We'd love help on these.

The check box is for if the OpInfo has added to PyTorch core. The ultimate goal is for all of these OpInfos to exist in PyTorch core.
The OpInfo is bolded if we have a poor man's version* of the OpInfo in the functorch repo (see https://github.com/facebookresearch/functorch/blob/main/test/functorch_additional_op_db.py).

Exists

• torch.nn.functional.softmax (OpInfo for nn.functional.softmax pytorch#62077)
• torch.nn.functional.relu (OpInfo for nn.functional.relu pytorch#62076)
• torch.nn.functional.interpolate (OpInfo for nn.functional.interpolate pytorch#61956)
• torch.nn.functional.pad ([opinfo] nn.functional.pad pytorch#62814)
• torch.nn.functional.normalize (OpInfo for torch.nn.functional.normalize pytorch#62635)
• torch.nn.functional.cross_entropy (OpInfo for nn.functional.cross_entropy pytorch#63547)
• torch.nn.functional.grid_sample (add OpInfo for torch.nn.functional.grid_sample pytorch#62311)
• torch.nn.functional.one_hot (add OpInfo for torch.nn.functional.one_hot pytorch#62253)
• torch.nn.functional.mse_loss
• torch.nn.functional.conv2d (OpInfo: nn.functional.conv2d pytorch#63517)
• torch.nn.functional.dropout (add OpInfo for torch.nn.functional.dropout pytorch#62315)
• torch.nn.functional.softplus (add OpInfo for torch.nn.functional.softplus pytorch#62317)
• torch.nn.functional.linear (nn.functional.linear OpInfo pytorch#61971)
• torch.nn.functional.avg_pool2d (OpInfo for nn.functional.avg_pool2d pytorch#62455)
• torch.nn.functional.max_pool2d (Add Maxpool to shape analysis / Opinfo pytorch#63530)
• torch.nn.functional.nll_loss (BUG Fixes regression for nllloss gradcheck pytorch#64203)
• torch.nn.functional.embedding (OpInfo nn.functional.embedding pytorch#63633)
• torch.nn.functional.adaptive_avg_pool2d (OpInfo for adaptive_avg_pool2d pytorch#62704)
• torch.nn.functional.cosine_similarity (Add OpInfo for nn.functional.cosine_similarity pytorch#62959)
• torch.nn.functional.unfold [opinfo] nn.functional.unfold pytorch#62705
• torch.nn.functional.batch_norm (OpInfo for nn.functional.batch_norm pytorch#63218)
• torch.nn.functional.conv_transpose2d OpInfo: nn.functional.conv_transpose2d pytorch#62882
• torch.nn.functional.layer_norm OpInfo for nn.functional.layer_norm pytorch#63276

*Why do we have poor man's version of these OpInfos? It's because right now we only care about float32 sample inputs on CPU and CUDA and OpInfos have a lot of flags that take some time to tweak.