Types & classes — github.com/airsplay/lxmert

CodeHub MCPcopy Create free account

hub / github.com/airsplay/lxmert / types & classes

Types & classes53 in github.com/airsplay/lxmert

⨍ Functions 214◇ Types & classes 53

↓ 4 callersClassBertAdam

Implements BERT version of Adam algorithm with weight decay fix. Params: lr: learning rate warmup: portion of t_total for the warm

src/lxrt/optimization.py:52

↓ 4 callersClassGeLU

Implementation of the gelu activation function. For information: OpenAI GPT's gelu is slightly different (and gives slightly different results

src/lxrt/modeling.py:121

↓ 3 callersClassBertIntermediate

src/lxrt/modeling.py:389

↓ 3 callersClassBertOutput

src/lxrt/modeling.py:404

↓ 3 callersClassBertSelfattLayer

src/lxrt/modeling.py:376

↓ 3 callersClassLXRTEncoder

src/lxrt/entry.py:80

↓ 2 callersClassAnswerTable

src/pretrain/qa_answer_table.py:8

↓ 2 callersClassBertAttOutput

src/lxrt/modeling.py:350

↓ 2 callersClassBertAttention

src/lxrt/modeling.py:295

↓ 2 callersClassBertLayer

src/lxrt/modeling.py:418

↓ 2 callersClassBertPredictionHeadTransform

src/lxrt/modeling.py:584

↓ 2 callersClassLXRTModel

src/lxrt/modeling.py:835

↓ 1 callersClassBasicTokenizer

Runs basic tokenization (punctuation splitting, lower casing, etc.).

src/lxrt/tokenization.py:174

↓ 1 callersClassBertConfig

Configuration class to store the configuration of a `BertModel`.

src/lxrt/modeling.py:172

↓ 1 callersClassBertCrossattLayer

src/lxrt/modeling.py:364

↓ 1 callersClassBertEmbeddings

Construct the embeddings from word, position and token_type embeddings.

src/lxrt/modeling.py:264

↓ 1 callersClassBertLMPredictionHead

src/lxrt/modeling.py:601

↓ 1 callersClassBertPooler

src/lxrt/modeling.py:569

↓ 1 callersClassBertPreTrainingHeads

src/lxrt/modeling.py:661

↓ 1 callersClassBertVisualAnswerHead

src/lxrt/modeling.py:620

↓ 1 callersClassBertVisualObjHead

src/lxrt/modeling.py:635

↓ 1 callersClassGQA

src/tasks/gqa.py:34

↓ 1 callersClassGQABufferLoader

src/tasks/gqa_data.py:63

↓ 1 callersClassGQADataset

A GQA data example in json file: { "img_id": "2375429", "label": { "pipe": 1.0 }, "question_id":

src/tasks/gqa_data.py:20

↓ 1 callersClassGQAEvaluator

src/tasks/gqa_data.py:158

↓ 1 callersClassGQAModel

src/tasks/gqa_model.py:14

↓ 1 callersClassGQATorchDataset

src/tasks/gqa_data.py:89

↓ 1 callersClassInputExample

A single training/test example for the language model.

src/pretrain/lxmert_data.py:26

↓ 1 callersClassInputFeatures

A single set of features of data.

src/lxrt/entry.py:27

↓ 1 callersClassInputFeatures

A single set of features of data.

src/pretrain/lxmert_pretrain.py:51

↓ 1 callersClassLXMERT

src/pretrain/lxmert_pretrain.py:214

↓ 1 callersClassLXMERTDataset

src/pretrain/lxmert_data.py:40

↓ 1 callersClassLXMERTEvaluator

src/pretrain/lxmert_data.py:198

↓ 1 callersClassLXMERTTorchDataset

src/pretrain/lxmert_data.py:87

↓ 1 callersClassLXRTEncoder

src/lxrt/modeling.py:520

↓ 1 callersClassLXRTXLayer

src/lxrt/modeling.py:439

↓ 1 callersClassNLVR2

src/tasks/nlvr2.py:32

↓ 1 callersClassNLVR2Dataset

An NLVR2 data example in json file: { "identifier": "train-10171-0-0", "img0": "train-10171-0-img0", "img1": "train-1

src/tasks/nlvr2_data.py:19

↓ 1 callersClassNLVR2Evaluator

src/tasks/nlvr2_data.py:129

↓ 1 callersClassNLVR2Model

src/tasks/nlvr2_model.py:10

↓ 1 callersClassNLVR2TorchDataset

src/tasks/nlvr2_data.py:58

↓ 1 callersClassVQA

src/tasks/vqa.py:33

↓ 1 callersClassVQADataset

A VQA data example in json file: { "answer_type": "other", "img_id": "COCO_train2014_000000458752", "

src/tasks/vqa_data.py:33

↓ 1 callersClassVQAEvaluator

src/tasks/vqa_data.py:153

↓ 1 callersClassVQAModel

src/tasks/vqa_model.py:14

↓ 1 callersClassVQATorchDataset

src/tasks/vqa_data.py:82

↓ 1 callersClassVisualConfig

src/lxrt/modeling.py:141

↓ 1 callersClassVisualFeatEncoder

src/lxrt/modeling.py:491

↓ 1 callersClassWordpieceTokenizer

Runs WordPiece tokenization.

src/lxrt/tokenization.py:291

ClassBertPreTrainedModel

An abstract class to handle weights initialization and a simple interface for dowloading and loading pretrained models.

src/lxrt/modeling.py:673

ClassBertTokenizer

Runs end-to-end tokenization: punctuation splitting + wordpiece

src/lxrt/tokenization.py:72

ClassLXRTFeatureExtraction

BERT model for classification.

src/lxrt/modeling.py:992

ClassLXRTPretraining

src/lxrt/modeling.py:889