dnd 5e – How can I scale down fire breath for use as a Creme Brulee torch?

I am starting a campaign in which my character is an accomplished and award-winning chef, and in an egg-themed competition won a dragon egg with a pseudodragon. I’ve talked with my DM, and he’s letting me replace the “Sting” ability with the ability to shoot small fire breaths. The entire point of this is to allow my chef character to be able to cook and use the dragon breath as a creme brulee torch.

The fire breath I am using in this example to scale down is for the Adult Red (or Brass) Dragon, whose actions state:

Fire Breath (Recharge 5-6): The Dragon exhales fire in a 60-foot cone. Each creature in that area must make a DC 21 Dexterity saving throw, taking 63 (18d6) fire damage on a failed save, or half as much damage on a successful one.

I don’t want to necessarily do damage to my tasty creations, but I would like for it to be a living being that assists me in my culinary creations.

How can I scale fire breath for use as a Creme Brulee torch?

How To Install Cuda for torch 1.8.1

I’m using ubuntu 20.04 LTS on Windows10.
And I’m going to use my GPU for torch.

What should I install?
cuda | cuDNN | cuda driver
Actually, I don’t know if cuDNN is necessary.

Following are my Computer SPECifications.

  2. Framework: torch 1.8.1 (Is 1.8.1+cu111 is latest version?)
  3. OS: Ubuntu 20.04 LTS

I tried a lot, but maybe version is not correct.

My main questions is as follows.

  1. Should I install cuda, cuDNN, driver all?
  2. What versions should I install?

dnd 5e – A torch in the dark: how high is the light level?

The lack of a light source does not affect the brightness of another

The brightness of a flashlight does not depend on whether its light overlaps with a background light source like moonlight. This is evidenced by the fact that a torch in a pitch-dark cave (where there is no moonlight for sure) works as well underground as in a moonlit (or not moonlit) night on the surface. In addition, moonlight is usually already considered dark, as explained in the description of darkness in the section Vision and Light of the Rules (emphasis added):

Darkness creates a heavily hidden area. Characters are exposed to darkness outdoors at night (even most moon nights)Within the confines of an unlit dungeon or subterranean vault, or within a range of magical darkness.

So there is no mechanical difference between a moonlit night and a moonless night anyway.

I can not find anything in the rules that explicitly states how the lights are made up of multiple light sources, but the most logical rule I can think of is that the brightest light source prevails. That is, when an object is illuminated by both bright and dim light, that object will be brightly lit. If another object is illuminated by two weak light sources, it is dimly lit.

Your DM can, however, set up new rules

Nevertheless, the DM can implement any rules. It is possible that the "non-magical" darkness of this moonless night is just another kind of magic. The Sage Advice Compendium explains:

You might think, "Dragons seem pretty magical to me." And yes, they are extraordinary! Their description even says that they are magical. However, our game distinguishes two types of magic:

  • the background magic that is part of the physics of the D & D multiverse and the physiology of many D & D creatures
  • The concentrated magical energy contained in a magical item or channeled to create a spell or other focused magical effect

In D & D, the first kind of magic is nature. It is not more compatible than the wind. A monster like a dragon exists because of this magically enhanced nature. The second kind of magic concerns the rules. When a rule refers to something magical, it refers to that second type.

In view of this, it would not be unreasonable for the DM to decide that the darkness of a moonless night in the world in which they build up is not only a lack of light, but beyond that, a little supernatural advantage has all the light a little darker , This would be particularly appropriate in a horror campaign.

Python – Share Torch Tensor by Flags

When I work with PyTorch tensors, I have to divide the stack of articles according to its flags, so that the articles in x_batch_one and x_batch_two are lists of couples with the same flags.

def pairs_by_flags (self, x_batch, y_batch, max_flag):
#split x_batch for couples

Sizes = (x_batch.size (0) // 2, * x_batch.size ()[1:] )

items_pos = torch.zeros ((max_flag,), dtype = torch.long, device = x_batch.device)

x_batch_one = torch.zeros (sizes, device = x_batch.device)
x_batch_one_pos = 0
x_batch_two = torch.zeros (sizes, device = x_batch.device)

for index, flag in enumerate (y_batch):
x_item = x_batch[index]
        i_flag = items_pos[flag]
        if i_flag:
x_batch_two[i_flag-1] = x_item
items_pos[flag] = 0
x_batch_one[x_batch_one_pos] = x_item
x_batch_one_pos + = 1
items_pos[flag] = x_batch_one_pos

Return x_batch_one, x_batch_two    

Here is a simple example with single digits:

by torch_helper.myt import thelp
Import torch

arr = torch.tensor ([2,7,6,2,7,3,2,8,8,6,2,3])

arr1, arr2 = thelp.pairs_by_flags (arr, arr, 9)

print (arr1)

print (arr2)


Tensor[2., 7., 6., 3., 2., 8.])
Tensor[2., 7., 6., 3., 2., 8.])

But after I have done this naive solution, I wonder if it could be improved by perhaps eliminating the loop with a tensor [:,,,:,,,:] Magic.

dnd 5e – Can the light of a torch become blind under the effect of a light?

The rules say this about torches:

torch, A torch burns for 1 hour and provides bright light in a radius of 20 feet. The light is dimmed for another 20 feet. When you make and strike a melee attack with a burning torch, it deals 1 Fire damage.

The magic lights have this description:

You touch an object that is no larger than 10 feet in any dimension. By the end of the spell, the object will emit bright light within a 20 foot radius and dim the light for an additional 20 feet. The light can be colored as desired. Full coverage of the object with something opaque blocks the light. The spell ends when you reactivate it or reject it as an action.

If you target an object held or carried by an enemy creature, then that
The creature must perform a skillful throw to avoid the spell.

The rules only cover these different levels of light: sunlight, bright light, dim light, darkness and magical darkness. There are no rules that say you can create extra bright light by stacking light sources on each other and dazzling a creature.

That being said, it is up to you to allow or forbid what the players are trying to do. If the dragon had lain in the dark for some time, it would be reasonable for a sudden bright light to be painful to him.

dnd 5e – Can the magician's hand be renewed and moved to carry a torch all the time?

Magician hand specified (PHB 256):

The hand disappears when […] You speak this spell again.

In addition,

You can use your action to control the hand.

Since casting the spell is an action, it takes 6 seconds after the first hand is gone before you can grab the torch again. Under normal circumstances, the torch would fall to the ground.

However, many DMs (including myself) dictate that they can do what you ask, even if that is not the strict reading.

How can you store cuda memory with torch in this case?

Like the code below, I try to find the mean of 47 repeated outputs of the model. However, it is always without memory. If I remove that z_proto_class_list.append (z_proto_class)then it is good. I think that's because the memory will be released if I do not attach the tensor. I always try to generate the 47 issue at once, but it's obviously more memory intensive than my current choice. Is there a way to solve my current problem? Many Thanks.

                                z_proto_class_list = []
        For support_input_ids, support_input_mask, support_segment_ids in the data loader:
s_z, s_pooled_output = model (support_input_ids, support_input_mask, support_segment_ids, output_all_encoded_layers = false)
sz_dim = s_z.size (-1)
index = torch.LongTensor (support_idx_list) .unsqueeze (1) .unsqueeze (2) .expand (len (support_idx_list), 1, sz_dim) .cuda ()
z_proto_raw = torch.gather (s_z, 1, index)
z_proto_class = z_proto_raw.view (1, n_support, sz_dim) .mean (1)
z_proto_class_list.append (z_proto_class)
torch.cuda.empty_cache ()
z_proto = torch.cat (z_proto_class_list, 0)

mac – Why do I get an error message when I am t = torch.rand (5) in torch?

I'm a beginner at pytorch and follow a basic course (https://www.youtube.com/watch?v=h8n6HrLfeJQ) when I get to the point where I'm supposed to type
t = torch.rand (5)
There is the following error message:
AttributeError: Module & # 39; Burner & # 39; has no attribute & # 39; margin & # 39;

The frustrating thing is that it worked yesterday. I tried to uninstall and reinstall Torch, but the same thing still happens.