from fastai.gen_doc.nbdoc import *
from fastai.text import *
The main thing here is RNNLearner. There are also some utility functions to help create and update text models.
show_doc(language_model_learner)
The model used is given by arch and config. It can be:
AWD_LSTM(Merity et al.)Transformer decoder (Vaswani et al.)TransformerXL (Dai et al.)They each have a default config for language modelling that is in {lower_case_class_name}_lm_config if you want to change the default parameter. At this stage, only the AWD LSTM support pretrained=True but we hope to add more pretrained models soon. drop_mult is applied to all the dropouts weights of the config, learn_kwargs are passed to the Learner initialization.
jekyll_note("Using QRNN (change the flag in the config of the AWD LSTM) requires to have cuda installed (same version as pytorch is using).")
path = untar_data(URLs.IMDB_SAMPLE)
data = TextLMDataBunch.from_csv(path, 'texts.csv')
learn = language_model_learner(data, AWD_LSTM, drop_mult=0.5)
show_doc(text_classifier_learner)
text_classifier_learner[source]
text_classifier_learner(data:DataBunch,arch:Callable,bptt:int=*70,max_len:int=1400,config:dict=None,pretrained:bool=True,drop_mult:float=1.0,lin_ftrs:Collection[int]=None,ps:Collection[float]=None, ***learn_kwargs**) →TextClassifierLearner
Create a Learner with a text classifier from data and arch.
Here again, the backbone of the model is determined by arch and config. The input texts are fed into that model by bunch of bptt and only the last max_len activations are considered. This gives us the backbone of our model. The head then consists of:
nn.BatchNorm1d, nn.Dropout, nn.Linear, nn.ReLU) layers.The blocks are defined by the lin_ftrs and drops arguments. Specifically, the first block will have a number of inputs inferred from the backbone arch and the last one will have a number of outputs equal to data.c (which contains the number of classes of the data) and the intermediate blocks have a number of inputs/outputs determined by lin_ftrs (of course a block has a number of inputs equal to the number of outputs of the previous block). The dropouts all have a the same value ps if you pass a float, or the corresponding values if you pass a list. Default is to have an intermediate hidden size of 50 (which makes two blocks model_activation -> 50 -> n_classes) with a dropout of 0.1.
path = untar_data(URLs.IMDB_SAMPLE)
data = TextClasDataBunch.from_csv(path, 'texts.csv')
learn = text_classifier_learner(data, AWD_LSTM, drop_mult=0.5)
show_doc(RNNLearner)
Handles the whole creation from data and a model with a text data using a certain bptt. The split_func is used to properly split the model in different groups for gradual unfreezing and differential learning rates. Gradient clipping of clip is optionally applied. alpha and beta are all passed to create an instance of RNNTrainer. Can be used for a language model or an RNN classifier. It also handles the conversion of weights from a pretrained model as well as saving or loading the encoder.
show_doc(RNNLearner.get_preds)
get_preds[source]
get_preds(ds_type:DatasetType=*<DatasetType.Valid: 2>,with_loss:bool=False,n_batch:Optional[int]=None,pbar:Union[MasterBar,ProgressBar,NoneType]=None,ordered:bool=False*) →List[Tensor]
Return predictions and targets on the valid, train, or test set, depending on ds_type.
If ordered=True, returns the predictions in the order of the dataset, otherwise they will be ordered by the sampler (from the longest text to the shortest). The other arguments are passed Learner.get_preds.
show_doc(RNNLearner.load_encoder)
show_doc(RNNLearner.save_encoder)
show_doc(RNNLearner.load_pretrained)
load_pretrained[source]
load_pretrained(wgts_fname:str,itos_fname:str,strict:bool=*True*)
Load a pretrained model and adapts it to the data vocabulary.
Opens the weights in the wgts_fname of self.model_dir and the dictionary in itos_fname then adapts the pretrained weights to the vocabulary of the data. The two files should be in the models directory of the learner.path.
show_doc(convert_weights)
convert_weights[source]
convert_weights(wgts:Weights,stoi_wgts:Dict[str,int],itos_new:StrList) →Weights
Convert the model wgts to go with a new vocabulary.
Uses the dictionary stoi_wgts (mapping of word to id) of the weights to map them to a new dictionary itos_new (mapping id to word).
show_doc(LanguageLearner, title_level=3)
class LanguageLearner[source]
LanguageLearner(data:DataBunch,model:Module,split_func:OptSplitFunc=*None,clip:float=None,alpha:float=2.0,beta:float=1.0,metrics=None, ***learn_kwargs**) ::RNNLearner
Subclass of RNNLearner for predictions.
show_doc(LanguageLearner.predict)
predict[source]
predict(text:str,n_words:int=*1,no_unk:bool=True,temperature:float=1.0,min_p:float=None,sep:str=' ',decoder='decode_spec_tokens'*)
Return the n_words that come after text.
If no_unk=True the unknown token is never picked. Words are taken randomly with the distribution of probabilities returned by the model. If min_p is not None, that value is the minimum probability to be considered in the pool of words. Lowering temperature will make the texts less randomized.
show_doc(LanguageLearner.beam_search)
beam_search[source]
beam_search(text:str,n_words:int,top_k:int=*10,beam_sz:int=1000,temperature:float=1.0,sep:str=' ',decoder='decode_spec_tokens'*)
Return the n_words that come after text using beam search.
show_doc(get_language_model)
get_language_model[source]
get_language_model(arch:Callable,vocab_sz:int,config:dict=*None,drop_mult:float=1.0*)
Create a language model from arch and its config, maybe pretrained.
show_doc(get_text_classifier)
This model uses an encoder taken from the arch on config. This encoder is fed the sequence by successive bits of size bptt and we only keep the last max_seq outputs for the pooling layers.
The decoder use a concatenation of the last outputs, a MaxPooling of all the outputs and an AveragePooling of all the outputs. It then uses a list of BatchNorm, Dropout, Linear, ReLU blocks (with no ReLU in the last one), using a first layer size of 3*emb_sz then following the numbers in n_layers. The dropouts probabilities are read in drops.
Note that the model returns a list of three things, the actual output being the first, the two others being the intermediate hidden states before and after dropout (used by the RNNTrainer). Most loss functions expect one output, so you should use a Callback to remove the other two if you're not using RNNTrainer.