1. 1. 1. 🐛 Describe the bug

Setting the float value `3.` to `input_size` of [nn.RNN()](https://pytorch.org/docs/stable/generated/torch.nn.RNN.html) gets the indirect error message as shown below:

```python
import torch
from torch import nn

my_tensor = torch.tensor([[8., -3., 5.]])

torch.manual_seed(42)

1. ↓↓
   rnn = nn.RNN(input_size=3., hidden_size=2) # Error

rnn(input=my_tensor)
```

> TypeError: empty(): argument 'size' failed to unpack the object at pos 2 with error "type must be tuple of ints,but got float"

And, setting the boolean value `True` to `input_size` of `nn.RNN()` gets the indirect error message as shown below:

```python
import torch
from torch import nn

my_tensor = torch.tensor([[8., -3., 5.]])

torch.manual_seed(42)

1. ↓↓↓↓
   rnn = nn.RNN(input_size=True, hidden_size=2)

rnn(input=my_tensor) # Error
```

> RuntimeError: input.size(-1) must be equal to input_size. Expected True, got 3

So, the error messages should be something direct like as shown below:

> TypeError: `input_size` argument must be `int` but got `float`

> TypeError: `input_size` argument must be `int` but got `bool`

1. 1. 1. Versions

```python
import torch

torch._version_ # '2.3.0'
```

cc @albanD @mruberry @jbschlosser @walterddr @mikaylagawarecki