From 3ab82ad36bce6fa698a13a029a0694b75a5947b7 Mon Sep 17 00:00:00 2001
From: Gustaf Rydholm <gustaf.rydholm@gmail.com>
Date: Fri, 6 Aug 2021 02:42:45 +0200
Subject: Fix VQVAE into en/decoder, bug in wandb artifact code uploading

---
 text_recognizer/networks/conv_transformer.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'text_recognizer/networks/conv_transformer.py')

diff --git a/text_recognizer/networks/conv_transformer.py b/text_recognizer/networks/conv_transformer.py
index f3ba49d..b1a101e 100644
--- a/text_recognizer/networks/conv_transformer.py
+++ b/text_recognizer/networks/conv_transformer.py
@@ -4,7 +4,6 @@ from typing import Tuple
 
 from torch import nn, Tensor
 
-from text_recognizer.networks.encoders.efficientnet import EfficientNet
 from text_recognizer.networks.transformer.layers import Decoder
 from text_recognizer.networks.transformer.positional_encodings import (
     PositionalEncoding,
@@ -18,15 +17,17 @@ class ConvTransformer(nn.Module):
     def __init__(
         self,
         input_dims: Tuple[int, int, int],
+        encoder_dim: int,
         hidden_dim: int,
         dropout_rate: float,
         num_classes: int,
         pad_index: Tensor,
-        encoder: EfficientNet,
+        encoder: nn.Module,
         decoder: Decoder,
     ) -> None:
         super().__init__()
         self.input_dims = input_dims
+        self.encoder_dim = encoder_dim
         self.hidden_dim = hidden_dim
         self.dropout_rate = dropout_rate
         self.num_classes = num_classes
@@ -38,7 +39,7 @@ class ConvTransformer(nn.Module):
         # positional encoding.
         self.latent_encoder = nn.Sequential(
             nn.Conv2d(
-                in_channels=self.encoder.out_channels,
+                in_channels=self.encoder_dim,
                 out_channels=self.hidden_dim,
                 kernel_size=1,
             ),
-- 
cgit v1.2.3-70-g09d2