This library is based on the Transformers library by HuggingFace. Simple Transformers lets you quickly train and evaluate Transformer models. Only 3 lines of code are needed to initialize a model, train the model, and evaluate a model.
Supports
- Sequence Classification
- Token Classification (NER)
- Question Answering
- Language Model Fine-Tuning
- Language Model Training
- Language Generation
- T5 Model
- Seq2Seq Tasks
- Multi-Modal Classification
- Conversational AI.
New documentation is now live at simpletransformers.ai
Currently added:
- Text classification
- NER
- Question answering
- Language model training
- Language model fine-tuning
- Training language models from scratch
Any feedback will be immensely helpful in improving the documentation! If you have any feedback, please leave a comment in the issue I've opened for this.
- T5 Model support added
- ELECTRA models can now be used with Language Model Training, Named Entity Recognition (Token Classification), Sequence Classification, and Question Answering.
- Simple Transformers
- Table of contents
- Setup
- Usage
- Text Classification
- Named Entity Recognition
- Question Answering
- Language Model Training
- Data format
- Minimal Example For Language Model Fine Tuning
- Minimal Example For Language Model Training From Scratch
- Minimal Example For Language Model Training With ELECTRA
- Real Dataset Example For Training a Language Model
- LanguageModelingModel
- Additional attributes for Language Modeling tasks
- config: dict
- generator_config: dict
- discriminator_config: dict
- Language Generation
- T5 Transformer
- Sequence-to-Sequence Models
- Conversational AI
- Multi-Modal Classification
- Regression
- Visualization Support
- Experimental Features
- Loading Saved Models
- Default Settings
- Args Explained
- output_dir: str
- cache_dir: str
- best_model_dir: str
- fp16: bool
- fp16_opt_level: str
- max_seq_length: int
- train_batch_size: int
- gradient_accumulation_steps: int
- eval_batch_size: int
- num_train_epochs: int
- weight_decay: float
- learning_rate: float
- adam_epsilon: float
- max_grad_norm: float
- do_lower_case: bool
- evaluate_during_training
- evaluate_during_training_steps
- evaluate_during_training_verbose
- use_cached_eval_features
- save_eval_checkpoints
- logging_steps: int
- save_steps: int
- no_cache: bool
- save_model_every_epoch: bool
- tensorboard_dir: str
- overwrite_output_dir: bool
- reprocess_input_data: bool
- process_count: int
- n_gpu: int
- silent: bool
- use_multiprocessing: bool
- wandb_project: str
- wandb_kwargs: dict
- use_early_stopping
- early_stopping_patience
- early_stopping_delta
- early_stopping_metric
- early_stopping_metric_minimize
- manual_seed
- encoding
- config
- Args Explained
- Current Pretrained Models
- Acknowledgements
- Contributors ✨
-
Install Anaconda or Miniconda Package Manager from here
-
Create a new virtual environment and install packages.
conda create -n transformers python pandas tqdm
conda activate transformers
If using cuda:
conda install pytorch cudatoolkit=10.1 -c pytorch
else:
conda install pytorch cpuonly -c pytorch
-
Install Apex if you are using fp16 training. Please follow the instructions here. (Installing Apex from pip has caused issues for several people.)
-
Install simpletransformers.
pip install simpletransformers
- Install Weights and Biases (wandb) for tracking and visualizing training in a web browser.
pip install wandb
Most available hyperparameters are common for all tasks. Any special hyperparameters will be listed in the docs section for the corresponding class. See Default Settings and Args Explained sections for more information.
Example scripts can be found in the examples
directory.
See the Changelog for up-to-date changes to the project.
The file structure has been updated starting with version 0.6.0. This should only affect import statements. The old import paths should still be functional although it is recommended to use the updated paths given below and in the minimal start examples.
simpletransformers.classification
- Includes all Classification models.ClassificationModel
MultiLabelClassificationModel
simpletransformers.ner
- Includes all Named Entity Recognition models.NERModel
simpletransformers.question_answering
- Includes all Question Answering models.QuestionAnsweringModel
Supports Binary Classification, Multiclass Classification, and Multilabel Classification.
Supported model types:
- ALBERT
- BERT
- CamemBERT
- RoBERTa
- DistilBERT
- ELECTRA
- FlauBERT
- XLM
- XLM-RoBERTa
- XLNet
- Set
'sliding_window': True
inargs
to prevent text being truncated. The default stride is'stride': 0.8
which is0.8 * max_seq_length
. Training text will be split using a sliding window and each window will be assigned the label from the original text. During evaluation and prediction, the mode of the predictions for each window will be the final prediction on each sample. Thetie_value
(default1
) will be used in the case of a tie.
Currently not available for Multilabel Classification
from simpletransformers.classification import ClassificationModel
import pandas as pd
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
# Train and Evaluation data needs to be in a Pandas Dataframe of two columns. The first column is the text with type str, and the second column is the label with type int.
train_data = [['Example sentence belonging to class 1', 1], ['Example sentence belonging to class 0', 0]]
train_df = pd.DataFrame(train_data)
eval_data = [['Example eval sentence belonging to class 1', 1], ['Example eval sentence belonging to class 0', 0]]
eval_df = pd.DataFrame(eval_data)
# Create a ClassificationModel
model = ClassificationModel('roberta', 'roberta-base') # You can set class weights by using the optional weight argument
# Train the model
model.train_model(train_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df)
If you wish to add any custom metrics, simply pass them as additional keyword arguments. The keyword is the name to be given to the metric, and the value is the function that will calculate the metric. Make sure that the function expects two parameters with the first one being the true label, and the second being the predictions. (This is the default for sklearn metrics)
import sklearn
result, model_outputs, wrong_predictions = model.eval_model(eval_df, acc=sklearn.metrics.accuracy_score)
To make predictions on arbitary data, the predict(to_predict)
function can be used. For a list of text, it returns the model predictions and the raw model outputs.
predictions, raw_outputs = model.predict(['Some arbitary sentence'])
For multiclass classification, simply pass in the number of classes to the num_labels
optional parameter of ClassificationModel
.
from simpletransformers.classification import ClassificationModel
import pandas as pd
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
# Train and Evaluation data needs to be in a Pandas Dataframe containing at least two columns. If the Dataframe has a header, it should contain a 'text' and a 'labels' column. If no header is present, the Dataframe should contain at least two columns, with the first column is the text with type str, and the second column in the label with type int.
train_data = [['Example sentence belonging to class 1', 1], ['Example sentence belonging to class 0', 0], ['Example eval senntence belonging to class 2', 2]]
train_df = pd.DataFrame(train_data)
eval_data = [['Example eval sentence belonging to class 1', 1], ['Example eval sentence belonging to class 0', 0], ['Example eval senntence belonging to class 2', 2]]
eval_df = pd.DataFrame(eval_data)
# Create a ClassificationModel
model = ClassificationModel('bert', 'bert-base-cased', num_labels=3, args={'reprocess_input_data': True, 'overwrite_output_dir': True})
# You can set class weights by using the optional weight argument
# Train the model
model.train_model(train_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df)
predictions, raw_outputs = model.predict(["Some arbitary sentence"])
For Multi-Label Classification, the labels should be multi-hot encoded. The number of classes can be specified (default is 2) by passing it to the num_labels
optional parameter of MultiLabelClassificationModel
.
Warning: Pandas can cause issues when saving and loading lists stored in a column. Check whether your list has been converted to a String!
The default evaluation metric used is Label Ranking Average Precision (LRAP) Score.
from simpletransformers.classification import MultiLabelClassificationModel
import pandas as pd
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
# Train and Evaluation data needs to be in a Pandas Dataframe containing at least two columns, a 'text' and a 'labels' column. The `labels` column should contain multi-hot encoded lists.
train_data = [['Example sentence 1 for multilabel classification.', [1, 1, 1, 1, 0, 1]]], [['This is another example sentence. ', [0, 1, 1, 0, 0, 0]]]
train_df = pd.DataFrame(train_data, columns=['text', 'labels'])
eval_data = [['Example eval sentence for multilabel classification.', [1, 1, 1, 1, 0, 1]], ['Example eval senntence belonging to class 2', [0, 1, 1, 0, 0, 0]]]
eval_df = pd.DataFrame(eval_data)
# Create a MultiLabelClassificationModel
model = MultiLabelClassificationModel('roberta', 'roberta-base', num_labels=6, args={'reprocess_input_data': True, 'overwrite_output_dir': True, 'num_train_epochs': 5})
# You can set class weights by using the optional weight argument
print(train_df.head())
# Train the model
model.train_model(train_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df)
print(result)
print(model_outputs)
predictions, raw_outputs = model.predict(['This thing is entirely different from the other thing. '])
print(predictions)
print(raw_outputs)
- The args dict of
MultiLabelClassificationModel
has an additionalthreshold
parameter with default value 0.5. The threshold is the value at which a given label flips from 0 to 1 when predicting. Thethreshold
may be a single value or a list of value with the same length as the number of labels. This enables the use of seperate threshold values for each label. MultiLabelClassificationModel
takes in an additional optional argumentpos_weight
. This should be a list with the same length as the number of labels. This enables using different weights for each label when calculating loss during training and evaluation.
- Training and evaluation Dataframes must contain a
text_a
,text_b
, and alabels
column. - The
predict()
function expects a list of lists in the format below. A single sample input should also be a list of lists like[[text_a, text_b]]
.
[
[sample_1_text_a, sample_1_text_b],
[sample_2_text_a, sample_2_text_b],
[sample_3_text_a, sample_3_text_b],
# More samples
]
from simpletransformers.classification import ClassificationModel
import pandas as pd
import sklearn
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
['Example sentence belonging to class 1', 'Yep, this is 1', 1],
['Example sentence belonging to class 0', 'Yep, this is 0', 0],
['Example 2 sentence belonging to class 0', 'Yep, this is 0', 0]
]
train_df = pd.DataFrame(train_data, columns=['text_a', 'text_b', 'labels'])
eval_data = [
['Example sentence belonging to class 1', 'Yep, this is 1', 1],
['Example sentence belonging to class 0', 'Yep, this is 0', 0],
['Example 2 sentence belonging to class 0', 'Yep, this is 0', 0]
]
eval_df = pd.DataFrame(eval_data, columns=['text_a', 'text_b', 'labels'])
train_args={
'reprocess_input_data': True,
'overwrite_output_dir': True,
'num_train_epochs': 3,
}
# Create a ClassificationModel
model = ClassificationModel('roberta', 'roberta-base', num_labels=2, use_cuda=True, cuda_device=0, args=train_args)
print(train_df.head())
# Train the model
model.train_model(train_df, eval_df=eval_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df, acc=sklearn.metrics.accuracy_score)
predictions, raw_outputs = model.predict([["I'd like to puts some CD-ROMS on my iPad, is that possible?'", "Yes, but wouldn't that block the screen?"]])
print(predictions)
print(raw_outputs)
- Yelp Reviews Dataset - Binary Classification
- AG News Dataset - Multiclass Classification
- Toxic Comments Dataset - Multilabel Classification
- Semantic Textual Similarity Benchmark - Sentence Pair
- AG News Dataset - BERT (base and distilled), RoBERTa (base and distilled), and XLNet compared
- Comparing ELECTRA, BERT, RoBERTa, and XLNET
class simpletransformers.classification.ClassificationModel (model_type, model_name, args=None, use_cuda=True)
This class is used for Text Classification tasks.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used.model_name
: model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.results
: A python dict of past evaluation results for the TransformerModel object.args
: A python dict of arguments used for training and evaluation.
cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
model_type
: (required) str - The type of model to use. Currently, BERT, XLNet, XLM, and RoBERTa models are available.model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models.num_labels
(optional): The number of labels or classes in the dataset.weight
(optional): A list of length num_labels containing the weights to assign to each label for loss calculation.args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values.use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used.
class methods
train_model(self, train_df, output_dir=None, show_running_loss=True, args=None, eval_df=None)
Trains the model using 'train_df'
Args:
-
train_df
: Pandas Dataframe containing at least two columns. If the Dataframe has a header, it should contain a 'text' and a 'labels' column. If no header is present, the Dataframe should contain at least two columns, with the first column containing the text, and the second column containing the label. The model will be trained on this Dataframe. -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
show_running_loss (optional): Set to False to disable printing running training loss to the terminal.
-
eval_df
(optional): A DataFrame against which evaluation will be performed whenevaluate_during_training
is enabled. Is required ifevaluate_during_training
is enabled. -
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). E.g. f1=sklearn.metrics.f1_score. A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
Returns:
- None
eval_model(self, eval_df, output_dir=None, verbose=False)
Evaluates the model on eval_df. Saves results to output_dir.
Args:
-
eval_df: Pandas Dataframe containing at least two columns. If the Dataframe has a header, it should contain a 'text' and a 'labels' column. If no header is present, the Dataframe should contain at least two columns, with the first column containing the text, and the second column containing the label. The model will be evaluated on this Dataframe.
-
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
verbose: If verbose, results will be printed to the console on completion of evaluation.
-
silent: If silent, tqdm progress bars will be hidden.
-
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). E.g. f1=sklearn.metrics.f1_score. A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
Returns:
-
result: Dictionary containing evaluation results. (Matthews correlation coefficient, tp, tn, fp, fn)
-
model_outputs: List of model outputs for each row in eval_df
-
wrong_preds: List of InputExample objects corresponding to each incorrect prediction by the model
predict(self, to_predict)
Performs predictions on a list of text.
Args:
- to_predict: A python list of text (str) to be sent to the model for prediction.
Returns:
- preds: A python list of the predictions (0 or 1) for each text.
- model_outputs: A python list of the raw model outputs for each text.
If config: {"output_hidden_states": True}
, two additional values will be returned.
- all_embedding_outputs: Numpy array of shape (batch_size, sequence_length, hidden_size)
- all_layer_hidden_states: Numpy array of shape (num_hidden_layers, batch_size, sequence_length, hidden_size)
train(self, train_dataset, output_dir)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_df, output_dir, prefix="")
Evaluates the model on eval_df. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, examples, evaluate=False)
Converts a list of InputExample objects to a TensorDataset containing InputFeatures. Caches the InputFeatures. Utility function for train() and eval() methods. Not intended to be used directly
compute_metrics(self, preds, labels, eval_examples, **kwargs):
Computes the evaluation metrics for the model predictions.
Args:
-
preds: Model predictions
-
labels: Ground truth labels
-
eval_examples: List of examples on which evaluation was performed
Returns:
-
result: Dictionary containing evaluation results. (Matthews correlation coefficient, tp, tn, fp, fn)
-
wrong: List of InputExample objects corresponding to each incorrect prediction by the model
This section describes how to use Simple Transformers for Named Entity Recognition. (If you are updating from a Simple Transformers before 0.5.0, note that seqeval
needs to be installed to perform NER.)
This model can also be used for any other NLP task involving token level classification. Make sure you pass in your list of labels to the model if they are different from the defaults.
Supported model types:
- BERT
- CamemBERT
- DistilBERT
- ELECTRA
- RoBERTa
- XLM-RoBERTa
model = NERModel('bert', 'bert-base-cased', labels=["LABEL_1", "LABEL_2", "LABEL_3"])
from simpletransformers.ner import NERModel
import pandas as pd
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
# Creating train_df and eval_df for demonstration
train_data = [
[0, 'Simple', 'B-MISC'], [0, 'Transformers', 'I-MISC'], [0, 'started', 'O'], [1, 'with', 'O'], [0, 'text', 'O'], [0, 'classification', 'B-MISC'],
[1, 'Simple', 'B-MISC'], [1, 'Transformers', 'I-MISC'], [1, 'can', 'O'], [1, 'now', 'O'], [1, 'perform', 'O'], [1, 'NER', 'B-MISC']
]
train_df = pd.DataFrame(train_data, columns=['sentence_id', 'words', 'labels'])
eval_data = [
[0, 'Simple', 'B-MISC'], [0, 'Transformers', 'I-MISC'], [0, 'was', 'O'], [1, 'built', 'O'], [1, 'for', 'O'], [0, 'text', 'O'], [0, 'classification', 'B-MISC'],
[1, 'Simple', 'B-MISC'], [1, 'Transformers', 'I-MISC'], [1, 'then', 'O'], [1, 'expanded', 'O'], [1, 'to', 'O'], [1, 'perform', 'O'], [1, 'NER', 'B-MISC']
]
eval_df = pd.DataFrame(eval_data, columns=['sentence_id', 'words', 'labels'])
# Create a NERModel
model = NERModel('bert', 'bert-base-cased', args={'overwrite_output_dir': True, 'reprocess_input_data': True})
# Train the model
model.train_model(train_df)
# Evaluate the model
result, model_outputs, predictions = model.eval_model(eval_df)
# Predictions on arbitary text strings
predictions, raw_outputs = model.predict(["Some arbitary sentence"])
print(predictions)
class simpletransformers.ner.ner_model.NERModel (model_type, model_name, labels=None, args=None, use_cuda=True)
This class is used for Named Entity Recognition.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.results
: A python dict of past evaluation results for the TransformerModel object.args
: A python dict of arguments used for training and evaluation.
cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
model_type
: (required) str - The type of model to use. Currently, BERT, XLNet, XLM, and RoBERTa models are available.model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models.labels
(optional): A list of all Named Entity labels. If not given, ["O", "B-MISC", "I-MISC", "B-PER", "I-PER", "B-ORG", "I-ORG", "B-LOC", "I-LOC"] will be used.args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values.use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used.
class methods
train_model(self, train_data, output_dir=None, args=None, eval_df=None)
Trains the model using 'train_data'
Args:
-
train_data: train_data should be the path to a .txt file containing the training data OR a pandas DataFrame with 3 columns. If a text file is used the data should be in the CoNLL format. i.e. One word per line, with sentences seperated by an empty line. The first word of the line should be a word, and the last should be a Name Entity Tag. If a DataFrame is given, each sentence should be split into words, with each word assigned a tag, and with all words from the same sentence given the same sentence_id.
-
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
show_running_loss (optional): Set to False to prevent running loss from being printed to console. Defaults to True.
-
args (optional): Optional changes to the args dict of the model. Any changes made will persist for the model.
-
eval_df (optional): A DataFrame against which evaluation will be performed when
evaluate_during_training
is enabled. Is required ifevaluate_during_training
is enabled.
Returns:
- None
eval_model(self, eval_data, output_dir=None, verbose=True)
Evaluates the model on eval_data. Saves results to output_dir.
Args:
-
eval_data: eval_data should be the path to a .txt file containing the evaluation data or a pandas DataFrame. If a text file is used the data should be in the CoNLL format. I.e. One word per line, with sentences seperated by an empty line. The first word of the line should be a word, and the last should be a Name Entity Tag. If a DataFrame is given, each sentence should be split into words, with each word assigned a tag, and with all words from the same sentence given the same sentence_id.
-
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
verbose: If verbose, results will be printed to the console on completion of evaluation.
Returns:
-
result: Dictionary containing evaluation results. (eval_loss, precision, recall, f1_score)
-
model_outputs: List of raw model outputs
-
preds_list: List of predicted tags
predict(self, to_predict)
Performs predictions on a list of text.
Args:
- to_predict: A python list of text (str) to be sent to the model for prediction.
Returns:
- preds: A Python list of lists with dicts containg each word mapped to its NER tag.
- model_outputs: A Python list of lists with dicts containing each word mapped to its list with raw model output.
train(self, train_dataset, output_dir)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_dataset, output_dir, prefix="")
Evaluates the model on eval_dataset. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, data, evaluate=False, no_cache=False, to_predict=None)
Converts a list of InputExample objects to a TensorDataset containing InputFeatures. Caches the InputFeatures. Utility function for train() and eval() methods. Not intended to be used directly
Supported model types:
- ALBERT
- BERT
- DistilBERT
- ELECTRA
- XLM
- XLNet
For question answering tasks, the input data can be in JSON files or in a Python list of dicts in the correct format.
The file should contain a single list of dictionaries. A dictionary represents a single context and its associated questions.
Each such dictionary contains two attributes, the "context"
and "qas"
.
context
: The paragraph or text from which the question is asked.qas
: A list of questions and answers.
Questions and answers are represented as dictionaries. Each dictionary in qas
has the following format.
id
: (string) A unique ID for the question. Should be unique across the entire dataset.question
: (string) A question.is_impossible
: (bool) Indicates whether the question can be answered correctly from the context.answers
: (list) The list of correct answers to the question.
A single answer is represented by a dictionary with the following attributes.
answer
: (string) The answer to the question. Must be a substring of the context.answer_start
: (int) Starting index of the answer in the context.
from simpletransformers.question_answering import QuestionAnsweringModel
import json
import os
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
# Create dummy data to use for training.
train_data = [
{
'context': "This is the first context",
'qas': [
{
'id': "00001",
'is_impossible': False,
'question': "Which context is this?",
'answers': [
{
'text': "the first",
'answer_start': 8
}
]
}
]
},
{
'context': "Other legislation followed, including the Migratory Bird Conservation Act of 1929, a 1937 treaty prohibiting the hunting of right and gray whales,
and the Bald Eagle Protection Act of 1940. These later laws had a low cost to society—the species were relatively rare—and little opposition was raised",
'qas': [
{
'id': "00002",
'is_impossible': False,
'question': "What was the cost to society?",
'answers': [
{
'text': "low cost",
'answer_start': 225
}
]
},
{
'id': "00003",
'is_impossible': False,
'question': "What was the name of the 1937 treaty?",
'answers': [
{
'text': "Bald Eagle Protection Act",
'answer_start': 167
}
]
}
]
}
]
# Save as a JSON file
os.makedirs('data', exist_ok=True)
with open('data/train.json', 'w') as f:
json.dump(train_data, f)
# Create the QuestionAnsweringModel
model = QuestionAnsweringModel('distilbert', 'distilbert-base-uncased-distilled-squad', args={'reprocess_input_data': True, 'overwrite_output_dir': True})
# Train the model with JSON file
model.train_model('data/train.json')
# The list can also be used directly
# model.train_model(train_data)
# Evaluate the model. (Being lazy and evaluating on the train data itself)
result, text = model.eval_model('data/train.json')
print(result)
print(text)
print('-------------------')
# Making predictions using the model.
to_predict = [{'context': 'This is the context used for demonstrating predictions.', 'qas': [{'question': 'What is this context?', 'id': '0'}]}]
print(model.predict(to_predict))
class simpletransformers.question_answering.QuestionAnsweringModel (model_type, model_name, args=None, use_cuda=True)
This class is used for Question Answering tasks.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.results
: A python dict of past evaluation results for the TransformerModel object.args
: A python dict of arguments used for training and evaluation.cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
model_type
: (required) str - The type of model to use.model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models.args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values.use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used.
class methods
train_model(self, train_df, output_dir=None, args=None, eval_df=None)
Trains the model using 'train_file'
Args:
-
train_df
: Path to JSON file containing training data. The model will be trained on this file. output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
show_running_loss
(Optional): Set to False to prevent training loss being printed. -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
eval_file
(optional): Path to JSON file containing evaluation data against which evaluation will be performed when evaluate_during_training is enabled. Is required if evaluate_during_training is enabled. -
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
Returns:
- None
eval_model(self, eval_df, output_dir=None, verbose=False)
Evaluates the model on eval_file. Saves results to output_dir.
Args:
-
eval_file
: Path to JSON file containing evaluation data. The model will be evaluated on this file. -
output_dir
: The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
verbose
: If verbose, results will be printed to the console on completion of evaluation. -
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
Returns:
-
result
: Dictionary containing evaluation results. (correct, similar, incorrect) -
text
: A dictionary containing the 3 dictionaries correct_text, similar_text (the predicted answer is a substring of the correct answer or vise versa), incorrect_text.
predict(self, to_predict)
Performs predictions on a list of text.
Args:
to_predict
: A python list of python dicts containing contexts and questions to be sent to the model for prediction.
E.g: predict([
{
'context': "Some context as a demo",
'qas': [
{'id': '0', 'question': 'What is the context here?'},
{'id': '1', 'question': 'What is this for?'}
]
}
])
n_best_size
(Optional): Number of predictions to return. args['n_best_size'] will be used if not specified.
Returns:
preds
: A python list containg the predicted answer, and id for each question in to_predict.
train(self, train_dataset, output_dir, show_running_loss=True, eval_file=None)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_df, output_dir, , verbose=False)
Evaluates the model on eval_df. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, examples, evaluate=False, no_cache=False, output_examples=False)
Converts a list of InputExample objects to a TensorDataset containing InputFeatures. Caches the InputFeatures. Utility function for train() and eval() methods. Not intended to be used directly
QuestionAnsweringModel has a few additional attributes in its args
dictionary, given below with their default values.
'doc_stride': 384,
'max_query_length': 64,
'n_best_size': 20,
'max_answer_length': 100,
'null_score_diff_threshold': 0.0
When splitting up a long document into chunks, how much stride to take between chunks.
Maximum token length for questions. Any questions longer than this will be truncated to this length.
The number of predictions given per question.
The maximum token length of an answer that can be generated.
If null_score - best_non_null is greater than the threshold predict null.
Supported model types:
- BERT
- CamemBERT
- DistilBERT
- ELECTRA
- GPT-2
- OpenAI-GPT
- RoBERTa
The data should simply be placed in a text file. E.g.: WikiText-2
The minimal example given below assumes that you have downloaded the WikiText-2 dataset.
from simpletransformers.language_modeling import LanguageModelingModel
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
}
model = LanguageModelingModel('bert', 'bert-base-cased', args=train_args)
model.train_model("wikitext-2/wiki.train.tokens", eval_file="wikitext-2/wiki.test.tokens")
model.eval_model("wikitext-2/wiki.test.tokens")
You can use any text file/files for training a new language model. Setting model_name
to None
will indicate that the language model should be trained from scratch.
Required for Language Model Training From Scratch:
train_files
must be specifief when creating theLanguagModelingModel
. This may be a path to a single file or a list of paths to multiple files.vocab_size
(in args dictionary)
from simpletransformers.language_modeling import LanguageModelingModel
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"vocab_size": 52000,
}
model = LanguageModelingModel('roberta', None, args=train_args)
model.train_model("wikitext-2/wiki.train.tokens", eval_file="wikitext-2/wiki.test.tokens")
model.eval_model("wikitext-2/wiki.test.tokens")
ELECTRA is a new approach to pretraining Transformer Language Models. This method is comparatively less compute-intensive.
You can use the save_discriminator()
and save_generator()
methods to extract the pretrained models. The two models will be saved to <output_dir>/discriminator_model
and <output_dir>/generator_model
by default.
from simpletransformers.language_modeling import LanguageModelingModel
import logging
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"vocab_size": 52000,
}
model = LanguageModelingModel('electra', None, args=train_args, train_files="wikitext-2/wiki.train.tokens")
# Mixing standard ELECTRA architectures example
# model = LanguageModelingModel(
# "electra",
# None,
# generator_name="google/electra-small-generator",
# discriminator_name="google/electra-large-discriminator",
# args=train_args,
# train_files="wikitext-2/wiki.train.tokens",
# )
model.train_model("wikitext-2/wiki.train.tokens", eval_file="wikitext-2/wiki.test.tokens")
model.eval_model("wikitext-2/wiki.test.tokens")
class simpletransformers.language_modeling.LanguageModelingModel (model_type, model_name, generator_name=None, discriminator_name=None, args=None, use_cuda=True, cuda_device=-1)
This class is used for language modeling tasks.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.results
: A python dict of past evaluation results for the TransformerModel object.args
: A python dict of arguments used for training and evaluation.cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
model_type
: (required) str - The type of model to use.model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models. Set toNone
for language model training from scratch.generator_name
: (optional) A pretrained model name or path to a directory containing an ELECTRA generator model.discriminator_name
: (optional) A pretrained model name or path to a directory containing an ELECTRA discriminator model.args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values.train_files
: (optional) List of files to be used when training the tokenizer.use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used.cuda_device
: (optional) Specific GPU that should be used. Will use the first available GPU by default.
class methods
train_model(self, train_file, output_dir=None, show_running_loss=True, args=None, eval_file=None, verbose=True,)
Trains the model using 'train_file'
Args:
-
train_file
: Path to text file containing the text to train the language model on. -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
show_running_loss
(Optional): Set to False to prevent training loss being printed. -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
eval_file
(optional): Path to eval file containing the text to evaluate the language model on. Is required if evaluate_during_training is enabled.
Returns:
- None
eval_model(self, eval_file, output_dir=None, verbose=True, silent=False,)
Evaluates the model on eval_file. Saves results to output_dir.
Args:
-
eval_file
: Path to eval file containing the text to evaluate the language model on. -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
verbose
: If verbose, results will be printed to the console on completion of evaluation. -
silent
: If silent, tqdm progress bars will be hidden.
Returns:
-
result
: Dictionary containing evaluation results. (correct, similar, incorrect) -
text
: A dictionary containing the 3 dictionaries correct_text, similar_text (the predicted answer is a substring of the correct answer or vise versa), incorrect_text.
**train_tokenizer(self, train_files, tokenizer_name=None, output_dir=None, use_trained_tokenizer=True)
Train a new tokenizer on train_files
.
Args:
-
train_files
: List of files to be used when training the tokenizer. -
tokenizer_name
: Name of a pretrained tokenizer or a path to a directory containing a tokenizer. -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
use_trained_tokenizer
(optional): Load the trained tokenizer once training completes.
Returns: None
train(self, train_dataset, output_dir, show_running_loss=True, eval_file=None)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_dataset, output_dir, , verbose=False)
Evaluates the model on eval_dataset. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, examples, evaluate=False, no_cache=False, output_examples=False)
Reads a text file from file_path and creates training features. Utility function for train() and eval() methods. Not intended to be used directly
LanguageModelingModel has a few additional attributes in its args
dictionary, given below with their default values.
"dataset_type": "None",
"dataset_class": None,
"custom_tokenizer": None,
"block_size": 512,
"mlm": True,
"mlm_probability": 0.15,
"max_steps": -1,
"config_name": None,
"tokenizer_name": None,
"min_frequency": 2,
"special_tokens": ["<s>", "<pad>", "</s>", "<unk>", "<mask>"],
"sliding_window": False,
"stride": 0.8
"config": {},
"generator_config": {},
"discriminator_config": {},
Used to specify the Dataset type to be used. The choices are given below.
-
simple
(or None) - Each line in the train files are considered to be a single, separate sample.sliding_window
can be set to True to automatically split longer sequences into samples of lengthmax_seq_length
. Uses multiprocessing for significantly improved performance on multicore systems. -
line_by_line
- Treats each line in the train files as a seperate sample. -
text
- Treats each file intrain_files
as a seperate sample.
Using simple
is recommended.
A custom dataset class to use.
Optional input sequence length after tokenization. The training dataset will be truncated in block of this size for training. Default to the model max input length for single sentence inputs (take into account special tokens).
Train with masked-language modeling loss instead of language modeling
Ratio of tokens to mask for masked language modeling loss
If > 0: set total number of training steps to perform. Override num_train_epochs.
Name of pretrained config or path to a directory containing a config.json
file.
Name of pretrained tokenizer or path to a directory containing tokenizer files.
Minimum frequency required for a word to be added to the vocabulary.
List of special tokens to be used when training a new tokenizer.
Whether sliding window technique should be used when preparing data. Only works with SimpleDataset.
A fraction of the max_seq_length
to use as the stride when using a sliding window
Key-values given here will override the default values used in a model Config
.
Key-values given here will override the default values used in an Electra generator model Config
.
Key-values given here will override the default values used in an Electra discriminator model Config
.
This section describes how to use Simple Transformers for Langauge Generation.
Supported model types:
- CTRL
- GPT-2
- OpenAI-GPT
- Transformer-XL
- XLM
- XLNet
import logging
from simpletransformers.language_generation import LanguageGenerationModel
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
model = LanguageGenerationModel("gpt2", "gpt2")
model.generate("Let's give a minimal start to the model like")
class simpletransformers.language_generation.language_generation_model.LanguageGenerationModel (self, model_type, model_name, args=None, use_cuda=True, cuda_device=-1, **kwargs)
This class is used for Language Generation.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.args
: A python dict of arguments used for training and evaluation.
cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
-
model_type
: (required) str - The type of model to use. -
model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models. -
args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values. -
use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used. -
cuda_device
(optional): Specific GPU that should be used. Will use the first available GPU by default. -
**kwargs
(optional): For providing proxies, force_download, resume_download, cache_dir and other options specific to the 'from_pretrained' implementation where this will be supplied.
class methods
generate(self, prompt=None, args=None, verbose=True)
Generate text using a LanguageGenerationModel
Args:
-
prompt
(optional): A prompt text for the model. If given, will override args["prompt"] -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
verbose
(optional): If verbose, generated text will be logged to the console.
Returns:
generated_sequences
: Sequences of text generated by the model.
LanguageGenerationModel has a few additional attributes in its args
dictionary, given below with their default values.
"do_sample": True,
"prompt": "",
"length": 20,
"stop_token": None,
"temperature": 1.0,
"repetition_penalty": 1.0,
"k": 0,
"p": 0.9,
"padding_text": "",
"xlm_language": "",
"num_return_sequences": 1,
"config_name": None,
"tokenizer_name": None,
If set to False
greedy decoding is used. Otherwise sampling is used. Defaults to False
as defined in configuration_utils.PretrainedConfig
.
A prompt text for the model.
Length of the text to generate
Token at which text generation is stopped
Temperature of 1.0 is the default. Lowering this makes the sampling greedier
Primarily useful for CTRL model; in that case, use 1.2
k value for top-k sampling
p value for top-p (nucleus) sampling
Padding text for Transfo-XL and XLNet.
Optional language when used with the XLM model.
The number of samples to generate.
Key-values given here will override the default values used in a model Config.
T5 model seems to be working fine, but please open an issue if you run across any problems
The T5 Transformer is an Encoder-Decoder architecture where both the input and targets are text sequences. The task that should be performed on the input is defined by a prefix. This means that the same T5 model can perform multiple tasks.
You can train the T5 model on a completely new task by simply using a new prefix
.
The input to a T5 model has the following pattern;
"<prefix>: <input_text> </s>"
The label sequence has the following pattern;
"<target_sequence> </s>"
The inputs to both the train_model()
and eval_model()
methods should be a Pandas DataFrame containing the 3 columns - prefix
, input_text
, target_text
.
prefix
: A string indicating the task to perform. (E.g."question"
,"stsb"
)input_text
: The input text sequence.prefix
is automatically prepended to form the full input. (<prefix>: <input_text>
)target_text
: The target sequence
If preprocess_inputs
is set to True
in the model args
, then the < /s>
tokens (including preceeding space) and the :
(prefix separator including trailing separator) between prefix
and input_text
are automatically added. Otherwise, the input DataFrames must contain the < /s>
tokens (including preceeding space) and the :
(prefix separator including trailing separator).
The prediction data should be a list of strings with the prefix
and the :
(prefix separator) included.
If preprocess_inputs
is set to True
in the model args
, then the < /s>
token (including preceeding space) is automatically added to each string in the list. Otherwise, the strings must have the < /s>
(including preceeding space) must be included.
import logging
import pandas as pd
from simpletransformers.t5 import T5Model
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
["convert", "one", "1"],
["convert", "two", "2"],
]
train_df = pd.DataFrame(train_data, columns=["prefix", "input_text", "target_text"])
eval_data = [
["convert", "three", "3"],
["convert", "four", "4"],
]
eval_df = pd.DataFrame(eval_data, columns=["prefix", "input_text", "target_text"])
eval_df = train_df.copy()
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 10,
"train_batch_size": 2,
"num_train_epochs": 200,
}
# Create T5 Model
model = T5Model("t5-base", args=model_args)
# Train T5 Model on new task
model.train_model(train_df)
# Evaluate T5 Model on new task
results = model.eval_model(eval_df)
# Predict with trained T5 model
print(model.predict(["convert: four"]))
You can evaluate the models' generated sequences using custom metric functions (including evaluation during training). However, due to the way T5 outputs are generated, this may be significantly slower than evaluation with other models.
Note, you must set evaluate_generated_text
to True
to evaluate generated sequences.
import logging
import pandas as pd
from simpletransformers.t5 import T5Model
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
["convert", "one", "1"],
["convert", "two", "2"],
]
train_df = pd.DataFrame(train_data, columns=["prefix", "input_text", "target_text"])
eval_data = [
["convert", "three", "3"],
["convert", "four", "4"],
]
eval_df = pd.DataFrame(eval_data, columns=["prefix", "input_text", "target_text"])
eval_df = train_df.copy()
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 10,
"train_batch_size": 2,
"num_train_epochs": 200,
"save_eval_checkpoints": False,
"save_model_every_epoch": False,
# "silent": True,
"evaluate_generated_text": True,
"evaluate_during_training": True,
"evaluate_during_training_verbose": True,
}
model = T5Model("t5-base", args=model_args)
def count_matches(labels, preds):
print(labels)
print(preds)
return sum([1 if label == pred else 0 for label, pred in zip(labels, preds)])
model.train_model(train_df, eval_data=eval_df, matches=count_matches)
print(model.eval_model(eval_df, matches=count_matches))
class simpletransformers.t5.t5_model.T5Model (self, model_name, args=None, use_cuda=True, cuda_device=-1, **kwargs)
This class is used for the T5 Transformer.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.args
: A python dict of arguments used for training and evaluation.
cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
-
model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models. -
args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values. -
cuda_device
: (optional) Specific GPU that should be used. Will use the first available GPU by default. -
use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used. -
**kwargs
: (optional) For providing proxies, force_download, resume_download, cache_dir and other options specific to the 'from_pretrained' implementation where this will be supplied.
class methods
train_model(self, train_data, output_dir=None, show_running_loss=True, args=None, eval_df=None)
Trains the model using 'train_data'
Args:
-
train_data
: Pandas DataFrame containing the 3 columns -prefix
,input_text
,target_text
. -prefix
: A string indicating the task to perform. (E.g."question"
,"stsb"
) -input_text
: The input text sequence.prefix
is automatically prepended to form the full input. (<prefix>: <input_text>
) -target_text
: The target sequence -
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
show_running_loss (optional): Set to False to disable printing running training loss to the terminal.
-
eval_data
(optional): A DataFrame against which evaluation will be performed whenevaluate_during_training
is enabled. Is required ifevaluate_during_training
is enabled. -
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down training significantly as the predicted sequences need to be generated.
Returns:
- None
eval_model(self, eval_data, output_dir=None, verbose=True, silent=False)
Evaluates the model on eval_data. Saves results to output_dir.
Args:
-
eval_data: Pandas DataFrame containing the 3 columns -
prefix
,input_text
,target_text
.prefix
: A string indicating the task to perform. (E.g."question"
,"stsb"
)input_text
: The input text sequence.prefix
is automatically prepended to form the full input. (<prefix>: <input_text>
)target_text
: The target sequence
-
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
verbose: If verbose, results will be printed to the console on completion of evaluation.
-
silent: If silent, tqdm progress bars will be hidden.
-
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down evaluation significantly as the predicted sequences need to be generated.
Returns:
- result: Dictionary containing evaluation results.
predict(self, to_predict)
Performs predictions on a list of text.
Args:
- to_predict: A python list of text (str) to be sent to the model for prediction. Note that the prefix should be prepended to the text.
Returns:
- preds: A python list of the generated sequences.
train(self, train_dataset, output_dir)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_dataset, output_dir, prefix="")
Evaluates the model on eval_dataset. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, examples, evaluate=False)
Creates a T5Dataset
from data.
Utility function for train() and eval() methods. Not intended to be used directly
compute_metrics(self, preds, labels, **kwargs):
Computes the evaluation metrics for the model predictions.
Args:
labels
: List of target sequencespreds
: List of model generated outputs**kwargs
: Custom metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down evaluation significantly as the predicted sequences need to be generated.
Returns:
- result: Dictionary containing evaluation results.
T5Model
has a few additional attributes in its args
dictionary, given below with their default values.
{
"dataset_class": None,
"do_sample": False,
"max_steps": -1,
"evaluate_generated_text": False,
"num_beams": 1,
"max_length": 20,
"repetition_penalty": 1.0,
"length_penalty": 2.0,
"early_stopping": True,
"preprocess_inputs": True,
}
A custom dataset class to use.
If set to False
greedy decoding is used. Otherwise sampling is used. Defaults to False
as defined in configuration_utils.PretrainedConfig
.
Maximum number of training steps. Will override the effect of num_train_epochs
.
Generate sequences for evaluation.
Number of beams for beam search. Must be between 1 and infinity. 1 means no beam search. Default to 1.
The max length of the sequence to be generated. Between min_length
and infinity. Default to 20.
The parameter for repetition penalty. Between 1.0 and infinity. 1.0 means no penalty. Default to 1.0.
Exponential penalty to the length. Default to 1.
if set to True
beam search is stopped when at least num_beams
sentences finished per batch.
Automatically add :
and < /s>
tokens to train_model()
and eval_model()
inputs. Automatically add < /s>
to each string in to_predict
in predict()
.
These models are Sequence-to-Sequence models (Seq2SeqModel
) where both the input and targets are text sequences. For example, translation and summarization are sequence-to-sequence tasks.
Currently, three main types of Sequence-to-Sequence models are available.
- BART (Summarization)
- Marian (Translation)
- Encoder-Decoder (Generic)
Note that these models are not restricted to the specifed task. The task is merely given as a starting point.
Commonly used for summarization tasks.
Commonly used for translation tasks.
Encoder-Decoder is a generic type of Sequence-to-Sequence model and it can be configured with different Encoder-Decoder combinations.
There is a known issue with loading saved Encoder-Decoder models. The loaded model seems to underperform compared to the model that was saved.
The following rules currently apply to Encoder-Decoder models:
- The decoder must be a
bert
model. - The encoder can be one of
[bert, roberta, distilbert, camembert, electra]
. - The encodr and the decoder must be of the same "size". (E.g.
roberta-base
encoder and abert-base-uncased
decoder)
The inputs to both the train_model()
and eval_model()
methods should be a Pandas DataFrame containing the 2 columns - input_text
and target_text
.
input_text
: The input text sequence.target_text
: The target text sequence.
The prediction data should be a list of strings.
The Seq2SeqModel
must be initialized with encoder_decoder_type="bart"
and encoder_decoder_name
set to a pre-trained model name or the path to a saved model directory.
import logging
import pandas as pd
from simpletransformers.seq2seq import Seq2SeqModel
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
["one", "1"],
["two", "2"],
]
train_df = pd.DataFrame(train_data, columns=["input_text", "target_text"])
eval_data = [
["three", "3"],
["four", "4"],
]
eval_df = pd.DataFrame(eval_data, columns=["input_text", "target_text"])
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 10,
"train_batch_size": 2,
"num_train_epochs": 10,
"save_eval_checkpoints": False,
"save_model_every_epoch": False,
"evaluate_during_training": True,
"evaluate_generated_text": True,
"evaluate_during_training_verbose": True,
"use_multiprocessing": False,
"max_length": 15,
"manual_seed": 4,
}
# Initialize model
model = Seq2SeqModel(
encoder_decoder_type="bart",
encoder_decoder_name="bart-large",
args=model_args,
)
# Train the model
model.train_model(train_df)
# Evaluate the model
results = model.eval_model(eval_df)
# Use the model for prediction
print(model.predict(["five"]))
# Load a saved model
model1 = Seq2SeqModel(
encoder_decoder_type="bart",
encoder_decoder_name="outputs",
args=model_args,
)
print(model1.predict(["five"]))
The Seq2SeqModel
must be initialized with encoder_decoder_type="marian"
and encoder_decoder_name
set to a pre-trained model name or the path to a saved model directory.
Everything else is identical to the Bart model usage.
import logging
import pandas as pd
from simpletransformers.seq2seq import Seq2SeqModel
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 50,
"train_batch_size": 2,
"num_train_epochs": 10,
"save_eval_checkpoints": False,
"save_model_every_epoch": False,
"evaluate_generated_text": True,
"evaluate_during_training_verbose": True,
"use_multiprocessing": False,
"max_length": 50,
"manual_seed": 4,
}
model = Seq2SeqModel(
encoder_decoder_type="marian",
encoder_decoder_name="Helsinki-NLP/opus-mt-en-de",
args=model_args,
)
src = [
"People say nothing is impossible, but I do nothing every day.",
"My opinions may have changed, but not the fact that I'm right.",
"He who laughs last didn't get the joke.",
]
predictions = model.predict(src)
for en, de in zip(src, predictions):
print("-------------")
print(en)
print(de)
print()
import logging
import pandas as pd
from simpletransformers.seq2seq import Seq2SeqModel
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
["one", "1"],
["two", "2"],
]
train_df = pd.DataFrame(train_data, columns=["input_text", "target_text"])
eval_data = [
["three", "3"],
["four", "4"],
]
eval_df = pd.DataFrame(eval_data, columns=["input_text", "target_text"])
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 10,
"train_batch_size": 2,
"num_train_epochs": 10,
"save_eval_checkpoints": False,
"save_model_every_epoch": False,
"evaluate_generated_text": True,
"evaluate_during_training_verbose": True,
"use_multiprocessing": False,
"max_length": 15,
"manual_seed": 4,
}
encoder_type = "roberta"
model = Seq2SeqModel(
encoder_type,
"roberta-base",
"bert-base-cased",
args=model_args,
use_cuda=True,
)
model.train_model(train_df)
results = model.eval_model(eval_df)
print(model.predict(["five"]))
model1 = Seq2SeqModel(
encoder_type,
encoder_decoder_name="outputs",
args=model_args,
use_cuda=True,
)
print(model1.predict(["five"]))
You can evaluate the models' generated sequences using custom metric functions (including evaluation during training). However, this may be significantly slower than evaluation with other models.
Note, you must set evaluate_generated_text
to True
to evaluate generated sequences.
import logging
import pandas as pd
from simpletransformers.t5 import T5Model
logging.basicConfig(level=logging.INFO)
transformers_logger = logging.getLogger("transformers")
transformers_logger.setLevel(logging.WARNING)
train_data = [
["convert", "one", "1"],
["convert", "two", "2"],
]
train_df = pd.DataFrame(train_data, columns=["prefix", "input_text", "target_text"])
eval_data = [
["convert", "three", "3"],
["convert", "four", "4"],
]
eval_df = pd.DataFrame(eval_data, columns=["prefix", "input_text", "target_text"])
eval_df = train_df.copy()
model_args = {
"reprocess_input_data": True,
"overwrite_output_dir": True,
"max_seq_length": 10,
"train_batch_size": 2,
"num_train_epochs": 200,
"save_eval_checkpoints": False,
"save_model_every_epoch": False,
"evaluate_generated_text": True,
"evaluate_during_training": True,
"evaluate_during_training_verbose": True,
}
# Initialize model
model = Seq2SeqModel(
encoder_decoder_type="bart",
encoder_decoder_name="bart-large",
args=model_args,
use_cuda=True,
)
def count_matches(labels, preds):
print(labels)
print(preds)
return sum([1 if label == pred else 0 for label, pred in zip(labels, preds)])
# Train the model
model.train_model(train_df, eval_data=eval_df, matches=count_matches)
class simpletransformers.seq2seq.seq2seq_model.Seq2SeqModel (self, encoder_type=None, encoder_name=None, decoder_name=None, encoder_decoder_type=None, encoder_decoder_name=None, config=None, args=None, use_cuda=True, cuda_device=-1, **kwargs)
This class is used for the T5 Transformer.
Class attributes
encoder_tokenizer
: The tokenizer to be used with the encoder model.decoder_tokenizer
: The tokenizer to be used with the decoder model.model
: The model to be used.device
: The device on which the model will be trained and evaluated.args
: A python dict of arguments used for training and evaluation.cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
Parameters
-
encoder_type
: (optional) str - The type of model to use as the encoder. -
encoder_name
: (optional) str - The exact model to use as the encoder. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models. -
decoder_name
: (optional) str - The exact model to use as the decoder. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models. -
encoder_decoder_type
: (optional) str - The type of encoder-decoder model. (E.g. bart) -
encoder_decoder_name
: (optional) str - The path to a directory containing the saved encoder and decoder of a Seq2SeqModel. (E.g. "outputs/") OR a valid BART model OR a valid Marian model. -
args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values. -
cuda_device
: (optional) Specific GPU that should be used. Will use the first available GPU by default. -
use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used. -
**kwargs
: (optional) For providing proxies, force_download, resume_download, cache_dir and other options specific to the 'from_pretrained' implementation where this will be supplied.
class methods
train_model(self, train_data, output_dir=None, show_running_loss=True, args=None, eval_df=None)
Trains the model using 'train_data'
Args:
-
train_data
: Pandas DataFrame containing the 2 columns -input_text
,target_text
.input_text
: The input text sequence.target_text
: The target sequence
-
output_dir
(optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used. -
args
(optional): Optional changes to the args dict of the model. Any changes made will persist for the model. -
show_running_loss (optional): Set to False to disable printing running training loss to the terminal.
-
eval_data
(optional): A DataFrame against which evaluation will be performed whenevaluate_during_training
is enabled. Is required ifevaluate_during_training
is enabled. -
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down training significantly as the predicted sequences need to be generated.
Returns:
- None
eval_model(self, eval_data, output_dir=None, verbose=True, silent=False)
Evaluates the model on eval_data. Saves results to output_dir.
Args:
-
eval_data: Pandas DataFrame containing the 2 columns -
input_text
,target_text
.input_text
: The input text sequence.target_text
: The target sequence
-
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
verbose: If verbose, results will be printed to the console on completion of evaluation.
-
silent: If silent, tqdm progress bars will be hidden.
-
**kwargs
: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down evaluation significantly as the predicted sequences need to be generated.
Returns:
- result: Dictionary containing evaluation results.
predict(self, to_predict)
Performs predictions on a list of text.
Args:
- to_predict: A python list of text (str) to be sent to the model for prediction. Note that the prefix should be prepended to the text.
Returns:
- preds: A python list of the generated sequences.
train(self, train_dataset, output_dir)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_dataset, output_dir, prefix="")
Evaluates the model on eval_dataset. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, examples, evaluate=False)
Creates a T5Dataset
from data.
Utility function for train() and eval() methods. Not intended to be used directly
compute_metrics(self, preds, labels, **kwargs):
Computes the evaluation metrics for the model predictions.
Args:
labels
: List of target sequencespreds
: List of model generated outputs**kwargs
: Custom metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions. Both inputs will be lists of strings. Note that this will slow down evaluation significantly as the predicted sequences need to be generated.
Returns:
- result: Dictionary containing evaluation results.
Seq2SeqModel
has a few additional attributes in its args
dictionary, given below with their default values.
{
"dataset_class": None,
"do_sample": False,
"max_steps": -1,
"evaluate_generated_text": False,
"num_beams": 1,
"max_length": 20,
"repetition_penalty": 1.0,
"length_penalty": 2.0,
"early_stopping": True,
}
A custom dataset class to use.
If set to False
greedy decoding is used. Otherwise sampling is used. Defaults to False
as defined in configuration_utils.PretrainedConfig
.
Maximum number of training steps. Will override the effect of num_train_epochs
.
Generate sequences for evaluation.
Number of beams for beam search. Must be between 1 and infinity. 1 means no beam search. Default to 1.
The max length of the sequence to be generated. Between min_length
and infinity. Default to 20.
The parameter for repetition penalty. Between 1.0 and infinity. 1.0 means no penalty. Default to 1.0.
Exponential penalty to the length. Default to 1.
if set to True
beam search is stopped when at least num_beams
sentences finished per batch.
Chatbot creation based on the Hugging Face State-of-the-Art Conversational AI.
Supported model types:
- GPT
- GPT2
Data format follows the Facebook Persona-Chat format. A JSON formatted version by Hugging Face is found here. The JSON file is directly compatible with this library (and it will be automatically downloaded and used if no dataset is specified).
Each entry in personachat is a dict with two keys personality
and utterances
, the dataset is a list of entries.
personality
: list of strings containing the personality of the agentutterances
: list of dictionaries, each of which has two keys which are lists of strings.candidates
: [next_utterance_candidate_1, ..., next_utterance_candidate_19]
The last candidate is the ground truth response observed in the conversational datahistory
: [dialog_turn_0, ... dialog_turn N], where N is an odd number since the other user starts every conversation.
Preprocessing:
- Spaces before periods at end of sentences
- everything lowercase
Example train data:
[
{
"personality": [
"i like computers .",
"i like reading books .",
"i like talking to chatbots .",
"i love listening to classical music ."
],
"utterances": [
{
"candidates": [
"i try to wear all black every day . it makes me feel comfortable .",
"well nursing stresses you out so i wish luck with sister",
"yeah just want to pick up nba nfl getting old",
"i really like celine dion . what about you ?",
"no . i live near farms .",
"mother taught me to cook ! we are looking for an exterminator .",
"i enjoy romantic movie . what is your favorite season ? mine is summer .",
"editing photos takes a lot of work .",
"you must be very fast . hunting is one of my favorite hobbies .",
"hi there . i'm feeling great! how about you ?"
],
"history": [
"hi , how are you ?"
]
},
{
"candidates": [
"i have trouble getting along with family .",
"i live in texas , what kind of stuff do you do in ",
"toronto ?",
"that's so unique ! veganism and line dancing usually don't mix !",
"no , it isn't that big . do you travel a lot",
"that's because they are real ; what do you do for work ?",
"i am lazy all day lol . my mom wants me to get a job and move out",
"i was born on arbor day , so plant a tree in my name",
"okay , i should not tell you , its against the rules ",
"i like to talk to chatbots too ! do you know why ? ."
],
"history": [
"hi , how are you ?",
"hi there . i'm feeling great! how about you ?",
"not bad ! i am trying out this chatbot ."
]
},
{
"candidates": [
"ll something like that . do you play games ?",
"does anything give you relief ? i hate taking medicine for mine .",
"i decorate cakes at a local bakery ! and you ?",
"do you eat lots of meat",
"i am so weird that i like to collect people and cats",
"how are your typing skills ?",
"yeah . i am headed to the gym in a bit to weight lift .",
"yeah you have plenty of time",
"metal is my favorite , but i can accept that people listen to country . haha",
"that's why you desire to be controlled . let me control you person one .",
"two dogs they are the best , how about you ?",
"you do art ? what kind of art do you do ?",
"i love watching baseball outdoors on sunny days .",
"oh i see . do you ever think about moving ? i do , it is what i want .",
"because i am a chatbot too, silly !"
],
"history": [
"hi , how are you ?",
"hi there . i'm feeling great! how about you ?",
"not bad ! i am trying out this chatbot .",
"i like to talk to chatbots too ! do you know why ? .",
"no clue, why don't you tell me ?"
]
}
]
}
]
You can download the pretrained (OpenAI GPT based) Conversation AI model open-sourced by Hugging Face here.
For the minimal example given below, you can download the model and extract it to gpt_personachat_cache
. Note that you can use any of the other GPT or GPT-2 models but they will require more training.
You will also need to create the JSON file given in the Data Format section and save it as data/minimal_train.json
.
from simpletransformers.conv_ai import ConvAIModel
train_args = {
"num_train_epochs": 50,
"save_model_every_epoch": False,
}
# Create a ConvAIModel
model = ConvAIModel("gpt", "gpt_personachat_cache", use_cuda=True, args=train_args)
# Train the model
model.train_model("data/minimal_train.json")
# Evaluate the model
model.eval_model()
# Interact with the trained model.
model.interact()
The interact()
method can be given a list of Strings which will be used to build a personality. If a list of Strings is not given, a random personality will be chosen from PERSONA-CHAT instead.
class simpletransformers.conv_ai.ConvAIModel ( model_type, model_name, args=None, use_cuda=True, cuda_device=-1, **kwargs)
This class is used to build Conversational AI.
Class attributes
tokenizer
: The tokenizer to be used.model
: The model to be used. model_name: Default Transformer model name or path to Transformer model file (pytorch_model.bin).device
: The device on which the model will be trained and evaluated.results
: A python dict of past evaluation results for the TransformerModel object.args
: A python dict of arguments used for training and evaluation.
Parameters
model_type
: (required) str - The type of model to use.model_name
: (required) str - The exact model to use. Could be a pretrained model name or path to a directory containing a model. See Current Pretrained Models for all available models.args
: (optional) python dict - A dictionary containing any settings that should be overwritten from the default values.use_cuda
: (optional) bool - Default = True. Flag used to indicate whether CUDA should be used.cuda_device
: (optional) int - Default = -1. Used to specify which GPU should be used.
class methods
train_model(self, train_file=None, output_dir=None, show_running_loss=True, args=None, eval_file=None)
Trains the model using 'train_file'
Args:
-
train_df: ath to JSON file containing training data. The model will be trained on this file. output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
output_dir (optional): The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
show_running_loss (Optional): Set to False to prevent training loss being printed.
-
args (optional): Optional changes to the args dict of the model. Any changes made will persist for the model.
-
eval_file (optional): Evaluation data against which evaluation will be performed when evaluate_during_training is enabled. If not given when evaluate_during_training is enabled, the evaluation data from PERSONA-CHAT will be used.
Returns:
- None
eval_model(self, eval_file, output_dir=None, verbose=True, silent=False)
Evaluates the model on eval_file. Saves results to output_dir.
Args:
-
eval_file: Path to JSON file containing evaluation data. The model will be evaluated on this file.
If not given, eval dataset from PERSONA-CHAT will be used. -
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
-
verbose: If verbose, results will be printed to the console on completion of evaluation.
-
silent: If silent, tqdm progress bars will be hidden.
Returns:
-
result: Dictionary containing evaluation results. (correct, similar, incorrect)
-
text: A dictionary containing the 3 dictionaries correct_text, similar_text (the predicted answer is a substring of the correct answer or vise versa), incorrect_text.
interact(self, personality=None)
Interact with a model in the terminal.
Args:
- personality (optional): A list of sentences that the model will use to build a personality.
If not given, a random personality from PERSONA-CHAT will be picked.
model.interact(
personality=[
"i like computers .",
"i like reading books .",
"i love classical music .",
"i am very social ."
]
)
Returns:
- None
train(self, train_dataloader, output_dir, show_running_loss=True, eval_dataloader=None, verbose=verbose)
Trains the model on train_dataset. Utility function to be used by the train_model() method. Not intended to be used directly.
evaluate(self, eval_file, output_dir, verbose=True, silent=False)
Evaluates the model on eval_file. Utility function to be used by the eval_model() method. Not intended to be used directly
load_and_cache_examples(self, dataset_path=None, evaluate=False, no_cache=False, verbose=True, silent=False)
Loads, tokenizes, and prepares data for training and/or evaluation. Utility function for train() and eval() methods. Not intended to be used directly
ConvAIModel has a few additional attributes in its args
dictionary, given below with their default values.
"num_candidates": 2,
"personality_permutations": 1,
"max_history": 2,
"lm_coef": 2.0,
"mc_coef": 1.0,
"no_sample": False,
"max_length": 20,
"min_length": 1,
"temperature": 0.7,
"top_k": 0,
"top_p": 0.9,
Number of candidates for training
Number of permutations of personality sentences".
Number of previous exchanges to keep in history
LM loss coefficient
Multiple-choice loss coefficient
Set to use greedy decoding instead of sampling
Maximum length of the output utterances
Minimum length of the output utterances
Sampling softmax temperature
Filter top-k tokens before sampling (<=0: no filtering)
Nucleus filtering (top-p) before sampling (<=0.0: no filtering)
Multi-Modal Classification fuses text and image data. This is performed using multi-modal bitransformer models introduced in the paper Supervised Multimodal Bitransformers for Classifying Images and Text.
Supported model types:
- BERT
There are several possible input formats you may use. The input formats are inspired by the MM-IMDb format. Note that several options for data preprocessing have been added for convenience and flexibility when dealing with complex datasets which can be found after the input format definitions.
Each subset of data (E.g: train and test) should be in its own directory. The path to the directory can then be given
directly to either train_model()
or eval_model()
.
Each data sample should have a text portion and an image associated with it (and a label/labels for training and evaluation data). The text for each sample should be in a separate JSON file. The JSON file may contain other fields in addition to the text itself but they will be ignored. The image associated with each sample should be in the same directory and both the text and the image must have the same identifier except for the file extension (E.g: 000001.json and 000001.jpg).
All data (including both train and test data) should be in the same directory. The path to this directory should be given
to both train_model()
and eval_model()
. A second argument, files_list
specifies which files should be taken from
the directory. files_list
can be a Python list or the path to a JSON file containing the list of files.
Each data sample should have a text portion and an image associated with it (and a label/labels for training and evaluation data). The text for each sample should be in a separate JSON file. The JSON file may contain other fields in addition to the text itself but they will be ignored. The image associated with each sample should be in the same directory and both the text and the image must have the same identifier except for the file extension (E.g: 000001.json and 000001.jpg).
Data can also be given in a Pandas DataFrame. When using this format, the image_path
argument must be specified and
it should be a String of the path to the directory containing the images. The DataFrame should contain at least 3
columns as detailed below.
text
(str) - The text associated with the sample.images
(str) - The relative path to the image file fromimage_path
directory.labels
(str) - The label (or list of labels for multilabel tasks) associated with the sample.
By default, Simple Transformers will look for column/field names text
, images
, and labels
. However, you can define
your own names to use in place of these names. This behaviour is controlled using the three attributes text_label
, labels_label
,
and images_label
in the args
dictionary.
You can set your custom names when creating the model by assigning the custom name to the corresponding attribute in the
args
dictionary.
You can also change these values at training and/or evaluation time (but not with the predict()
method) by passing the
names to the arguments text_label
, labels_label
, and images_label
. Note that the change will persist even after
the method call terminates. That is, the args
dictionary of the model itself will be modified.
By default, Simple Transformers will assume that any paths will also include the file type extension (E.g: .json or .jpg).
Alternatively, you can specify the extensions using the image_type_extension
and data_type_extension
attributes (for
image file extensions and text file extensions respectively) in the args
dictionary.
This too can be done when creating the model or when running the train_model()
or eval_model()
methods. The changes
will persist in the args
dictionary when using these methods.
The image_type_extension
can be specified when using the predict()
method but the change WILL NOT persist.
With Multi-Modal Classification, labels are always given as strings. You may specify a list of labels by passing in the
list to label_list
argument when creating the model. If label_list
is given, num_labels
is not required.
If label_list
is not given, num_labels
is required and the labels should be Strings starting from "0"
up to
"<num_labels>"
.
Create a MultiModalClassificationModel
.
from simpletransformers.classification.multi_modal_classification_model import MultiModalClassificationModel
model = MultiModalClassificationModel("bert", "bert-base-uncased")
Available arguments:
"""
Args:
model_type: The type of model (bert, xlnet, xlm, roberta, distilbert, albert)
model_name: Default Transformer model name or path to a directory containing Transformer model file (pytorch_model.bin).
multi_label (optional): Set to True for multi label tasks.
label_list (optional) : A list of all the labels (str) in the dataset.
num_labels (optional): The number of labels or classes in the dataset.
pos_weight (optional): A list of length num_labels containing the weights to assign to each label for loss calculation.
args (optional): Default args will be used if this parameter is not provided. If provided, it should be a dict containing the args that should be changed in the default args.
use_cuda (optional): Use GPU if available. Setting to False will force model to use CPU only.
cuda_device (optional): Specific GPU that should be used. Will use the first available GPU by default.
**kwargs (optional): For providing proxies, force_download, resume_download, cache_dir and other options specific to the 'from_pretrained' implementation where this will be supplied.
"""
Use the train_model()
method to train. You can use the auto_weights
feature to balance out unbalanced datasets.
Available arguments:
"""
Args:
data: Path to data directory containing text files (JSON) and image files OR a Pandas DataFrame.
If a DataFrame is given, it should contain the columns [text, labels, images]. When using a DataFrame,
image_path MUST be specified. The image column of the DataFrame should contain the relative path from
image_path to the image.
E.g:
For an image file 1.jpeg located in "data/train/";
image_path = "data/train/"
images = "1.jpeg"
files_list (optional): If given, only the files specified in this list will be taken from data directory.
files_list can be a Python list or the path (str) to a JSON file containing a list of files.
image_path (optional): Must be specified when using DataFrame as input. Path to the directory containing the
images.
text_label (optional): Column name to look for instead of the default "text"
labels_label (optional): Column name to look for instead of the default "labels"
images_label (optional): Column name to look for instead of the default "images"
image_type_extension (optional): If given, this will be added to the end of each value in "images".
data_type_extension (optional): If given, this will be added to the end of each value in "files_list".
auto_weights (optional): If True, weights will be used to balance the classes.
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
show_running_loss (optional): Set to False to prevent running loss from being printed to console. Defaults to True.
args (optional): Optional changes to the args dict of the model. Any changes made will persist for the model.
eval_data (optional): A DataFrame against which evaluation will be performed when evaluate_during_training is enabled. Is required if evaluate_during_training is enabled.
**kwargs: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). E.g. f1=sklearn.metrics.f1_score.
A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
"""
Use the eval_model()
method to evaluate. You can load a saved model by giving the path to the model directory as
model_name
. Note that you need to provide the same arguments when loading a saved model as you did when creating the
original model.
model = MultiModalClassificationModel("bert", "outputs")
results, _ = model.eval_model("data/dataset/", "data/dev.json")
Available arguments:
"""
Args:
data: Path to data directory containing text files (JSON) and image files OR a Pandas DataFrame.
If a DataFrame is given, it should contain the columns [text, labels, images]. When using a DataFrame,
image_path MUST be specified. The image column of the DataFrame should contain the relative path from
image_path to the image.
E.g:
For an image file 1.jpeg located in "data/train/";
image_path = "data/train/"
images = "1.jpeg"
files_list (optional): If given, only the files specified in this list will be taken from data directory.
files_list can be a Python list or the path (str) to a JSON file containing a list of files.
image_path (optional): Must be specified when using DataFrame as input. Path to the directory containing the
images.
text_label (optional): Column name to look for instead of the default "text"
labels_label (optional): Column name to look for instead of the default "labels"
images_label (optional): Column name to look for instead of the default "images"
image_type_extension (optional): If given, this will be added to the end of each value in "images".
data_type_extension (optional): If given, this will be added to the end of each value in "files_list".
output_dir: The directory where model files will be saved. If not given, self.args['output_dir'] will be used.
verbose: If verbose, results will be printed to the console on completion of evaluation.
silent: If silent, tqdm progress bars will be hidden.
**kwargs: Additional metrics that should be used. Pass in the metrics as keyword arguments (name of metric: function to use). E.g. f1=sklearn.metrics.f1_score.
A metric function should take in two parameters. The first parameter will be the true labels, and the second parameter will be the predictions.
"""
Use the predict()
method to make predictions. You can load a saved model by giving the path to the model directory as
model_name
. Note that you need to provide the same arguments when loading a saved model as you did when creating the
original model.
model = MultiModalClassificationModel("bert", "outputs")
model.predict(
{
"text": [
"A lawyer is forced to defend a guilty judge, while defending other innocent clients, and trying to find punishment for the guilty and provide justice for the innocent."
],
"labels": ["Crime"],
"images": ["0078718"]
},
image_path="data/dataset",
image_type_extension=".jpeg"
)
Regression tasks also use the ClassificationModel with 2 caveats.
num_labels
should be 1.regression
should beTrue
inargs
dict.
Regression can be used with either single sentence or sentence pair tasks.
from simpletransformers.classification import ClassificationModel
import pandas as pd
train_data = [
['Example sentence belonging to class 1', 'Yep, this is 1', 1.8],
['Example sentence belonging to class 0', 'Yep, this is 0', 0.2],
['Example 2 sentence belonging to class 0', 'Yep, this is 0', 4.5]
]
train_df = pd.DataFrame(train_data, columns=['text_a', 'text_b', 'labels'])
eval_data = [
['Example sentence belonging to class 1', 'Yep, this is 1', 1.9],
['Example sentence belonging to class 0', 'Yep, this is 0', 0.1],
['Example 2 sentence belonging to class 0', 'Yep, this is 0', 5]
]
eval_df = pd.DataFrame(eval_data, columns=['text_a', 'text_b', 'labels'])
train_args={
'reprocess_input_data': True,
'overwrite_output_dir': True,
'num_train_epochs': 3,
'regression': True,
}
# Create a ClassificationModel
model = ClassificationModel('roberta', 'roberta-base', num_labels=1, use_cuda=True, cuda_device=0, args=train_args)
print(train_df.head())
# Train the model
model.train_model(train_df, eval_df=eval_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df)
predictions, raw_outputs = model.predict([["I'd like to puts some CD-ROMS on my iPad, is that possible?'", "Yes, but wouldn't that block the screen?"]])
print(predictions)
print(raw_outputs)
The Weights & Biases framework is supported for visualizing model training.
To use this, simply set a project name for W&B in the wandb_project
attribute of the args
dictionary. This will log all hyperparameter values, training losses, and evaluation metrics to the given project.
model = ClassificationModel('roberta', 'roberta-base', args={'wandb_project': 'project-name'})
For a complete example, see here.
To use experimental features, import from simpletransformers.experimental.X
from simpletransformers.experimental.classification import ClassificationModel
Normally, sequences longer than max_seq_length
are unceremoniously truncated.
This experimental feature moves a sliding window over each sequence and generates sub-sequences with length max_seq_length
. The model output for each sub-sequence is averaged into a single output before being sent to the linear classifier.
Currently available on binary and multiclass classification models of the following types:
- BERT
- DistilBERT
- RoBERTa
- AlBERT
- XLNet
- CamemBERT
Set sliding_window
to True
for the ClassificationModel to enable this feature.
from simpletransformers.classification import ClassificationModel
import pandas as pd
import sklearn
# Train and Evaluation data needs to be in a Pandas Dataframe of two columns. The first column is the text with type str, and the second column in the label with type int.
train_data = [['Example sentence belonging to class 1' * 50, 1], ['Example sentence belonging to class 0', 0], ['Example 2 sentence belonging to class 0', 0]] + [['Example sentence belonging to class 0', 0] for i in range(12)]
train_df = pd.DataFrame(train_data, columns=['text', 'labels'])
eval_data = [['Example eval sentence belonging to class 1', 1], ['Example eval sentence belonging to class 0', 0]]
eval_df = pd.DataFrame(eval_data)
train_args={
'sliding_window': True,
'reprocess_input_data': True,
'overwrite_output_dir': True,
'evaluate_during_training': True,
'logging_steps': 5,
'stride': 0.8,
'max_seq_length': 128
}
# Create a TransformerModel
model = ClassificationModel('camembert', 'camembert-base', args=train_args, use_cuda=False)
print(train_df.head())
# Train the model
model.train_model(train_df, eval_df=eval_df)
# Evaluate the model
result, model_outputs, wrong_predictions = model.eval_model(eval_df, acc=sklearn.metrics.accuracy_score)
predictions, raw_outputs = model.predict(["I'd like to puts some CD-ROMS on my iPad, is that possible?' — Yes, but wouldn't that block the screen?" * 25])
print(predictions)
print(raw_outputs)
To load a saved model, provide the path to the directory containing the saved model as the model_name
.
Note that you will need to specify the correct (usually the same used in training) args
when loading the model
model = ClassificationModel('roberta', 'outputs/', args={})
model = NERModel('bert', 'outputs/', args={})
The default args used are given below. Any of these can be overridden by passing a dict containing the corresponding key: value pairs to the the init method of a Model class.
self.args = {
"output_dir": "outputs/",
"cache_dir": "cache/",
"best_model_dir": "outputs/best_model/",
"fp16": True,
"fp16_opt_level": "O1",
"max_seq_length": 128,
"train_batch_size": 8,
"eval_batch_size": 8,
"gradient_accumulation_steps": 1,
"num_train_epochs": 1,
"weight_decay": 0,
"learning_rate": 4e-5,
"adam_epsilon": 1e-8,
"warmup_ratio": 0.06,
"warmup_steps": 0,
"max_grad_norm": 1.0,
"do_lower_case": False,
"logging_steps": 50,
"evaluate_during_training": False,
"evaluate_during_training_steps": 2000,
"evaluate_during_training_verbose": False,
"use_cached_eval_features": False,
"save_eval_checkpoints": True
"save_steps": 2000,
"no_cache": False,
"save_model_every_epoch": True,
"tensorboard_dir": None,
"overwrite_output_dir": False,
"reprocess_input_data": True,
"process_count": cpu_count() - 2 if cpu_count() > 2 else 1
"n_gpu": 1,
"silent": False,
"use_multiprocessing": True,
"wandb_project": None,
"wandb_kwargs": {},
"use_early_stopping": True,
"early_stopping_patience": 3,
"early_stopping_delta": 0,
"early_stopping_metric": "eval_loss",
"early_stopping_metric_minimize": True,
"manual_seed": None,
"encoding": None,
"config": {},
}
The directory where all outputs will be stored. This includes model checkpoints and evaluation results.
The directory where cached files will be saved.
The directory where the best model (model checkpoints) will be saved if evaluate_during_training is enabled and the training loop achieves a lowest evaluation loss calculated after every evaluate_during_training_steps, or an epoch.
Whether or not fp16 mode should be used. Requires NVidia Apex library.
Can be '01', '02', '03'. See the Apex docs for an explanation of the different optimization levels (opt_levels).
Maximum sequence level the model will support.
The training batch size.
The number of training steps to execute before performing a optimizer.step()
. Effectively increases the training batch size while sacrificing training time to lower memory consumption.
The evaluation batch size.
The number of epochs the model will be trained for.
Adds L2 penalty.
The learning rate for training.
Epsilon hyperparameter used in AdamOptimizer.
Maximum gradient clipping.
Set to True when using uncased models.
Set to True to perform evaluation while training models. Make sure eval_df
is passed to the training method if enabled.
Perform evaluation at every specified number of steps. A checkpoint model and the evaluation results will be saved.
Print results from evaluation during training.
Evaluation during training uses cached features. Setting this to False
will cause features to be recomputed at every evaluation step.
Save a model checkpoint for every evaluation performed.
Log training loss and learning at every specified number of steps.
Save a model checkpoint at every specified number of steps.
Cache features to disk.
Save a model at the end of every epoch.
The directory where Tensorboard events will be stored during training. By default, Tensorboard events will be saved in a subfolder inside runs/
like runs/Dec02_09-32-58_36d9e58955b0/
.
If True, the trained model will be saved to the ouput_dir and will overwrite existing saved models in the same directory.
If True, the input data will be reprocessed even if a cached file of the input data exists in the cache_dir.
Number of cpu cores (processes) to use when converting examples to features. Default is (number of cores - 2) or 1 if (number of cores <= 2)
Number of GPUs to use.
Disables progress bars.
If True, multiprocessing will be used when converting data into features. Disabling can reduce memory usage, but may substantially slow down processing.
Name of W&B project. This will log all hyperparameter values, training losses, and evaluation metrics to the given project.
Dictionary of keyword arguments to be passed to the W&B project.
Use early stopping to stop training when early_stopping_metric
doesn't improve (based on early_stopping_patience
, and early_stopping_delta
)
Terminate training after this many evaluations without an improvement in eval_loss
greater then early_stopping_delta
.
The improvement over best_eval_loss
necessary to count as a better checkpoint.
The metric that should be used with early stopping. (Should be computed during eval_during_training
).
Whether early_stopping_metric
should be minimized (or maximized).
Set a manual seed if necessary for reproducible results.
Specify an encoding to be used when reading text files.
A dictionary containing configuration options that should be overriden in a model's config.
For a list of pretrained models, see Hugging Face docs.
The model_types
available for each task can be found under their respective section. Any pretrained model of that type
found in the Hugging Face docs should work. To use any of them set the correct model_type
and model_name
in the args
dictionary.
None of this would have been possible without the hard work by the HuggingFace team in developing the Pytorch-Transformers library.
Thanks goes to these wonderful people (emoji key):
This project follows the all-contributors specification. Contributions of any kind welcome!
If you should be on this list but you aren't, or you are on the list but don't want to be, please don't hesitate to contact me!