Tensor extension to torch.clamp https://github.com/pytorch/pytorch/issues/2793#issuecomment-428784713

Compute the md5sum of a file in chunks. Parameters ---------- fname : Union[str, BinaryIO] Filename or file object chunksize

Compute the reward for a list of histories.

Concatenate the chosen and rejected inputs into a single tensor. Args: batch: A batch of data. Must contain the keys 'chosen_inpu

Copy/paste/tweak model's weights to our SigLIP structure.

Converts the stats from a flattened dict to single scalar dicts

Creates and saves a model card for a TRL model. Args: path (`str`): The path to save the model card to. model_name (`

Creates and saves a model card for a TRL model. Args: path (`str`): The path to save the model card to. model_name (`

Setup the optimizer. We provide a reasonable default that works well. If you want to use something else, you can pass a tuple in the

Removes the `_require_grads_hook`.

Download a file from a remote url to a local path. The remote url can be a pipe: url, in which case the remainder of the url is treated as a c

Compute the DPO loss for a batch of policy and reference model log probabilities. Args: policy_chosen_logps: Log probabilities of

DropBlock. See https://arxiv.org/pdf/1810.12890.pdf DropBlock with an experimental gaussian noise option. This layer has been tested on a few tra

DropBlock. See https://arxiv.org/pdf/1810.12890.pdf DropBlock with an experimental gaussian noise option. Simplied from above without concern for

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for E

Returns the size (in bytes) occupied by one parameter of type `dtype`. Example: ```py >>> dtype_byte_size(torch.float32) 4

Calculate entropy from logits.

Helper function to estimate the total number of tokens from the model inputs. Args: inputs (`dict`): The model inputs.

Evaluate a multiple choice instance.

Evaluate an open question instance

Evaluate your results against the ground truth Args: - your_results_path (str): Path to your results file - video_types (Optional[Li

Evaluate your results against the ground truth Args: - your_results_path (str): Path to your results file - video_types (Optional[Li

Batch evaluation for multiple choice and open questions.

Expand a device map to return the correspondance parameter name to device.

Exact all forms of numbers from a string with regex.

A helper util to find the last sub-module and the param/buffer name. If `start_prefix` is supplied it'll be removed from the start of the key

Huggingface will call model.train() at each training_step. To ensure the expected behaviors for modules like dropout, batchnorm, etc., we nee

Gather stats from all processes. Useful in the context of distributed training. Args: stats (dict[str, Any]):

r""" A simple wrapper around the `generate` method of the wrapped model. Please refer to the [`generate`](https://huggingface.co/docs/

Compute the log probabilities of the given labels under the given logits. Args: logits: Logits of the model (unnormalized). Shape

Generate samples from the model and reference model for the given batch of inputs.