Functions — github.com/ahangchen/GDLnotes

hub / github.com/ahangchen/GDLnotes / functions

Functions224 in github.com/ahangchen/GDLnotes

⨍ Functions 224◇ Types & classes 7

↓ 40 callersFunctionaccuracy

(predictions, labels)

src/neural/full_connect.py:16

↓ 28 callersFunctionmaxpool2d

(data, k=2, s=2)

src/convnet/conv_mnist.py:41

↓ 20 callersFunctionvariable_summary

Attach a lot of summaries to a Tensor (for TensorBoard visualization).

src/util/board.py:17

↓ 13 callersFunctionchar2id

(char)

src/rnn/seq2seq.py:46

↓ 12 callersFunctionload_pickle

(pickle_name)

src/not_mnist/img_pickle.py:77

↓ 11 callersFunctionmaybe_download

Download a file if not present, and make sure it's the right size.

src/not_mnist/load_data.py:42

↓ 11 callersFunctionsave_obj

(pickle_file, obj)

src/not_mnist/img_pickle.py:84

↓ 9 callersFunctionformat_mnist

()

src/util/mnist.py:25

↓ 9 callersFunctionmodel

(data)

src/convnet/conv_mnist.py:79

↓ 7 callersFunctionid2char

(dictid)

src/rnn/seq2seq.py:58

↓ 6 callersFunctionbi2id

(char)

src/rnn/bigram_lstm.py:53

↓ 6 callersFunctionchar2id

(char)

src/rnn/embed_bigram_lstm.py:47

↓ 6 callersFunctionmaybe_pickle

(data_folders, min_num_images_per_class, force=False)

src/not_mnist/img_pickle.py:45

↓ 6 callersMethodnext

Generate the next array of batches from the data. The array consists of the last batch of the previous array, followed by num_unrollings new o

src/rnn/bigram_lstm.py:95

↓ 6 callersMethodnext

Generate the next array of batches from the data. The array consists of the last batch of the previous array, followed by num_unrollings new o

src/rnn/singlew_lstm.py:94

↓ 6 callersMethodnext

(self)

src/rnn/embed_bigram_lstm.py:95

↓ 6 callersMethodnext

Generate the next array of batches from the data. The array consists of the last batch of the previous array, followed by num_unrollings new o

src/rnn/lstm.py:91

↓ 5 callersFunctionchar2id

(char)

src/rnn/singlew_lstm.py:47

↓ 5 callersFunctionchar2id

(char)

src/rnn/lstm.py:47

↓ 5 callersFunctionid2bi

(dictid)

src/rnn/bigram_lstm.py:61

↓ 5 callersFunctionid2char

(dictid)

src/rnn/embed_bigram_lstm.py:57

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/cnn_long_optimize.py:13

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/cnn_half_optimize.py:13

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/cnn_prophet.py:14

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/cnn_step_optimize.py:13

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/cnn_optimize.py:13

↓ 5 callersFunctionlarge_data_size

(data)

src/optimize/random_param_cnn.py:13

↓ 5 callersFunctionlarge_data_size

(data)

src/convnet/cnn_board.py:11

↓ 5 callersFunctionlarge_data_size

(data)

src/convnet/hyper_conv_mnist.py:12

↓ 5 callersFunctionread2mem

(path)

src/util/file_helper.py:6

↓ 4 callersFunction_slice

(_x, n, dim)

src/rnn/bigram_lstm.py:182

↓ 4 callersFunction_slice

(_x, n, dim)

src/rnn/lstm_regular.py:88

↓ 4 callersFunctionbatches2string

Convert a sequence of batches back into their (most likely) string representation.

src/rnn/bigram_lstm.py:112

↓ 4 callersFunctionbatches2string

Convert a sequence of batches back into their (most likely) string representation.

src/rnn/singlew_lstm.py:112

↓ 4 callersFunctionbatches2string

Convert a sequence of batches back into their (most likely) string representation.

src/rnn/lstm.py:109

↓ 4 callersFunctionbibatches2string

(batches)

src/rnn/embed_bigram_lstm.py:111

↓ 4 callersFunctiondata_to_token_ids

Tokenize data file and turn into token-ids using given vocabulary file. This function loads data line-by-line from data_path, calls the above sen

src/rnn/data_utils.py:216

↓ 4 callersMethodget_batch

Get a random batch of data from the specified bucket, prepare for step. To feed data in step(..) it must be a list of batch-major vectors, while

src/rnn/seq2seq_model.py:257

↓ 4 callersFunctionid2char

(dictid)

src/rnn/singlew_lstm.py:57

↓ 4 callersFunctionid2char

(dictid)

src/rnn/lstm.py:57

↓ 4 callersFunctionmaybe_extract

(filename, force=False)

src/not_mnist/extract.py:16

↓ 4 callersFunctionmerge_datasets

(pickle_files, train_size, valid_size=0)

src/not_mnist/merge_prune.py:24

↓ 4 callersFunctionread_lines

(path)

src/util/file_helper.py:22

↓ 4 callersMethodstep

Run a step of the model feeding the given inputs. Args: session: tensorflow session to use. encoder_inputs: list of numpy int vectors

src/rnn/seq2seq_model.py:197

↓ 3 callersFunctionaccuracy

(predictions, labels, is_test=False)

src/app/caltech/dnn_caltech_board.py:29

↓ 3 callersFunctionbi2str

(encoding)

src/rnn/embed_bigram_lstm.py:103

↓ 3 callersFunctioncharacters

Turn a 1-hot encoding or a probability distribution over the possible characters back into its (mostl likely) character representation.

src/rnn/bigram_lstm.py:106

↓ 3 callersFunctioncharacters

Turn a 1-hot encoding or a probability distribution over the possible characters back into its (most likely) character representation.

src/rnn/singlew_lstm.py:105

↓ 3 callersFunctioncharacters

Turn a 1-hot encoding or a probability distribution over the possible characters back into its (most likely) character representation.

src/rnn/lstm.py:102

↓ 3 callersFunctionimg_reshape

(data, length)

src/util/mnist.py:10

↓ 3 callersFunctionimgs_idx_hash_except

(left, right)

src/not_mnist/clean_overlap.py:76

↓ 3 callersFunctionimread

(path)

src/app/neural_style/neural_style.py:142

↓ 3 callersFunctionlabel_reshape

(data, length)

src/util/mnist.py:18

↓ 3 callersFunctionmodel

(data, init=False)

src/optimize/cnn_half_optimize.py:60

↓ 3 callersFunctionmodel

(data, init=False)

src/optimize/cnn_prophet.py:55

↓ 3 callersFunctionmodel

(data, init=False)

src/optimize/cnn_step_optimize.py:60

↓ 3 callersFunctionmodel

(data, init=False)

src/optimize/cnn_optimize.py:54

↓ 3 callersFunctionmodel

(data, init=False)

src/optimize/random_param_cnn.py:55

↓ 3 callersFunctionmodel

(data, model_drop=True, init=True)

src/convnet/hyper_conv_mnist.py:54

↓ 3 callersFunctionrandomize

(dataset, labels)

src/not_mnist/merge_prune.py:59

↓ 3 callersFunctionread_caltech

()

src/app/caltech/data.py:16

↓ 3 callersFunctionrecall_rate

(predictions, labels, is_test=False)

src/app/caltech/dnn_caltech_board.py:12

↓ 3 callersFunctionreformat

(dataset, labels, image_size, num_labels, num_channels)

src/convnet/conv_mnist.py:8

↓ 3 callersFunctionreformat

(dataset, labels, image_size, num_labels)

src/neural/full_connect.py:9

↓ 2 callersMethod_next_batch

Generate a single batch from the current cursor position in the data.

src/rnn/bigram_lstm.py:87

↓ 2 callersMethod_next_batch

Generate a single batch from the current cursor position in the data.

src/rnn/seq2seq.py:85

↓ 2 callersMethod_next_batch

Generate a single batch from the current cursor position in the data.

src/rnn/singlew_lstm.py:82

↓ 2 callersMethod_next_batch

(self)

src/rnn/embed_bigram_lstm.py:81

↓ 2 callersMethod_next_batch

Generate a single batch from the current cursor position in the data.

src/rnn/lstm.py:82

↓ 2 callersFunction_tensor_size

(tensor)

src/app/neural_style/stylize.py:131

↓ 2 callersFunctionbasic_tokenizer

Very basic tokenizer: split the sentence into a list of tokens.

src/rnn/data_utils.py:106

↓ 2 callersFunctionbetter_hyper

(hypers, loss)

src/util/request.py:33

↓ 2 callersFunctionclean

()

src/not_mnist/clean_overlap.py:96

↓ 2 callersFunctionconv_train

(train_dataset, train_labels, valid_dataset, valid_labels, test_dataset, test_labels, image_size,

src/optimize/cnn_half_optimize.py:17

↓ 2 callersFunctionconv_train

(train_dataset, train_labels, valid_dataset, valid_labels, test_dataset, test_labels, image_size,

src/optimize/cnn_step_optimize.py:17

↓ 2 callersFunctionconv_train

(train_dataset, train_labels, valid_dataset, valid_labels, test_dataset, test_labels, image_size,

src/optimize/cnn_optimize.py:17

↓ 2 callersFunctioncreate_lstm_graph_bi

(num_nodes, num_unrollings, batch_size, embedding_size=bi_voc_size)

src/rnn/embed_bigram_lstm.py:176

↓ 2 callersFunctioncreate_vocabulary

Create vocabulary file (if it does not exist yet) from data file. Data file is assumed to contain one sentence per line. Each sentence is tokeniz

src/rnn/data_utils.py:114

↓ 2 callersFunctiondiv

(xt)

src/neural/digit_nn.py:6

↓ 2 callersFunctionfit_loss

(reset, hypers, loss)

src/util/request.py:17

↓ 2 callersFunctionfit_more

(reset, hypers, loss)

src/util/request.py:25

↓ 2 callersFunctiongenerate_batch

(batch_size, num_skips, skip_window)

src/rnn/word2vec.py:42

↓ 2 callersFunctiongenerate_batch

(batch_size, num_skips, skip_window)

src/rnn/cbow.py:42

↓ 2 callersFunctiongunzip_file

Unzips from gz_path into new_path.

src/rnn/data_utils.py:66

↓ 2 callersFunctionimg_diff

(pix_s1, pix_s2)

src/not_mnist/clean_overlap.py:10

↓ 2 callersFunctionimg_hash

(pix_s)

src/not_mnist/clean_overlap.py:43

↓ 2 callersFunctionimg_in

(img, imgs)

src/not_mnist/clean_overlap.py:28

↓ 2 callersFunctionload_reformat_not_mnist

(image_size, num_labels)

src/neural/full_connect.py:213

↓ 2 callersFunctionload_train

()

src/not_mnist/logistic_train.py:10

↓ 2 callersFunctionlogprob

Log-probability of the true labels in a predicted batch.

src/rnn/bigram_lstm.py:130

↓ 2 callersFunctionlogprob

Log-probability of the true labels in a predicted batch.

src/rnn/lstm_regular.py:65

↓ 2 callersFunctionlogprob

Log-probability of the true labels in a predicted batch.

src/rnn/singlew_lstm.py:130

↓ 2 callersFunctionlogprob

Log-probability of the true labels in a predicted batch.

src/rnn/embed_bigram_lstm.py:138

↓ 2 callersFunctionlogprob

Log-probability of the true labels in a predicted batch.

src/rnn/lstm.py:127

↓ 2 callersFunctionlstm_cell

(i, o, state)

src/rnn/bigram_lstm.py:187

↓ 2 callersFunctionlstm_cell

(cur_input, last_output, last_state, drop)

src/rnn/lstm_regular.py:92

↓ 2 callersFunctionlstm_cell

Create a LSTM cell. See e.g.: http://arxiv.org/pdf/1402.1128v1.pdf Note that in this formulation, we omit the various connections between the

src/rnn/singlew_lstm.py:183

↓ 2 callersFunctionlstm_cell

(i, o, state)

src/rnn/embed_bigram_lstm.py:200

↓ 2 callersFunctionlstm_cell

Create a LSTM cell. See e.g.: http://arxiv.org/pdf/1402.1128v1.pdf Note that in this formulation, we omit the various connections between the

src/rnn/lstm.py:191

↓ 2 callersFunctionmake_arrays

(nb_rows, img_size)

src/not_mnist/merge_prune.py:15

next →1–100 of 224, ranked by callers