Laurent Mazare 539ead927a Update the Phi model to use the updated architecture. (#1580) ...

* Update the Phi model to use the updated architecture.

* Add more of the phi model.

* Repeat KV + caching.

* Apply the rotary embeddings.

* Add support for the new phi model in the phi example.

* Fix a couple glitches.

* Fix a couple more glitches.

2024-01-13 17:38:27 +01:00

..

activation.rs

Update the Phi model to use the updated architecture. (#1580)

2024-01-13 17:38:27 +01:00

batch_norm.rs

Do not implement Module for BatchNorm. (#1513)

2024-01-01 10:13:13 +01:00

conv.rs

Transposed conv1d in candle-nn. (#1252)

2023-11-03 11:18:25 +01:00

embedding.rs

Add clone to various nn layers. (#910)

2023-09-20 11:33:51 +01:00

encoding.rs

Simplify the one-hot implementation, support arbitrary rank. (#1514)

2024-01-01 11:40:17 +01:00

func.rs

Allow for different behavior between training and eval (#1213)

2023-10-29 07:53:09 +01:00

group_norm.rs

Add clone to various nn layers. (#910)

2023-09-20 11:33:51 +01:00

init.rs

VarBuilder cleanup (#627)

2023-08-27 18:03:26 +01:00

layer_norm.rs

Add weight and bias functions to LayerNorm (#1306)

2023-11-09 16:09:01 +01:00

lib.rs

Add one-hot/cold encoding (#1489)

2024-01-01 11:18:40 +01:00

linear.rs

Add the prelu layer. (#1402)

2023-12-03 16:06:09 +00:00

loss.rs

Add Binary Cross Entropy With Logit Loss to nn crate (#1157)

2023-10-23 17:12:44 +01:00

ops.rs

Clippy pass.

2023-12-18 15:22:43 +01:00

optim.rs

Expose AdamW parameters (#1449)

2023-12-16 18:41:56 -06:00

rnn.rs

More general seq forward functions for RNNs. (#1050)

2023-10-07 15:08:01 +01:00

sequential.rs

Make func cloneable. (#1137)

2023-10-20 16:28:50 +01:00

var_builder.rs

Fix a couple typos (#1451)

2023-12-17 05:20:05 -06:00

var_map.rs

Use AsRef<str> for set_one. (#1033)

2023-10-05 06:05:44 +01:00