Create tf.data Instance from txt file Args: path_txt: (string) path containing one example per line vocab: (tf.lookuptable)

Update word and tag vocabulary from dataset Args: txt_path: (string) path to file, one sentence per line vocab: (dict or Counter)

Update word and tag vocabulary from dataset Args: txt_path: (string) path to file, one sentence per line vocab: (dict or Counter)

Saves parameters to json file

Loads parameters from json file

Returns a generator that yields batches data with labels. Batch size is params.batch_size. Expires after one pass over the data.

Input function for the SIGNS dataset. The filenames have format "{label}_IMG_{id}.jpg". For instance: "data_dir/2_IMG_4584.jpg". Args:

Input function for NER Args: mode: (string) 'train', 'eval' or any other mode you can think of At training, we shuff

Model function defining the graph operations. Args: mode: (string) can be 'train' or 'eval' inputs: (dict) contains the inputs of

Model function defining the graph operations. Args: mode: (string) 'train', 'eval', etc. inputs: (dict) contains the inputs of th

Saves parameters to json file

Writes sentences.txt and labels.txt files in save_dir from dataset Args: dataset: ([(["a", "cat"], ["O", "O"]), ...]) save_dir: (

Writes sentences.txt and labels.txt files in save_dir from dataset Args: dataset: ([(["a", "cat"], ["O", "O"]), ...]) save_dir: (

Saves dict of floats in json file Args: d: (dict) of float-castable values (np.float, int, float, etc.) json_path: (string) path

Saves dict of floats in json file Args: d: (dict) of float-castable values (np.float, int, float, etc.) json_path: (string) path

Loads parameters from json file

Evaluate the model on `num_steps` batches. Args: model: (torch.nn.Module) the neural network loss_fn: a function that takes batch

Evaluate the model on `num_steps` batches. Args: model: (torch.nn.Module) the neural network loss_fn: a function that takes batch

Train the model on `num_steps` batches. Args: sess: (tf.Session) current session model_spec: (dict) contains the graph operations

Train the model on `num_steps` batches. Args: sess: (tf.Session) current session model_spec: (dict) contains the graph operations

Loads the data for each type in types from data_dir. Args: types: (list) has one or more of 'train', 'val', 'test' depen

Compute the cross entropy loss given outputs and labels. Args: outputs: (Variable) dimension batch_size x 6 - output of the model

Compute the cross entropy loss given outputs from the model and labels for all tokens. Exclude loss terms for PADding tokens. Args:

Writes one token per line, 0-based line id corresponds to the id of the token. Args: vocab: (iterable object) yields token txt_pa

Writes one token per line, 0-based line id corresponds to the id of the token. Args: vocab: (iterable object) yields token txt_pa

Sets the logger to log info in terminal and file `log_path`. In general, it is useful to have a logger so that every output to the terminal is sa

Sets the logger to log info in terminal and file `log_path`. In general, it is useful to have a logger so that every output to the terminal is sa

Loads parameters from json file

Obtain the image from the filename (for both training and validation). The following operations are applied: - Decode the image from jpeg

Aggregate the metrics of all experiments in folder `parent_dir`. Assumes that `parent_dir` contains multiple experiments, with their results stor

Aggregate the metrics of all experiments in folder `parent_dir`. Assumes that `parent_dir` contains multiple experiments, with their results stor

Aggregate the metrics of all experiments in folder `parent_dir`. Assumes that `parent_dir` contains multiple experiments, with their results stor

Aggregate the metrics of all experiments in folder `parent_dir`. Assumes that `parent_dir` contains multiple experiments, with their results stor

Compute logits of the model (output distribution) Args: is_training: (bool) whether we are training or not inputs: (dict) contain

Compute logits of the model (output distribution) Args: mode: (string) 'train', 'eval', etc. inputs: (dict) contains the inputs o

Evaluate the model Args: model_spec: (dict) contains the graph operations or nodes needed for evaluation model_dir: (string) dire

Evaluate the model Args: model_spec: (dict) contains the graph operations or nodes needed for evaluation model_dir: (string) dire

Launch training of the model with a set of hyperparameters in parent_dir/job_name Args: parent_dir: (string) directory containing config,

Launch training of the model with a set of hyperparameters in parent_dir/job_name Args: parent_dir: (string) directory containing config,

Launch training of the model with a set of hyperparameters in parent_dir/job_name Args: model_dir: (string) directory containing config,

Launch training of the model with a set of hyperparameters in parent_dir/job_name Args: model_dir: (string) directory containing config,

Loads dataset into memory from csv file

Loads dataset into memory from csv file

Loads sentences and labels from their corresponding files. Maps tokens and tags to their indices and stores them in the provided dict

Resize the image contained in `filename` and save it to the `output_dir`

Resize the image contained in `filename` and save it to the `output_dir`

Saves dict to json file Args: d: (dict) json_path: (string) path to json file

Saves dict to json file Args: d: (dict) json_path: (string) path to json file

Train the model on `num_steps` batches Args: model: (torch.nn.Module) the neural network optimizer: (torch.optim) optimizer for p

Train the model on `num_steps` batches Args: model: (torch.nn.Module) the neural network optimizer: (torch.optim) optimizer for p

Train the model and evaluate every epoch. Args: train_model_spec: (dict) contains the graph operations or nodes needed for training

Train the model and evaluate every epoch. Args: train_model_spec: (dict) contains the graph operations or nodes needed for training

Train the model and evaluate every epoch. Args: model: (torch.nn.Module) the neural network train_dataloader: (DataLoader) a torc

Train the model and evaluate every epoch. Args: model: (torch.nn.Module) the neural network train_data: (dict) training data with

Image preprocessing for training. Apply the following operations: - Horizontally flip the image with probability 1/2 - Apply rand

Train the model on `num_steps` batches Args: sess: (tf.Session) current session model_spec: (dict) contains the graph operations

Train the model on `num_steps` batches Args: sess: (tf.Session) current session model_spec: (dict) contains the graph operations

Loads parameters from json file

Fetch index idx image and labels from dataset. Perform transforms on image. Args: idx: (int) index in [0, 1, ..., size_o

We define an convolutional network that predicts the sign from an image. The components required are: - an embedding layer:

Store the filenames of the jpgs to use. Specifies transforms to apply on images. Args: data_dir: (string) directory cont

We define an recurrent network that predicts the NER tags for each token in the sentence. The components required are: - an

Loads dataset_params, vocabulary and tags. Ensure you have run `build_vocab.py` on data_dir before using this class. Args:

Compute the accuracy, given the outputs and labels for all images. Args: outputs: (np.ndarray) dimension batch_size x 6 - log softma

Compute the accuracy, given the outputs and labels for all tokens. Exclude PADding terms. Args: outputs: (np.ndarray) dimension batc

Gives dict-like access to Params instance by `params.dict['learning_rate']`

Gives dict-like access to Params instance by `params.dict['learning_rate']`

Gives dict-like access to Params instance by `params.dict['learning_rate']

Gives dict-like access to Params instance by `params.dict['learning_rate']

Fetches the DataLoader object for each type in types from data_dir. Args: types: (list) has one or more of 'train', 'val', 'test' de

This function defines how we use the components of our network to operate on an input batch. Args: s: (Variable) contain

This function defines how we use the components of our network to operate on an input batch. Args: s: (Variable) contain

Loads model parameters (state_dict) from file_path. If optimizer is provided, loads state_dict of optimizer assuming it is present in checkpoint.

Loads model parameters (state_dict) from file_path. If optimizer is provided, loads state_dict of optimizer assuming it is present in checkpoint.

Saves model and training parameters at checkpoint + 'last.pth.tar'. If is_best==True, also saves checkpoint + 'best.pth.tar' Args: st

Saves model and training parameters at checkpoint + 'last.pth.tar'. If is_best==True, also saves checkpoint + 'best.pth.tar' Args: st

Saves dict of floats in json file Args: d: (dict) of float-castable values (np.float, int, float, etc.) json_path: (string) path

Saves dict of floats in json file Args: d: (dict) of float-castable values (np.float, int, float, etc.) json_path: (string) path

Set the logger to log info in terminal and file `log_path`. In general, it is useful to have a logger so that every output to the terminal is sav

Set the logger to log info in terminal and file `log_path`. In general, it is useful to have a logger so that every output to the terminal is sav