From 2120cb24d4585270158e885070bb53992079879c Mon Sep 17 00:00:00 2001 From: ydshieh Date: Wed, 16 Feb 2022 17:11:16 +0100 Subject: [PATCH 1/2] fix doc --- src/transformers/models/funnel/configuration_funnel.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/transformers/models/funnel/configuration_funnel.py b/src/transformers/models/funnel/configuration_funnel.py index 9496a63f48d2..1fe5f1c7a6d3 100644 --- a/src/transformers/models/funnel/configuration_funnel.py +++ b/src/transformers/models/funnel/configuration_funnel.py @@ -77,7 +77,7 @@ class FunnelConfig(PretrainedConfig): type_vocab_size (`int`, *optional*, defaults to 3): The vocabulary size of the `token_type_ids` passed when calling [`FunnelModel`] or [`TFFunnelModel`]. initializer_range (`float`, *optional*, defaults to 0.1): - The standard deviation of the *uniform initializer* for initializing all weight matrices in attention + The upper bound of the *uniform initializer* for initializing all weight matrices in attention layers. initializer_std (`float`, *optional*): The standard deviation of the *normal initializer* for initializing the embedding matrix and the weight of From 4a372acc6273601b578254000b6ef04d522c3f32 Mon Sep 17 00:00:00 2001 From: ydshieh Date: Wed, 16 Feb 2022 17:38:22 +0100 Subject: [PATCH 2/2] make style --- src/transformers/models/funnel/configuration_funnel.py | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/transformers/models/funnel/configuration_funnel.py b/src/transformers/models/funnel/configuration_funnel.py index 1fe5f1c7a6d3..5684427cb7a7 100644 --- a/src/transformers/models/funnel/configuration_funnel.py +++ b/src/transformers/models/funnel/configuration_funnel.py @@ -77,8 +77,7 @@ class FunnelConfig(PretrainedConfig): type_vocab_size (`int`, *optional*, defaults to 3): The vocabulary size of the `token_type_ids` passed when calling [`FunnelModel`] or [`TFFunnelModel`]. initializer_range (`float`, *optional*, defaults to 0.1): - The upper bound of the *uniform initializer* for initializing all weight matrices in attention - layers. + The upper bound of the *uniform initializer* for initializing all weight matrices in attention layers. initializer_std (`float`, *optional*): The standard deviation of the *normal initializer* for initializing the embedding matrix and the weight of linear layers. Will default to 1 for the embedding matrix and the value given by Xavier initialization for