issues
search
huggingface
/
pytorch-openai-transformer-lm
🐥A PyTorch implementation of OpenAI's finetuned transformer language model with a script to import the weights pre-trained by OpenAI
MIT License
1.51k
stars
285
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Implementation of Similarity Head
#60
TEGELB
opened
3 years ago
0
Training from scratch: Repeated and mangled words
#59
maruker
opened
5 years ago
0
Instructions for encoding own sentences
#58
izaskr
opened
5 years ago
1
Running on new dataset similar to rocstories
#57
priyanka-chaudhary
opened
5 years ago
0
ConvAI
#55
bkj
opened
5 years ago
0
vocab = n_vocab + n_special + n_ctx means?
#54
JiahangOK
opened
5 years ago
1
Why do we need to apply mask while fine tuning?
#53
pranoy-k
closed
5 years ago
4
Implementation of Seq2Seq with Transformer
#52
bhedayat
opened
5 years ago
0
why it didn't use softmax in computing multichoice loss
#51
eveliao
closed
5 years ago
1
problems about DoubleHeadModel implementation
#50
eveliao
closed
5 years ago
1
Use the embedding dropout
#49
sgugger
closed
5 years ago
1
Potentially incorrect regex in text_utils.py
#48
schmmd
opened
5 years ago
0
Why is output vocab including positional embeddings?
#47
OanaMariaCamburu
opened
6 years ago
2
Retrain the LM on new dataset?
#46
fabrahman
opened
6 years ago
0
Encoder paddings influence results?
#45
OanaMariaCamburu
opened
6 years ago
0
How does position embedding implementation work?
#44
bcserna
closed
6 years ago
2
How should one modify the code to successfully run text classification?
#43
davidefiocco
closed
5 years ago
7
help to understand bpe logic
#42
BogdanDidenko
closed
6 years ago
2
How is the file "cloze_test_test__spring2016 - cloze_test_ALL_test.csv" created?
#41
luffycodes
opened
6 years ago
5
How to create transforms for entailment task?
#40
lordzuko
opened
6 years ago
12
Confused by multiply n to both clf_logits and clf_losses?
#39
Vimos
opened
6 years ago
3
Vocabulary size code explanation and occasionally shape error
#38
Vimos
closed
6 years ago
2
Removing unused decoder for TransformedModel class.
#37
rodgzilla
closed
6 years ago
1
Results and questions on text generation experiments with pretrained LM model
#36
xiaoda99
closed
6 years ago
10
Text generation with pretrained LM model
#35
xiaoda99
closed
5 years ago
2
So we can not change the word embedding with the pretrained LM?
#34
herbertchen1
opened
6 years ago
1
How to use Inference?
#33
masati91
opened
6 years ago
4
Redundant decoder
#32
joshim5
closed
6 years ago
1
Avoid model overfitting
#31
BangLiu
opened
6 years ago
9
finetuning the model for NLI, but using sentence embeddings instead of word embeddings
#30
saurabhvyas
closed
6 years ago
2
Instructions on how to train a language model from scratch
#29
Froskekongen
opened
6 years ago
9
loading pretrained open ai model
#28
mehdimashayekhi
opened
6 years ago
3
Add comment explaining the encoding of the position information.
#27
sharpsy
closed
6 years ago
1
Clarifying last step of the 'transform_roc' function
#26
sharpsy
opened
6 years ago
7
Simplifying the use of the model to perform different tasks
#25
rodgzilla
closed
6 years ago
0
Having various network heads
#24
rodgzilla
opened
6 years ago
1
Removing unused field of the Transformer class.
#23
rodgzilla
closed
6 years ago
1
Noise shape dropout
#22
rodgzilla
closed
6 years ago
3
Can someone explain this line?
#21
teucer
opened
6 years ago
4
Removing python 2.7 check from Travis.
#20
rodgzilla
closed
6 years ago
1
what is the use of dropout in the Transformer?
#19
teucer
opened
6 years ago
2
Using the ellipsis to allow the function to work with dim 3 or 4 tensors
#18
rodgzilla
closed
6 years ago
1
Cleaning the data encoding code.
#17
rodgzilla
closed
6 years ago
1
Updating README.md
#16
rodgzilla
closed
6 years ago
1
Fixing issue #13 and #14.
#15
rodgzilla
closed
6 years ago
1
DoubleHeadModel is not instanciated when n_gpu <= 1
#14
rodgzilla
closed
6 years ago
1
Object is not specified
#13
Oktai15
closed
6 years ago
4
In `transform_roc`, why do we need `xmb[:, :, :, 1] `?
#12
FrankWork
opened
6 years ago
3
How does Dropout2d help in cloze task?
#11
sai-prasanna
closed
6 years ago
12
[WIP] Add multi-gpu support
#10
nottombrown
closed
6 years ago
2
Next