mygpt.git
2022-08-08 Francois FleuretAdded args.learning_rate_end for an exponential decay.
2022-08-08 Francois FleuretAdded the small weight embedding + id layer norm inits.
2022-08-07 Francois FleuretAdded the rng state in the checkpoint.
2022-08-07 Francois FleuretAdded the small-weight embedding initialization.
2022-07-30 Francois FleuretUpdate.
2022-07-30 Francois FleuretUpdate.
2022-07-30 Francois FleuretUpdate.
2022-07-30 Francois FleuretFixed a bug when there are no squares.
2022-07-29 Francois FleuretOCDC
2022-07-29 Francois FleuretUpdate.
2022-07-29 Francois FleuretUpdate.
2022-07-29 Francois FleuretOCDC
2022-07-29 Francois FleuretUpdate.
2022-07-28 Francois FleuretOCDC
2022-07-28 Francois FleuretUpdate.
2022-07-27 Francois FleuretFixed stuff.
2022-07-27 Francois FleuretOCDC
2022-07-27 Francois FleuretOCDC
2022-07-27 Francois FleuretOCDC
2022-07-27 Francois FleuretCleaning up more.
2022-07-27 Francois FleuretOCD cosmectics
2022-07-27 Francois FleuretCosmetics.
2022-07-27 Francois FleuretUpdate.
2022-07-27 Francois FleuretOCD cosmetics.
2022-07-27 Francois FleuretCompute both the average number of requested and obtain...
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretFixed the size of w_o.
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretAdded a null token, which is the one to predict.
2022-07-26 Francois FleuretRemoved the Linear transformation since there is now...
2022-07-26 Francois FleuretUpdate.
2022-07-26 Francois FleuretMoved the input/output shift in the forward of the...
2022-07-25 Francois FleuretUpdate.
2022-07-25 Francois FleuretOCD update
2022-07-25 Francois FleuretAdded --no_checkpoint
2022-07-25 Francois FleuretUpdate.
2022-07-25 Francois FleuretInitialize properly w_o.
2022-07-25 Francois FleuretUpdate.
2022-07-25 Francois FleuretUpdate.
2022-07-25 Francois FleuretAdded the (missing) W_o
2022-07-25 Francois FleuretRemoved the default image size.
2022-07-16 Francois FleuretUpdate.
2022-07-16 Francois FleuretUpdate.
2022-07-16 Francois FleuretUpdate.
2022-07-16 Francois FleuretUpdate.
2022-07-16 Francois FleuretUpdate.
2022-07-16 Francois FleuretUpdate.
2022-07-15 Francois FleuretUpdate.
2022-07-15 Francois FleuretUpdate.
2022-07-15 Francois FleuretUpdate.
2022-07-02 Francois FleuretUpdate.
2022-07-01 Francois FleuretUpdate.
2022-07-01 Francois FleuretAdd cross-attention to QKVAttention.
2022-06-20 Francois FleuretFinalized PicoCLVR with "many colors".
2022-06-17 Francois FleuretUpdate.
2022-06-17 Francois FleuretUpdate.
2022-06-17 Francois FleuretUpdate.
2022-06-13 Francois FleuretUpdate.
2022-06-10 Francois FleuretUpdate.
2022-05-12 Francois FleuretUpdate.
2022-05-11 Francois FleuretUpdate.
2022-04-29 Francois FleuretUpdate.
2022-04-29 Francois FleuretUpdate.
2022-04-29 Francois FleuretUpdate.
2022-04-26 Francois FleuretUpdate.
2022-04-26 Francois FleuretUpdate.
2022-04-26 Francois FleuretUpdate.
2022-04-25 Francois FleuretFixed an asymmetry between top / bottom and between...
2022-04-25 Francois FleuretUpdate.
2022-04-24 Francois FleuretUpdate.
2022-04-24 Francois FleuretUpdate.
2022-04-24 Francois FleuretUpdate.
2022-04-24 Francois FleuretInitial commit