text-recognizer.git - A neural network that reads handwritten paragraphs.

Age	Commit message (Expand)	Author
2021-11-28	Refactor attention layer module	Gustaf Rydholm
2021-11-27	Revert "Remove ViT"	Gustaf Rydholm
2021-11-27	Revert "Revert "Remove default transformer""	Gustaf Rydholm
2021-11-27	Revert "Remove default transformer"	Gustaf Rydholm
2021-11-22	Fix import efficientnet	Gustaf Rydholm
2021-11-22	Remove unused utils	Gustaf Rydholm
2021-11-22	Format conv transformer	Gustaf Rydholm
2021-11-22	Move efficientnet from encoder dir	Gustaf Rydholm
2021-11-21	Remove criterion and ctc loss	Gustaf Rydholm
2021-11-21	Format files	Gustaf Rydholm
2021-11-21	Add axial encoder to conv transformer	Gustaf Rydholm
2021-11-21	Remove VQVAE stuff, did not work...	Gustaf Rydholm
2021-11-21	Add axial transformer	Gustaf Rydholm
2021-11-21	Remove label smoothing loss	Gustaf Rydholm
2021-11-17	Update VqTransformer	Gustaf Rydholm
2021-11-17	Update encoder fun in conv_transformer	Gustaf Rydholm
2021-11-17	Update vqvae with new quantizer	Gustaf Rydholm
2021-11-17	Add new quantizer	Gustaf Rydholm
2021-11-17	Remove local attention	Gustaf Rydholm
2021-11-17	Replace last two layers in efficientnet with one	Gustaf Rydholm
2021-11-17	Format efficientnet	Gustaf Rydholm
2021-11-17	Update vq transformer lit model	Gustaf Rydholm
2021-11-17	Fix import path vqganloss	Gustaf Rydholm
2021-11-07	Fix dropout in efficientnet	Gustaf Rydholm
2021-11-07	Add dropout to final layer in efficientnet	Gustaf Rydholm
2021-11-07	Fix mbconv sub modules	Gustaf Rydholm
2021-11-06	Fix efficientnet incorrect channel calculation	Gustaf Rydholm
2021-11-06	Format efficientnet	Gustaf Rydholm
2021-11-05	Remove conv attention	Gustaf Rydholm
2021-11-05	Rename mask to input_mask	Gustaf Rydholm
2021-11-05	Remove out_channels as a settable parameter in effnet	Gustaf Rydholm
2021-11-05	Rename iam lines transforms	Gustaf Rydholm
2021-11-05	Format pad transform	Gustaf Rydholm
2021-11-03	Add pad transform	Gustaf Rydholm
2021-11-03	Remove unused args from conv transformer	Gustaf Rydholm
2021-11-03	Remove pixelcnn	Gustaf Rydholm
2021-11-03	Fix output from attn modules	Gustaf Rydholm
2021-11-03	Fix local attn to work with any input length	Gustaf Rydholm
2021-11-03	Update output shape from attn module	Gustaf Rydholm
2021-11-01	Add check for positional encoding in attn layer	Gustaf Rydholm
2021-11-01	Fix bugs in local attention	Gustaf Rydholm
2021-11-01	Refactor transformer layer	Gustaf Rydholm
2021-11-01	Fix self attention module	Gustaf Rydholm
2021-11-01	Update rotary embedding	Gustaf Rydholm
2021-10-28	Update rotary embedding	Gustaf Rydholm
2021-10-28	Remove 2D positional embedding	Gustaf Rydholm
2021-10-28	Remove absolute positional embedding	Gustaf Rydholm
2021-10-28	Add check for position embedding	Gustaf Rydholm
2021-10-28	Fix multihead local attention	Gustaf Rydholm
2021-10-28	Fix bug with local attention	Gustaf Rydholm