var_mean.html?highlight=var_mean#torch.var_mean) | 无对应实现 | [组合实现](./torch.var_mean.md) | +| 105 | [torch.allclose](https://pytorch.org/docs/stable/generated/torch.allclose.html?highlight=allclose#torch.allclose) | [paddle.allclose](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/allclose_cn.html) | 功能一致，参数名不一致 | +| 106 | [torch.argsort](https://pytorch.org/docs/stable/generated/torch.argsort.html?highlight=argsort#torch.argsort) | [paddle.argsort](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/argsort_cn.html) | 功能一致，参数名不一致 | +| 107 | [torch.eq](https://pytorch.org/docs/stable/generated/torch.eq.html?highlight=eq#torch.eq) | [paddle.equal](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/equal_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 108 | [torch.equal](https://pytorch.org/docs/stable/generated/torch.equal.html?highlight=eq#torch.equal) | [paddle.equal_all](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/equal_all_cn.html) | 功能一致，参数名不一致 | +| 109 | [torch.ge](https://pytorch.org/docs/stable/generated/torch.ge.html?highlight=ge#torch.ge) | [paddle.greater_equal](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/greater_equal_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 110 | [torch.gt](https://pytorch.org/docs/stable/generated/torch.gt.html?highlight=gt#torch.gt) | [paddle.greater_than](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/greater_than_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 111 | [torch.isfinite](https://pytorch.org/docs/stable/generated/torch.isfinite.html?highlight=isfinite#torch.isfinite) | [paddle.isfinite](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/isfinite_cn.html) | 功能一致，参数名不一致 | +| 112 | [torch.isinf](https://pytorch.org/docs/stable/generated/torch.isinf.html?highlight=isinf#torch.isinf) | [paddle.isinf](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/isinf_cn.html) | 功能一致，参数名不一致 | +| 113 | [torch.isnan](https://pytorch.org/docs/stable/generated/torch.isnan.html?highlight=isnan#torch.isnan) | [paddle.isnan](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/isnan_cn.html) | 功能一致，参数名不一致 | +| 114 | [torch.kthvalue](https://pytorch.org/docs/stable/generated/torch.kthvalue.html?highlight=kthvalue#torch.kthvalue) | 无对应实现 | [组合实现](./torch.kthvalue.md) | +| 115 | [torch.le](https://pytorch.org/docs/stable/generated/torch.le.html?highlight=le#torch.le) | [paddle.less_equal](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/less_equal_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 116 | [torch.lt](https://pytorch.org/docs/stable/generated/torch.lt.html?highlight=lt#torch.lt) | [paddle.less_than](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/less_than_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 117 | [torch.maximum](https://pytorch.org/docs/stable/generated/torch.maximum.html?highlight=max#torch.maximum) | [paddle.maximum](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/api/paddle/tensor/math/maximum_cn.html#cn-api-paddle-tensor-maximum) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 118 | [torch.minimum](https://pytorch.org/docs/stable/generated/torch.minimum.html?highlight=minimum#torch.minimum) | [paddle.minimum](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/api/paddle/tensor/math/minimum_cn.html#cn-api-paddle-tensor-minimum) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 119 | [torch.ne](https://pytorch.org/docs/stable/generated/torch.ne.html?highlight=ne#torch.ne) | [paddle.not_equal](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/not_equal_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 120 | [torch.sort](https://pytorch.org/docs/stable/generated/torch.sort.html?highlight=sort#torch.sort) | [paddle.argsort](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/argsort_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 121 | [torch.topk](https://pytorch.org/docs/stable/generated/torch.topk.html?highlight=topk#torch.topk) | [paddle.topk](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/topk_cn.html#topk) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 122 | [torch.cross](https://pytorch.org/docs/stable/generated/torch.cross.html?highlight=cross#torch.cross) | [paddle.cross](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/cross_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 123 | [torch.any](https://pytorch.org/docs/stable/generated/torch.any.html?highlight=any#torch.any) | [paddle.any](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/any_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 124 | [torch.cumsum](https://pytorch.org/docs/stable/generated/torch.cumsum.html?highlight=cumsum#torch.cumsum) | [paddle.cumsum](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/cumsum_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 125 | [torch.diag](https://pytorch.org/docs/stable/generated/torch.diag.html?highlight=diag#torch.diag) | [paddle.diag](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/diag_cn.html) | [差异对比](./torch.diag.md) | +| 126 | [torch.diag_embed](https://pytorch.org/docs/stable/generated/torch.diag_embed.html?highlight=diag_embed#torch.diag_embed) | [paddle.nn.functional.diag_embed](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/nn/functional/diag_embed_cn.html) | 功能一致 | +| 127 | [torch.einsum](https://pytorch.org/docs/stable/generated/torch.einsum.html?highlight=einsum#torch.einsum) | paddlenlp.ops.einsum | 功能一致，需要安装[paddlenlp](https://github.com/PaddlePaddle/PaddleNLP#pip%E5%AE%89%E8%A3%85) | +| 128 | [torch.flatten](https://pytorch.org/docs/stable/generated/torch.flatten.html?highlight=flatten#torch.flatten) | [paddle.flatten](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/flatten_cn.html) | 功能一致，参数名不一致 | +| 129 | [torch.flip](https://pytorch.org/docs/stable/generated/torch.flip.html?highlight=flip#torch.flip) | [paddle.flip](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/flip_cn.html) | 功能一致，参数名不一致 | +| 130 | [torch.rot90](https://pytorch.org/docs/stable/generated/torch.rot90.html?highlight=rot90#torch.rot90) | 无对应实现 | [组合实现](./torch.rot90.md) | +| 131 | [torch.meshgrid](https://pytorch.org/docs/stable/generated/torch.meshgrid.html?highlight=meshgrid#torch.meshgrid) | [paddle.meshgrid](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/meshgrid_cn.html) | 功能一致 | +| 132 | [torch.roll](https://pytorch.org/docs/stable/generated/torch.roll.html?highlight=roll#torch.roll) | [paddle.roll](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/roll_cn.html) | 功能一致，参数名不一致 | +| 133 | [torch.tril](https://pytorch.org/docs/stable/generated/torch.tril.html?highlight=tril#torch.tril) | [paddle.tril](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/tril_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 134 | [torch.triu](https://pytorch.org/docs/stable/generated/torch.triu.html?highlight=triu#torch.triu) | [paddle.triu](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/triu_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 135 | [torch.bmm](https://pytorch.org/docs/stable/generated/torch.bmm.html?highlight=bmm#torch.bmm) | [paddle.bmm](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/bmm_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 136 | [torch.cholesky](https://pytorch.org/docs/stable/generated/torch.cholesky.html?highlight=cholesky#torch.cholesky) | [paddle.cholesky](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/cholesky_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 137 | [torch.dot](https://pytorch.org/docs/stable/generated/torch.dot.html?highlight=dot#torch.dot) | [paddle.dot](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/dot_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | +| 138 | [torch.inverse](https://pytorch.org/docs/stable/generated/torch.inverse.html?highlight=inverse#torch.inverse) | [paddle.inverse](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/inverse_cn.html) | 功能一致，参数名不一致，PaddlePaddle未定义`out`参数代表输出Tensor | ***持续更新...*** diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.diag.md b/docs/pytorch_project_convertor/API_docs/ops/torch.diag.md new file mode 100644 index 0000000000000000000000000000000000000000..5a4863250c0bee623aaaa97405668f3c83a73fe1 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.diag.md @@ -0,0 +1,16 @@ +## torch.diag +### [torch.diag](https://pytorch.org/docs/stable/generated/torch.diag.html?highlight=diag#torch.diag) +```python +torch.diag(input, diagonal=0, *, out=None) +``` +### [paddle.diag](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/diag_cn.html) +```python +paddle.diag(x, offset=0, padding_value=0, name=None) +``` + +### 参数差异 +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| input | x | 表示输出的Tensor。 | +| diagonal | offset | 对角线偏移量。正值表示上对角线，0表示主对角线，负值表示下对角线。 | +| - | padding_value | 表示填充指定对角线以外的区域，PyTorch无此参数，其用0填充。 | diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.kthvalue.md b/docs/pytorch_project_convertor/API_docs/ops/torch.kthvalue.md new file mode 100644 index 0000000000000000000000000000000000000000..432acab0beb5187dac8b522d3e44b89332e8493f --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.kthvalue.md @@ -0,0 +1,23 @@ +## torch.kthvalue + +### [torch.kthvalue](https://pytorch.org/docs/stable/generated/torch.kthvalue.html?highlight=kthvalue#torch.kthvalue) + +```python +torch.kthvalue(input, k, dim=None, keepdim=False, out=None) +``` +### 功能介绍 +用于获取某一维度上第k大的数值，PaddlePaddle目前无对应API，可使用如下代码组合实现该API。 +```python +import paddle + +def kthvalue(input, k, dim=None, keepdim=False, out=None): + values = paddle.sort(input, axis=dim) + indices = paddle.argsort(input, axis=dim) + values = paddle.slice(values, axes=[dim], starts=[k-1], ends=[k]) + indices = paddle.slice(indices, axes=[dim], starts=[k-1], ends=[k]) + if not keepdim: + values = paddle.flatten(values) + indices = paddle.flatten(indices) + return values, indices +``` + diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.rot90.md b/docs/pytorch_project_convertor/API_docs/ops/torch.rot90.md new file mode 100644 index 0000000000000000000000000000000000000000..b244ab7362c17f7bfafb953760f71971a8ce7176 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.rot90.md @@ -0,0 +1,22 @@ +## torch.rot90 + +### [torch.rot90](https://pytorch.org/docs/stable/generated/torch.rot90.html?highlight=rot90#torch.rot90) +```python +torch.rot90(input, k, dims) +``` + +### 功能介绍 +用于实现对矩阵进行k次90度旋转，PaddlePaddle目前无对应API，可使用如下代码组合实现该API。 +```python +import paddle +def rot90(input, k, dims): + l = len(input.shape) + new_dims = list(range(l)) + new_dims[dims[0]] = dims[1] + new_dims[dims[1]] = dims[0] + flip_dim = min(dims) + for i in range(k): + input = paddle.transpose(input, new_dims) + input = paddle.flip(input, [flip_dim]) + return input +``` \ No newline at end of file diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.std_mean.md b/docs/pytorch_project_convertor/API_docs/ops/torch.std_mean.md new file mode 100644 index 0000000000000000000000000000000000000000..f3edbbd1b692ddcd0f332a2c3fecdf8e7bc7a8f9 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.std_mean.md @@ -0,0 +1,24 @@ +## torch.std_mean + +### [torch.std_mean](https://pytorch.org/docs/stable/generated/torch.std_mean.html?highlight=std_mean#torch.std_mean) +```python +# 用法一： +torch.std_mean(input, unbiased=True) +# 用法二： +torch.std_mean(input, dim, unbiased=True, keepdim=False) +``` + +### 功能介绍 +用于实现返回Tensor的标准差和均值，PaddlePaddle目前无对应API，可使用如下代码组合实现该API。 + +```python +import paddle + +def std_mean(input, dim=None, unbiased=True, keepdim=False): + std = paddle.std(input, axis=dim, + unbiased=unbiased, keepdim=keepdim) + mean = paddle.mean(input, + axis=dim, + keepdim=keepdim) + return std, mean +``` diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.sub.md b/docs/pytorch_project_convertor/API_docs/ops/torch.sub.md new file mode 100644 index 0000000000000000000000000000000000000000..2b050fc7393819eb16f07828789aab1aa0574000 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.sub.md @@ -0,0 +1,23 @@ +## torch.sub +### [torch.sub](https://pytorch.org/docs/stable/generated/torch.sub.html?highlight=torch%20sub#torch.sub) +```python +torch.sub(input, other, *, alpha=1, out=None) +``` +### [paddle.subtract](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/subtract_cn.html#subtract) +```python +paddle.subtract(x, y, name=None) +``` +### 参数差异 +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| alpha | - | 表示`other`的乘数，PaddlePaddle无此参数。 | +| out | - | 表示输出的Tensor，PaddlePaddle无此参数。 | + +### 功能差异 + +#### 计算差异 +***PyTorch***： +$ out = input - alpha * other $ + +***PaddlePaddle***： +$ out = x - y $ \ No newline at end of file diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.subtract.md b/docs/pytorch_project_convertor/API_docs/ops/torch.subtract.md new file mode 100644 index 0000000000000000000000000000000000000000..7a06ed77e09f34bc2105038ed1176465c2f6a2d0 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.subtract.md @@ -0,0 +1,23 @@ +## torch.subtract +### [torch.subtract](https://pytorch.org/docs/stable/generated/torch.subtract.html?highlight=subtract#torch.subtract) +```python +torch.subtract(input, other, *, alpha=1, out=None) +``` +### [paddle.subtract](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/subtract_cn.html#subtract) +```python +paddle.subtract(x, y, name=None) +``` +### 参数差异 +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| alpha | - | 表示`other`的乘数，PaddlePaddle无此参数。 | +| out | - | 表示输出的Tensor，PaddlePaddle无此参数。 | + +### 功能差异 + +#### 计算差异 +***PyTorch***： +$ out = input - alpha * other $ + +***PaddlePaddle***： +$ out = x - y $ \ No newline at end of file diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.unique.md b/docs/pytorch_project_convertor/API_docs/ops/torch.unique.md new file mode 100644 index 0000000000000000000000000000000000000000..7e5d1dd6a06adff8c60926f7f77c50bc7b0f112e --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.unique.md @@ -0,0 +1,73 @@ +## torch.unique +### [torch.unique](https://pytorch.org/docs/stable/generated/torch.unique.html?highlight=unique#torch.unique) +```python +torch.unique(input, sorted, return_inverse, return_counts, dim=None) +``` +### [paddle.unique](https://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/unique_cn.html#unique) +```python +paddle.unique(x, return_index=False, return_inverse=False, return_counts=False, axis=None, dtype='int64', name=None) +``` +### 参数差异 +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| input | x | 表示输入Tensor。 | +| sorted | - | 表示升序或者降序，PaddlePaddle无此参数。 | +| - | return_index | 表示是否返回独有元素在输入Tensor中的索引，PyTorch无此参数。 | +| return_inverse | return_inverse | 表示是否返回输入Tensor的元素对应在独有元素中的索引。 | +| return_counts | return_counts | 表示是否返回每个独有元素在输入Tensor中的个数。 | +| dim | axis | 表示指定选取独有元素的轴。 | +| dtype | - | 表示返回值的类型，PyTorch无此参数。 | + +### 功能差异 +#### 返回值差异 +| PyTorch | PaddlePaddle | 备注 | +| ------------- | ------------ | ------------------------------------------------------ | +| output | out | 表示独有元素构成的Tensor。 | +| inverse_indices | inverse | （可选）表示输入Tensor的元素对应在独有元素中的索引。 | +| counts | counts | （可选）表示每个独有元素在输入Tensor中的个数。 | +| - | index | （可选）表示独有元素在输入Tensor中的索引，PyTorch无此返回值。 | + +***PyTorch***：（1）可以通过`sorted`设置是返回升序还是降序结果；（2）`inverse_indices`形状与原形状相同。 +***PaddlePaddle***：（1）只能返回升序结果；（2）`inverse`为`inverse_indices`展平后形状。 + +### 代码示例 +``` python +# PyTorch示例： +import torch +import torch +x = torch.tensor([[[1, 3], [2, 3]], [[1, 6], [2, 3]]], dtype=torch.float32) +output, inverse_indices, counts = torch.unique(x, + sorted=True, + return_inverse=True, + return_counts=True) +# 输出 +# output: +# tensor([1., 2., 3., 6.]) +# inverse_indices: +# tensor([[[0, 2], +# [1, 2]], + +# [[0, 3], +# [1, 2]]]) +# counts: +# tensor([2, 2, 3, 1]) +``` + +``` python +# PaddlePaddle示例： +import paddle +x = paddle.to_tensor([[[1, 3], [2, 3]], [[1, 6], [2, 3]]], dtype="float32") +out, inverse, counts = paddle.unique(x, + return_inverse=True, + return_counts=True) +# 输出 +# out: +# Tensor(shape=[4], dtype=float32, place=CUDAPlace(0), stop_gradient=True, +# [1., 2., 3., 6.]) +# inverse: +# Tensor(shape=[8], dtype=int64, place=CUDAPlace(0), stop_gradient=True, +# [0, 2, 1, 2, 0, 3, 1, 2]) +# counts: +# Tensor(shape=[4], dtype=int64, place=CUDAPlace(0), stop_gradient=True, +# [2, 2, 3, 1]) +``` diff --git a/docs/pytorch_project_convertor/API_docs/ops/torch.var_mean.md b/docs/pytorch_project_convertor/API_docs/ops/torch.var_mean.md new file mode 100644 index 0000000000000000000000000000000000000000..2ce9003e2732fb37bc30c8ceb2cf0ee9eff49026 --- /dev/null +++ b/docs/pytorch_project_convertor/API_docs/ops/torch.var_mean.md @@ -0,0 +1,24 @@ +## torch.var_mean + +### [torch.var_mean](https://pytorch.org/docs/stable/generated/torch.var_mean.html?highlight=var_mean#torch.var_mean) +```python +# 用法一： +torch.var_mean(input, unbiased=True) +# 用法二： +torch.var_mean(input, dim, keepdim=False, unbiased=True) +``` + +### 功能介绍 +用于实现返回Tensor的方差和均值，PaddlePaddle目前无对应API，可使用如下代码组合实现该API。 + +```python +import paddle + +def var_mean(input, dim=None, unbiased=True, keepdim=False): + var = paddle.var(input, axis=dim, + unbiased=unbiased, keepdim=keepdim) + mean = paddle.mean(input, + axis=dim, + keepdim=keepdim) + return var, mean +```