Working on fixing decoder transformer

13 files changed, 260 insertions, 1092 deletions

diff --git a/notebooks/00-scratch-pad.ipynb b/notebooks/00-scratch-pad.ipynb
index 3c44f2b..8db843c 100644
--- a/notebooks/00-scratch-pad.ipynb
+++ b/notebooks/00-scratch-pad.ipynb
@@ -2,18 +2,9 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 1,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "The autoreload extension is already loaded. To reload it, use:\n",
-      "  %reload_ext autoreload\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "%load_ext autoreload\n",
     "%autoreload 2\n",
@@ -34,7 +25,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
@@ -43,7 +34,7 @@
        "True"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -54,7 +45,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -63,187 +54,224 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "decoder = Decoder(dim=128, depth=4, num_heads=8, ff_kwargs={}, attn_kwargs={}, cross_attend=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "decoder.cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from text_recognizer.networks.transformer.transformer import Transformer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
-    "decoder = Decoder(dim=256, depth=4, num_heads=8, ff_kwargs={}, attn_kwargs={}, cross_attend=True)"
+    "transformer_decoder = Transformer(num_tokens=90, max_seq_len=690, attn_layers=decoder, emb_dim=128, emb_dropout=0.1)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "Decoder(\n",
-       "  (layers): ModuleList(\n",
-       "    (0): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "Transformer(\n",
+       "  (attn_layers): Decoder(\n",
+       "    (layers): ModuleList(\n",
+       "      (0): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (1): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (1): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (2): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): FeedForward(\n",
-       "        (mlp): Sequential(\n",
-       "          (0): GEGLU(\n",
-       "            (fc): Linear(in_features=256, out_features=2048, bias=True)\n",
+       "      (2): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): FeedForward(\n",
+       "          (mlp): Sequential(\n",
+       "            (0): GEGLU(\n",
+       "              (fc): Linear(in_features=128, out_features=1024, bias=True)\n",
+       "            )\n",
+       "            (1): Dropout(p=0.0, inplace=False)\n",
+       "            (2): Linear(in_features=512, out_features=128, bias=True)\n",
        "          )\n",
-       "          (1): Dropout(p=0.0, inplace=False)\n",
-       "          (2): Linear(in_features=1024, out_features=256, bias=True)\n",
        "        )\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (3): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (3): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (4): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (4): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (5): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): FeedForward(\n",
-       "        (mlp): Sequential(\n",
-       "          (0): GEGLU(\n",
-       "            (fc): Linear(in_features=256, out_features=2048, bias=True)\n",
+       "      (5): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): FeedForward(\n",
+       "          (mlp): Sequential(\n",
+       "            (0): GEGLU(\n",
+       "              (fc): Linear(in_features=128, out_features=1024, bias=True)\n",
+       "            )\n",
+       "            (1): Dropout(p=0.0, inplace=False)\n",
+       "            (2): Linear(in_features=512, out_features=128, bias=True)\n",
        "          )\n",
-       "          (1): Dropout(p=0.0, inplace=False)\n",
-       "          (2): Linear(in_features=1024, out_features=256, bias=True)\n",
        "        )\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (6): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (6): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (7): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (7): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (8): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): FeedForward(\n",
-       "        (mlp): Sequential(\n",
-       "          (0): GEGLU(\n",
-       "            (fc): Linear(in_features=256, out_features=2048, bias=True)\n",
+       "      (8): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): FeedForward(\n",
+       "          (mlp): Sequential(\n",
+       "            (0): GEGLU(\n",
+       "              (fc): Linear(in_features=128, out_features=1024, bias=True)\n",
+       "            )\n",
+       "            (1): Dropout(p=0.0, inplace=False)\n",
+       "            (2): Linear(in_features=512, out_features=128, bias=True)\n",
        "          )\n",
-       "          (1): Dropout(p=0.0, inplace=False)\n",
-       "          (2): Linear(in_features=1024, out_features=256, bias=True)\n",
        "        )\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (9): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (9): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (10): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): Attention(\n",
-       "        (qkv_fn): Sequential(\n",
-       "          (0): Linear(in_features=256, out_features=49152, bias=False)\n",
-       "          (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "      (10): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): Attention(\n",
+       "          (qkv_fn): Sequential(\n",
+       "            (0): Linear(in_features=128, out_features=24576, bias=False)\n",
+       "            (1): Rearrange('b n (qkv h d) -> qkv b h n d', qkv=3, h=8)\n",
+       "          )\n",
+       "          (dropout): Dropout(p=0.0, inplace=False)\n",
+       "          (fc): Linear(in_features=8192, out_features=128, bias=True)\n",
        "        )\n",
-       "        (dropout): Dropout(p=0.0, inplace=False)\n",
-       "        (fc): Linear(in_features=16384, out_features=256, bias=True)\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
-       "    )\n",
-       "    (11): ModuleList(\n",
-       "      (0): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n",
-       "      (1): FeedForward(\n",
-       "        (mlp): Sequential(\n",
-       "          (0): GEGLU(\n",
-       "            (fc): Linear(in_features=256, out_features=2048, bias=True)\n",
+       "      (11): ModuleList(\n",
+       "        (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "        (1): FeedForward(\n",
+       "          (mlp): Sequential(\n",
+       "            (0): GEGLU(\n",
+       "              (fc): Linear(in_features=128, out_features=1024, bias=True)\n",
+       "            )\n",
+       "            (1): Dropout(p=0.0, inplace=False)\n",
+       "            (2): Linear(in_features=512, out_features=128, bias=True)\n",
        "          )\n",
-       "          (1): Dropout(p=0.0, inplace=False)\n",
-       "          (2): Linear(in_features=1024, out_features=256, bias=True)\n",
        "        )\n",
+       "        (2): Residual()\n",
        "      )\n",
-       "      (2): Residual()\n",
        "    )\n",
        "  )\n",
+       "  (token_emb): Embedding(90, 128)\n",
+       "  (emb_dropout): Dropout(p=0.1, inplace=False)\n",
+       "  (pos_emb): AbsolutePositionalEmbedding(\n",
+       "    (emb): Embedding(690, 128)\n",
+       "  )\n",
+       "  (project_emb): Identity()\n",
+       "  (norm): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
+       "  (logits): Linear(in_features=128, out_features=90, bias=True)\n",
        ")"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "decoder.cuda()"
+    "transformer_decoder.cuda()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -252,16 +280,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from functools import partial"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -275,7 +294,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -284,7 +303,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -298,29 +317,95 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 12,
    "metadata": {},
    "outputs": [],
    "source": [
-    "t = torch.randn(32, 1, 576, 640).cuda()"
+    "t = torch.randn(16, 1, 576, 640).cuda()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 13,
    "metadata": {},
    "outputs": [],
    "source": [
-    "v(t).shape"
+    "o = v(t)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 22,
    "metadata": {},
    "outputs": [],
    "source": [
-    "from text_recognizer.networks.encoders.efficientnet import EfficientNet"
+    "caption = torch.randint(0, 90, (16, 690)).cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([16, 360, 128])"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "o.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([16, 690])"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "caption.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "forward() missing 2 required positional arguments: 'context' and 'context_mask'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m<ipython-input-29-2290911ad81b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtransformer_decoder\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcaption\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcontext\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# (1, 1024, 20000)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[0;32m~/.cache/pypoetry/virtualenvs/text-recognizer-ejNaVa9M-py3.9/lib/python3.9/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    887\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 889\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    890\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    891\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m~/projects/text-recognizer/text_recognizer/networks/transformer/transformer.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x, mask, return_embeddings, **kwargs)\u001b[0m\n\u001b[1;32m     58\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     59\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mproject_emb\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 60\u001b[0;31m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattn_layers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     61\u001b[0m         \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogits\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mreturn_embeddings\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     62\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mout\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m~/.cache/pypoetry/virtualenvs/text-recognizer-ejNaVa9M-py3.9/lib/python3.9/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    887\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 889\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    890\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    891\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m~/projects/text-recognizer/text_recognizer/networks/transformer/layers.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x, context, mask, context_mask)\u001b[0m\n\u001b[1;32m     89\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     90\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mlayer_type\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"a\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 91\u001b[0;31m                 \u001b[0mout\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mblock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mrotary_pos_emb\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mrotary_pos_emb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     92\u001b[0m             \u001b[0;32melif\u001b[0m \u001b[0mlayer_type\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"c\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     93\u001b[0m                 \u001b[0mout\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mblock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcontext\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcontext\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmask\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcontext_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcontext_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;32m~/.cache/pypoetry/virtualenvs/text-recognizer-ejNaVa9M-py3.9/lib/python3.9/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m    887\u001b[0m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_slow_forward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    888\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 889\u001b[0;31m             \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    890\u001b[0m         for hook in itertools.chain(\n\u001b[1;32m    891\u001b[0m                 \u001b[0m_global_forward_hooks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+      "\u001b[0;31mTypeError\u001b[0m: forward() missing 2 required positional arguments: 'context' and 'context_mask'"
+     ]
+    }
+   ],
+   "source": [
+    "transformer_decoder(caption, context = o) # (1, 1024, 20000)"
    ]
   },
   {
@@ -329,7 +414,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "en = EfficientNet()"
+    "from text_recognizer.networks.encoders.efficientnet import EfficientNet"
    ]
   },
   {
@@ -338,7 +423,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "(576, 640) // (8, 8)"
+    "en = EfficientNet()"
    ]
   },
   {
@@ -347,7 +432,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "(576 // 32) ** 2"
+    "en.cuda()"
    ]
   },
   {
@@ -482,24 +567,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from text_recognizer.networks.backbones.efficientnet import EfficientNet"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "en = EfficientNet()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
     "datum = torch.randn([2, 1, 576, 640])"
    ]
   },
@@ -536,7 +603,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "en(datum).shape"
+    "en(t).shape"
    ]
   },
   {
@@ -752,7 +819,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.1"
+   "version": "3.9.5"
   }
  },
  "nbformat": 4,
diff --git a/notebooks/03-look-at-iam-paragraphs.ipynb b/notebooks/03-look-at-iam-paragraphs.ipynb
index b72019b..e05704d 100644
--- a/notebooks/03-look-at-iam-paragraphs.ipynb
+++ b/notebooks/03-look-at-iam-paragraphs.ipynb
@@ -505,7 +505,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.1"
+   "version": "3.9.5"
   }
  },
  "nbformat": 4,
diff --git a/pyproject.toml b/pyproject.toml
index 33c7287..c3f6ebc 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -2,7 +2,7 @@
 name = "text-recognizer"
 version = "0.1.0"
 authors = ["aktersnurra <gustaf.rydholm@gmail.com>"]
-description = "A text recognizer using best pratices in python and deep learning."
+description = "Text recognition software using best pratices in python."
 license = "MIT"
 readme = "README.md"
 homepage = "https://github.com/aktersnurra/text-recognizer"
diff --git a/text_recognizer/networks/cnn_transformer.py b/text_recognizer/networks/cnn_transformer.py
deleted file mode 100644
index 80798e1..0000000
--- a/text_recognizer/networks/cnn_transformer.py
+++ /dev/null
@@ -1,182 +0,0 @@
-# """A Transformer with a cnn backbone.
-#
-# The network encodes a image with a convolutional backbone to a latent representation,
-# i.e. feature maps. A 2d positional encoding is applied to the feature maps for
-# spatial information. The resulting feature are then set to a transformer decoder
-# together with the target tokens.
-#
-# TODO: Local attention for lower layer in attention.
-#
-# """
-# import importlib
-# import math
-# from typing import Dict, Optional, Union, Sequence, Type
-#
-# from einops import rearrange
-# from omegaconf import DictConfig, OmegaConf
-# import torch
-# from torch import nn
-# from torch import Tensor
-#
-# from text_recognizer.data.emnist import NUM_SPECIAL_TOKENS
-# from text_recognizer.networks.transformer import (
-#     Decoder,
-#     DecoderLayer,
-#     PositionalEncoding,
-#     PositionalEncoding2D,
-#     target_padding_mask,
-# )
-#
-# NUM_WORD_PIECES = 1000
-#
-#
-# class CNNTransformer(nn.Module):
-#     def __init__(
-#         self,
-#         input_dim: Sequence[int],
-#         output_dims: Sequence[int],
-#         encoder: Union[DictConfig, Dict],
-#         vocab_size: Optional[int] = None,
-#         num_decoder_layers: int = 4,
-#         hidden_dim: int = 256,
-#         num_heads: int = 4,
-#         expansion_dim: int = 1024,
-#         dropout_rate: float = 0.1,
-#         transformer_activation: str = "glu",
-#         *args,
-#         **kwargs,
-#     ) -> None:
-#         super().__init__()
-#         self.vocab_size = (
-#             NUM_WORD_PIECES + NUM_SPECIAL_TOKENS if vocab_size is None else vocab_size
-#         )
-#         self.pad_index = 3  # TODO: fix me
-#         self.hidden_dim = hidden_dim
-#         self.max_output_length = output_dims[0]
-#
-#         # Image backbone
-#         self.encoder = self._configure_encoder(encoder)
-#         self.encoder_proj = nn.Conv2d(256, hidden_dim, kernel_size=1)
-#         self.feature_map_encoding = PositionalEncoding2D(
-#             hidden_dim=hidden_dim, max_h=input_dim[1], max_w=input_dim[2]
-#         )
-#
-#         # Target token embedding
-#         self.trg_embedding = nn.Embedding(self.vocab_size, hidden_dim)
-#         self.trg_position_encoding = PositionalEncoding(
-#             hidden_dim, dropout_rate, max_len=output_dims[0]
-#         )
-#
-#         # Transformer decoder
-#         self.decoder = Decoder(
-#             decoder_layer=DecoderLayer(
-#                 hidden_dim=hidden_dim,
-#                 num_heads=num_heads,
-#                 expansion_dim=expansion_dim,
-#                 dropout_rate=dropout_rate,
-#                 activation=transformer_activation,
-#             ),
-#             num_layers=num_decoder_layers,
-#             norm=nn.LayerNorm(hidden_dim),
-#         )
-#
-#         # Classification head
-#         self.head = nn.Linear(hidden_dim, self.vocab_size)
-#
-#         # Initialize weights
-#         self._init_weights()
-#
-#     def _init_weights(self) -> None:
-#         """Initialize network weights."""
-#         self.trg_embedding.weight.data.uniform_(-0.1, 0.1)
-#         self.head.bias.data.zero_()
-#         self.head.weight.data.uniform_(-0.1, 0.1)
-#
-#         nn.init.kaiming_normal_(
-#             self.encoder_proj.weight.data,
-#             a=0,
-#             mode="fan_out",
-#             nonlinearity="relu",
-#         )
-#         if self.encoder_proj.bias is not None:
-#             _, fan_out = nn.init._calculate_fan_in_and_fan_out(
-#                 self.encoder_proj.weight.data
-#             )
-#             bound = 1 / math.sqrt(fan_out)
-#             nn.init.normal_(self.encoder_proj.bias, -bound, bound)
-#
-#     @staticmethod
-#     def _configure_encoder(encoder: Union[DictConfig, Dict]) -> Type[nn.Module]:
-#         encoder = OmegaConf.create(encoder)
-#         args = encoder.args or {}
-#         network_module = importlib.import_module("text_recognizer.networks")
-#         encoder_class = getattr(network_module, encoder.type)
-#         return encoder_class(**args)
-#
-#     def encode(self, image: Tensor) -> Tensor:
-#         """Extracts image features with backbone.
-#
-#         Args:
-#             image (Tensor): Image(s) of handwritten text.
-#
-#         Retuns:
-#             Tensor: Image features.
-#
-#         Shapes:
-#             - image: :math: `(B, C, H, W)`
-#             - latent: :math: `(B, T, C)`
-#
-#         """
-#         # Extract image features.
-#         image_features = self.encoder(image)
-#         image_features = self.encoder_proj(image_features)
-#
-#         # Add 2d encoding to the feature maps.
-#         image_features = self.feature_map_encoding(image_features)
-#
-#         # Collapse features maps height and width.
-#         image_features = rearrange(image_features, "b c h w -> b (h w) c")
-#         return image_features
-#
-#     def decode(self, memory: Tensor, trg: Tensor) -> Tensor:
-#         """Decodes image features with transformer decoder."""
-#         trg_mask = target_padding_mask(trg=trg, pad_index=self.pad_index)
-#         trg = self.trg_embedding(trg) * math.sqrt(self.hidden_dim)
-#         trg = rearrange(trg, "b t d -> t b d")
-#         trg = self.trg_position_encoding(trg)
-#         trg = rearrange(trg, "t b d -> b t d")
-#         out = self.decoder(trg=trg, memory=memory, trg_mask=trg_mask, memory_mask=None)
-#         logits = self.head(out)
-#         return logits
-#
-#     def forward(self, image: Tensor, trg: Tensor) -> Tensor:
-#         image_features = self.encode(image)
-#         output = self.decode(image_features, trg)
-#         output = rearrange(output, "b t c -> b c t")
-#         return output
-#
-#     def predict(self, image: Tensor) -> Tensor:
-#         """Transcribes text in image(s)."""
-#         bsz = image.shape[0]
-#         image_features = self.encode(image)
-#
-#         output_tokens = (
-#             (torch.ones((bsz, self.max_output_length)) * self.pad_index)
-#             .type_as(image)
-#             .long()
-#         )
-#         output_tokens[:, 0] = self.start_index
-#         for i in range(1, self.max_output_length):
-#             trg = output_tokens[:, :i]
-#             output = self.decode(image_features, trg)
-#             output = torch.argmax(output, dim=-1)
-#             output_tokens[:, i] = output[-1:]
-#
-#         # Set all tokens after end token to be padding.
-#         for i in range(1, self.max_output_length):
-#             indices = output_tokens[:, i - 1] == self.end_index | (
-#                 output_tokens[:, i - 1] == self.pad_index
-#             )
-#             output_tokens[indices, i] = self.pad_index
-#
-#         return output_tokens
diff --git a/text_recognizer/networks/transducer/__init__.py b/text_recognizer/networks/transducer/__init__.py
deleted file mode 100644
index 8c19a01..0000000
--- a/text_recognizer/networks/transducer/__init__.py
+++ /dev/null
@@ -1,3 +0,0 @@
-"""Transducer modules."""
-from .tds_conv import TDS2d
-from .transducer import load_transducer_loss, Transducer
diff --git a/text_recognizer/networks/transducer/tds_conv.py b/text_recognizer/networks/transducer/tds_conv.py
deleted file mode 100644
index 5fb8ba9..0000000
--- a/text_recognizer/networks/transducer/tds_conv.py
+++ /dev/null
@@ -1,208 +0,0 @@
-"""Time-Depth Separable Convolutions.
-
-References:
-    https://arxiv.org/abs/1904.02619
-    https://arxiv.org/pdf/2010.01003.pdf
-
-Code stolen from:
-    https://github.com/facebookresearch/gtn_applications
-
-
-"""
-from typing import List, Tuple
-
-from einops import rearrange
-import gtn
-import numpy as np
-import torch
-from torch import nn
-from torch import Tensor
-
-
-class TDSBlock2d(nn.Module):
-    """Internal block of a 2D TDSC network."""
-
-    def __init__(
-        self,
-        in_channels: int,
-        img_depth: int,
-        kernel_size: Tuple[int],
-        dropout_rate: float,
-    ) -> None:
-        super().__init__()
-
-        self.in_channels = in_channels
-        self.img_depth = img_depth
-        self.kernel_size = kernel_size
-        self.dropout_rate = dropout_rate
-        self.fc_dim = in_channels * img_depth
-
-        # Network placeholders.
-        self.conv = None
-        self.mlp = None
-        self.instance_norm = None
-
-        self._build_block()
-
-    def _build_block(self) -> None:
-        # Convolutional block.
-        self.conv = nn.Sequential(
-            nn.Conv3d(
-                in_channels=self.in_channels,
-                out_channels=self.in_channels,
-                kernel_size=(1, self.kernel_size[0], self.kernel_size[1]),
-                padding=(0, self.kernel_size[0] // 2, self.kernel_size[1] // 2),
-            ),
-            nn.ReLU(inplace=True),
-            nn.Dropout(self.dropout_rate),
-        )
-
-        # MLP block.
-        self.mlp = nn.Sequential(
-            nn.Linear(self.fc_dim, self.fc_dim),
-            nn.ReLU(inplace=True),
-            nn.Dropout(self.dropout_rate),
-            nn.Linear(self.fc_dim, self.fc_dim),
-            nn.Dropout(self.dropout_rate),
-        )
-
-        # Instance norm.
-        self.instance_norm = nn.ModuleList(
-            [
-                nn.InstanceNorm2d(self.fc_dim, affine=True),
-                nn.InstanceNorm2d(self.fc_dim, affine=True),
-            ]
-        )
-
-    def forward(self, x: Tensor) -> Tensor:
-        """Forward pass.
-
-        Args:
-            x (Tensor): Input tensor.
-
-        Shape:
-            - x: :math: `(B, CD, H, W)`
-
-        Returns:
-            Tensor: Output tensor.
-
-        """
-        B, CD, H, W = x.shape
-        C, D = self.in_channels, self.img_depth
-        residual = x
-        x = rearrange(x, "b (c d) h w -> b c d h w", c=C, d=D)
-        x = self.conv(x)
-        x = rearrange(x, "b c d h w -> b (c d) h w")
-        x += residual
-
-        x = self.instance_norm[0](x)
-
-        x = self.mlp(x.transpose(1, 3)).transpose(1, 3) + x
-        x + self.instance_norm[1](x)
-
-        # Output shape: [B, CD, H, W]
-        return x
-
-
-class TDS2d(nn.Module):
-    """TDS Netowrk.
-
-    Structure is the following:
-        Downsample layer -> TDS2d group -> ... -> Linear output layer
-
-
-    """
-
-    def __init__(
-        self,
-        input_dim: int,
-        output_dim: int,
-        depth: int,
-        tds_groups: Tuple[int],
-        kernel_size: Tuple[int],
-        dropout_rate: float,
-        in_channels: int = 1,
-    ) -> None:
-        super().__init__()
-
-        self.in_channels = in_channels
-        self.input_dim = input_dim
-        self.output_dim = output_dim
-        self.depth = depth
-        self.tds_groups = tds_groups
-        self.kernel_size = kernel_size
-        self.dropout_rate = dropout_rate
-
-        self.tds = None
-        self.fc = None
-
-        self._build_network()
-
-    def _build_network(self) -> None:
-        in_channels = self.in_channels
-        modules = []
-        stride_h = np.prod([grp["stride"][0] for grp in self.tds_groups])
-        if self.input_dim % stride_h:
-            raise RuntimeError(
-                f"Image height not divisible by total stride {stride_h}."
-            )
-
-        for tds_group in self.tds_groups:
-            # Add downsample layer.
-            out_channels = self.depth * tds_group["channels"]
-            modules.extend(
-                [
-                    nn.Conv2d(
-                        in_channels=in_channels,
-                        out_channels=out_channels,
-                        kernel_size=self.kernel_size,
-                        padding=(self.kernel_size[0] // 2, self.kernel_size[1] // 2),
-                        stride=tds_group["stride"],
-                    ),
-                    nn.ReLU(inplace=True),
-                    nn.Dropout(self.dropout_rate),
-                    nn.InstanceNorm2d(out_channels, affine=True),
-                ]
-            )
-
-            for _ in range(tds_group["num_blocks"]):
-                modules.append(
-                    TDSBlock2d(
-                        tds_group["channels"],
-                        self.depth,
-                        self.kernel_size,
-                        self.dropout_rate,
-                    )
-                )
-
-            in_channels = out_channels
-
-        self.tds = nn.Sequential(*modules)
-        self.fc = nn.Linear(in_channels * self.input_dim // stride_h, self.output_dim)
-
-    def forward(self, x: Tensor) -> Tensor:
-        """Forward pass.
-
-        Args:
-            x (Tensor): Input tensor.
-
-        Shape:
-            - x: :math: `(B, H, W)`
-
-        Returns:
-            Tensor: Output tensor.
-
-        """
-        if len(x.shape) == 4:
-            x = x.squeeze(1)  # Squeeze the channel dim away.
-
-        B, H, W = x.shape
-        x = rearrange(
-            x, "b (h1 h2) w -> b h1 h2 w", h1=self.in_channels, h2=H // self.in_channels
-        )
-        x = self.tds(x)
-
-        # x shape: [B, C, H, W]
-        x = rearrange(x, "b c h w -> b w (c h)")
-
-        return self.fc(x)
diff --git a/text_recognizer/networks/transducer/test.py b/text_recognizer/networks/transducer/test.py
deleted file mode 100644
index cadcecc..0000000
--- a/text_recognizer/networks/transducer/test.py
+++ /dev/null
@@ -1,60 +0,0 @@
-import torch
-from torch import nn
-
-from text_recognizer.networks.transducer import load_transducer_loss, Transducer
-import unittest
-
-
-class TestTransducer(unittest.TestCase):
-    def test_viterbi(self):
-        T = 5
-        N = 4
-        B = 2
-
-        # fmt: off
-        emissions1 = torch.tensor((
-            0, 4, 0, 1,
-            0, 2, 1, 1,
-            0, 0, 0, 2,
-            0, 0, 0, 2,
-            8, 0, 0, 2,
-            ),
-            dtype=torch.float,
-        ).view(T, N)
-        emissions2 = torch.tensor((
-            0, 2, 1, 7,
-            0, 2, 9, 1,
-            0, 0, 0, 2,
-            0, 0, 5, 2,
-            1, 0, 0, 2,
-            ),
-            dtype=torch.float,
-        ).view(T, N)
-        # fmt: on
-
-        # Test without blank:
-        labels = [[1, 3, 0], [3, 2, 3, 2, 3]]
-        transducer = Transducer(
-            tokens=["a", "b", "c", "d"],
-            graphemes_to_idx={"a": 0, "b": 1, "c": 2, "d": 3},
-            blank="none",
-        )
-        emissions = torch.stack([emissions1, emissions2], dim=0)
-        predictions = transducer.viterbi(emissions)
-        self.assertEqual([p.tolist() for p in predictions], labels)
-
-        # Test with blank without repeats:
-        labels = [[1, 0], [2, 2]]
-        transducer = Transducer(
-            tokens=["a", "b", "c"],
-            graphemes_to_idx={"a": 0, "b": 1, "c": 2},
-            blank="optional",
-            allow_repeats=False,
-        )
-        emissions = torch.stack([emissions1, emissions2], dim=0)
-        predictions = transducer.viterbi(emissions)
-        self.assertEqual([p.tolist() for p in predictions], labels)
-
-
-if __name__ == "__main__":
-    unittest.main()
diff --git a/text_recognizer/networks/transducer/transducer.py b/text_recognizer/networks/transducer/transducer.py
deleted file mode 100644
index d7e3d08..0000000
--- a/text_recognizer/networks/transducer/transducer.py
+++ /dev/null
@@ -1,410 +0,0 @@
-"""Transducer and the transducer loss function.py
-
-Stolen from:
-    https://github.com/facebookresearch/gtn_applications/blob/master/transducer.py
-
-"""
-from pathlib import Path
-import itertools
-from typing import Dict, List, Optional, Union, Tuple
-
-from loguru import logger
-import gtn
-import torch
-from torch import nn
-from torch import Tensor
-
-from text_recognizer.datasets.iam_preprocessor import Preprocessor
-
-
-def make_scalar_graph(weight) -> gtn.Graph:
-    scalar = gtn.Graph()
-    scalar.add_node(True)
-    scalar.add_node(False, True)
-    scalar.add_arc(0, 1, 0, 0, weight)
-    return scalar
-
-
-def make_chain_graph(sequence) -> gtn.Graph:
-    graph = gtn.Graph(False)
-    graph.add_node(True)
-    for i, s in enumerate(sequence):
-        graph.add_node(False, i == (len(sequence) - 1))
-        graph.add_arc(i, i + 1, s)
-    return graph
-
-
-def make_transitions_graph(
-    ngram: int, num_tokens: int, calc_grad: bool = False
-) -> gtn.Graph:
-    transitions = gtn.Graph(calc_grad)
-    transitions.add_node(True, ngram == 1)
-
-    state_map = {(): 0}
-
-    # First build transitions which include <s>:
-    for n in range(1, ngram):
-        for state in itertools.product(range(num_tokens), repeat=n):
-            in_idx = state_map[state[:-1]]
-            out_idx = transitions.add_node(False, ngram == 1)
-            state_map[state] = out_idx
-            transitions.add_arc(in_idx, out_idx, state[-1])
-
-    for state in itertools.product(range(num_tokens), repeat=ngram):
-        state_idx = state_map[state[:-1]]
-        new_state_idx = state_map[state[1:]]
-        # p(state[-1] | state[:-1])
-        transitions.add_arc(state_idx, new_state_idx, state[-1])
-
-    if ngram > 1:
-        # Build transitions which include </s>:
-        end_idx = transitions.add_node(False, True)
-        for in_idx in range(end_idx):
-            transitions.add_arc(in_idx, end_idx, gtn.epsilon)
-
-    return transitions
-
-
-def make_lexicon_graph(word_pieces: List, graphemes_to_idx: Dict) -> gtn.Graph:
-    """Constructs a graph which transduces letters to word pieces."""
-    graph = gtn.Graph(False)
-    graph.add_node(True, True)
-    for i, wp in enumerate(word_pieces):
-        prev = 0
-        for l in wp[:-1]:
-            n = graph.add_node()
-            graph.add_arc(prev, n, graphemes_to_idx[l], gtn.epsilon)
-            prev = n
-        graph.add_arc(prev, 0, graphemes_to_idx[wp[-1]], i)
-    graph.arc_sort()
-    return graph
-
-
-def make_token_graph(
-    token_list: List, blank: str = "none", allow_repeats: bool = True
-) -> gtn.Graph:
-    """Constructs a graph with all the individual token transition models."""
-    if not allow_repeats and blank != "optional":
-        raise ValueError("Must use blank='optional' if disallowing repeats.")
-
-    ntoks = len(token_list)
-    graph = gtn.Graph(False)
-
-    # Creating nodes
-    graph.add_node(True, True)
-    for i in range(ntoks):
-        # We can consume one or more consecutive word
-        # pieces for each emission:
-        # E.g. [ab, ab, ab] transduces to [ab]
-        graph.add_node(False, blank != "forced")
-
-    if blank != "none":
-        graph.add_node()
-
-    # Creating arcs
-    if blank != "none":
-        # Blank index is assumed to be last (ntoks)
-        graph.add_arc(0, ntoks + 1, ntoks, gtn.epsilon)
-        graph.add_arc(ntoks + 1, 0, gtn.epsilon)
-
-    for i in range(ntoks):
-        graph.add_arc((ntoks + 1) if blank == "forced" else 0, i + 1, i)
-        graph.add_arc(i + 1, i + 1, i, gtn.epsilon)
-
-        if allow_repeats:
-            if blank == "forced":
-                # Allow transitions from token to blank only
-                graph.add_arc(i + 1, ntoks + 1, ntoks, gtn.epsilon)
-            else:
-                # Allow transition from token to blank and all other tokens
-                graph.add_arc(i + 1, 0, gtn.epsilon)
-
-        else:
-            # allow transitions to blank and all other tokens except the same token
-            graph.add_arc(i + 1, ntoks + 1, ntoks, gtn.epsilon)
-            for j in range(ntoks):
-                if i != j:
-                    graph.add_arc(i + 1, j + 1, j, j)
-
-    return graph
-
-
-class TransducerLossFunction(torch.autograd.Function):
-    @staticmethod
-    def forward(
-        ctx,
-        inputs,
-        targets,
-        tokens,
-        lexicon,
-        transition_params=None,
-        transitions=None,
-        reduction="none",
-    ) -> Tensor:
-        B, T, C = inputs.shape
-
-        losses = [None] * B
-        emissions_graphs = [None] * B
-
-        if transitions is not None:
-            if transition_params is None:
-                raise ValueError("Specified transitions, but not transition params.")
-
-            cpu_data = transition_params.cpu().contiguous()
-            transitions.set_weights(cpu_data.data_ptr())
-            transitions.calc_grad = transition_params.requires_grad
-            transitions.zero_grad()
-
-        def process(b: int) -> None:
-            # Create emission graph:
-            emissions = gtn.linear_graph(T, C, inputs.requires_grad)
-            cpu_data = inputs[b].cpu().contiguous()
-            emissions.set_weights(cpu_data.data_ptr())
-            target = make_chain_graph(targets[b])
-            target.arc_sort(True)
-
-            # Create token tot grapheme decomposition graph
-            tokens_target = gtn.remove(gtn.project_output(gtn.compose(target, lexicon)))
-            tokens_target.arc_sort()
-
-            # Create alignment graph:
-            aligments = gtn.project_input(
-                gtn.remove(gtn.compose(tokens, tokens_target))
-            )
-            aligments.arc_sort()
-
-            # Add transitions scores:
-            if transitions is not None:
-                aligments = gtn.intersect(transitions, aligments)
-                aligments.arc_sort()
-
-            loss = gtn.forward_score(gtn.intersect(emissions, aligments))
-
-            # Normalize if needed:
-            if transitions is not None:
-                norm = gtn.forward_score(gtn.intersect(emissions, transitions))
-                loss = gtn.subtract(loss, norm)
-
-            losses[b] = gtn.negate(loss)
-
-            # Save for backward:
-            if emissions.calc_grad:
-                emissions_graphs[b] = emissions
-
-        gtn.parallel_for(process, range(B))
-
-        ctx.graphs = (losses, emissions_graphs, transitions)
-        ctx.input_shape = inputs.shape
-
-        # Optionally reduce by target length
-        if reduction == "mean":
-            scales = [(1 / len(t) if len(t) > 0 else 1.0) for t in targets]
-        else:
-            scales = [1.0] * B
-
-        ctx.scales = scales
-
-        loss = torch.tensor([l.item() * s for l, s in zip(losses, scales)])
-        return torch.mean(loss.to(inputs.device))
-
-    @staticmethod
-    def backward(ctx, grad_output) -> Tuple:
-        losses, emissions_graphs, transitions = ctx.graphs
-        scales = ctx.scales
-
-        B, T, C = ctx.input_shape
-        calc_emissions = ctx.needs_input_grad[0]
-        input_grad = torch.empty((B, T, C)) if calc_emissions else None
-
-        def process(b: int) -> None:
-            scale = make_scalar_graph(scales[b])
-            gtn.backward(losses[b], scale)
-            emissions = emissions_graphs[b]
-            if calc_emissions:
-                grad = emissions.grad().weights_to_numpy()
-                input_grad[b] = torch.tensor(grad).view(1, T, C)
-
-        gtn.parallel_for(process, range(B))
-
-        if calc_emissions:
-            input_grad = input_grad.to(grad_output.device)
-            input_grad *= grad_output / B
-
-        if ctx.needs_input_grad[4]:
-            grad = transitions.grad().weights_to_numpy()
-            transition_grad = torch.tensor(grad).to(grad_output.device)
-            transition_grad *= grad_output / B
-        else:
-            transition_grad = None
-
-        return (
-            input_grad,
-            None,  # target
-            None,  # tokens
-            None,  # lexicon
-            transition_grad,  # transition params
-            None,  # transitions graph
-            None,
-        )
-
-
-TransducerLoss = TransducerLossFunction.apply
-
-
-class Transducer(nn.Module):
-    def __init__(
-        self,
-        tokens: List,
-        graphemes_to_idx: Dict,
-        ngram: int = 0,
-        transitions: str = None,
-        blank: str = "none",
-        allow_repeats: bool = True,
-        reduction: str = "none",
-    ) -> None:
-        """A generic transducer loss function.
-
-        Args:
-            tokens (List) : A list of iterable objects (e.g. strings, tuples, etc)
-                representing the output tokens of the model (e.g. letters,
-                word-pieces, words). For example ["a", "b", "ab", "ba", "aba"]
-                could be a list of sub-word tokens.
-            graphemes_to_idx (dict) : A dictionary mapping grapheme units (e.g.
-                "a", "b", ..) to their corresponding integer index.
-            ngram (int) : Order of the token-level transition model. If `ngram=0`
-                then no transition model is used.
-            blank (string) : Specifies the usage of blank token
-                'none' - do not use blank token
-                'optional' - allow an optional blank inbetween tokens
-                'forced' - force a blank inbetween tokens (also referred to as garbage token)
-            allow_repeats (boolean) : If false, then we don't allow paths with
-                consecutive tokens in the alignment graph. This keeps the graph
-                unambiguous in the sense that the same input cannot transduce to
-                different outputs.
-        """
-        super().__init__()
-        if blank not in ["optional", "forced", "none"]:
-            raise ValueError(
-                "Invalid value specified for blank. Must be in ['optional', 'forced', 'none']"
-            )
-        self.tokens = make_token_graph(tokens, blank=blank, allow_repeats=allow_repeats)
-        self.lexicon = make_lexicon_graph(tokens, graphemes_to_idx)
-        self.ngram = ngram
-        if ngram > 0 and transitions is not None:
-            raise ValueError("Only one of ngram and transitions may be specified")
-
-        if ngram > 0:
-            transitions = make_transitions_graph(
-                ngram, len(tokens) + int(blank != "none"), True
-            )
-
-        if transitions is not None:
-            self.transitions = transitions
-            self.transitions.arc_sort()
-            self.transitions_params = nn.Parameter(
-                torch.zeros(self.transitions.num_arcs())
-            )
-        else:
-            self.transitions = None
-            self.transitions_params = None
-        self.reduction = reduction
-
-    def forward(self, inputs: Tensor, targets: Tensor) -> TransducerLoss:
-        TransducerLoss(
-            inputs,
-            targets,
-            self.tokens,
-            self.lexicon,
-            self.transitions_params,
-            self.transitions,
-            self.reduction,
-        )
-
-    def viterbi(self, outputs: Tensor) -> List[Tensor]:
-        B, T, C = outputs.shape
-
-        if self.transitions is not None:
-            cpu_data = self.transition_params.cpu().contiguous()
-            self.transitions.set_weights(cpu_data.data_ptr())
-            self.transitions.calc_grad = False
-
-        self.tokens.arc_sort()
-
-        paths = [None] * B
-
-        def process(b: int) -> None:
-            emissions = gtn.linear_graph(T, C, False)
-            cpu_data = outputs[b].cpu().contiguous()
-            emissions.set_weights(cpu_data.data_ptr())
-
-            if self.transitions is not None:
-                full_graph = gtn.intersect(emissions, self.transitions)
-            else:
-                full_graph = emissions
-
-            # Find the best path and remove back-off arcs:
-            path = gtn.remove(gtn.viterbi_path(full_graph))
-
-            # Left compose the viterbi path with the "aligment to token"
-            # transducer to get the outputs:
-            path = gtn.compose(path, self.tokens)
-
-            # When there are ambiguous paths (allow_repeats is true), we take
-            # the shortest:
-            path = gtn.viterbi_path(path)
-            path = gtn.remove(gtn.project_output(path))
-            paths[b] = path.labels_to_list()
-
-        gtn.parallel_for(process, range(B))
-        predictions = [torch.IntTensor(path) for path in paths]
-        return predictions
-
-
-def load_transducer_loss(
-    num_features: int,
-    ngram: int,
-    tokens: str,
-    lexicon: str,
-    transitions: str,
-    blank: str,
-    allow_repeats: bool,
-    prepend_wordsep: bool = False,
-    use_words: bool = False,
-    data_dir: Optional[Union[str, Path]] = None,
-    reduction: str = "mean",
-) -> Tuple[Transducer, int]:
-    if data_dir is None:
-        data_dir = (
-            Path(__file__).resolve().parents[4] / "data" / "raw" / "iam" / "iamdb"
-        )
-        logger.debug(f"Using data dir: {data_dir}")
-        if not data_dir.exists():
-            raise RuntimeError(f"Could not locate iamdb directory at {data_dir}")
-    else:
-        data_dir = Path(data_dir)
-    processed_path = (
-        Path(__file__).resolve().parents[4] / "data" / "processed" / "iam_lines"
-    )
-    tokens_path = processed_path / tokens
-    lexicon_path = processed_path / lexicon
-
-    if transitions is not None:
-        transitions = gtn.load(str(processed_path / transitions))
-
-    preprocessor = Preprocessor(
-        data_dir, num_features, tokens_path, lexicon_path, use_words, prepend_wordsep,
-    )
-
-    num_tokens = preprocessor.num_tokens
-
-    criterion = Transducer(
-        preprocessor.tokens,
-        preprocessor.graphemes_to_index,
-        ngram=ngram,
-        transitions=transitions,
-        blank=blank,
-        allow_repeats=allow_repeats,
-        reduction=reduction,
-    )
-
-    return criterion, num_tokens + int(blank != "none")
diff --git a/text_recognizer/networks/transformer/__init__.py b/text_recognizer/networks/transformer/__init__.py
index a3f3011..d9e63ef 100644
--- a/text_recognizer/networks/transformer/__init__.py
+++ b/text_recognizer/networks/transformer/__init__.py
@@ -1 +1,3 @@
 """Transformer modules."""
+from .nystromer.nystromer import Nystromer
+from .vit import ViT
diff --git a/text_recognizer/networks/transformer/layers.py b/text_recognizer/networks/transformer/layers.py
index b2c703f..a44a525 100644
--- a/text_recognizer/networks/transformer/layers.py
+++ b/text_recognizer/networks/transformer/layers.py
@@ -1,8 +1,6 @@
 """Generates the attention layer architecture."""
 from functools import partial
-from typing import Any, Dict, Optional, Type
-
-from click.types import Tuple
+from typing import Any, Dict, Optional, Tuple, Type
 
 from torch import nn, Tensor
 
@@ -30,6 +28,7 @@ class AttentionLayers(nn.Module):
         pre_norm: bool = True,
     ) -> None:
         super().__init__()
+        self.dim = dim
         attn_fn = partial(attn_fn, dim=dim, num_heads=num_heads, **attn_kwargs)
         norm_fn = partial(norm_fn, dim)
         ff_fn = partial(ff_fn, dim=dim, **ff_kwargs)
diff --git a/text_recognizer/networks/transformer/positional_encodings/absolute_embedding.py b/text_recognizer/networks/transformer/positional_encodings/absolute_embedding.py
index 9466f6e..7140537 100644
--- a/text_recognizer/networks/transformer/positional_encodings/absolute_embedding.py
+++ b/text_recognizer/networks/transformer/positional_encodings/absolute_embedding.py
@@ -1,4 +1,5 @@
 """Absolute positional embedding."""
+import torch
 from torch import nn, Tensor
 
 
diff --git a/text_recognizer/networks/transformer/transformer.py b/text_recognizer/networks/transformer/transformer.py
index 60ab1ce..31088b4 100644
--- a/text_recognizer/networks/transformer/transformer.py
+++ b/text_recognizer/networks/transformer/transformer.py
@@ -19,7 +19,9 @@ class Transformer(nn.Module):
         emb_dropout: float = 0.0,
         use_pos_emb: bool = True,
     ) -> None:
+        super().__init__()
         dim = attn_layers.dim
+        self.attn_layers = attn_layers
         emb_dim = emb_dim if emb_dim is not None else dim
         self.max_seq_len = max_seq_len
 
@@ -32,7 +34,6 @@ class Transformer(nn.Module):
         )
 
         self.project_emb = nn.Linear(emb_dim, dim) if emb_dim != dim else nn.Identity()
-        self.attn_layers = attn_layers
         self.norm = nn.LayerNorm(dim)
 
         self._init_weights()
@@ -45,12 +46,12 @@ class Transformer(nn.Module):
     def forward(
         self,
         x: Tensor,
-        mask: Optional[Tensor],
+        mask: Optional[Tensor] = None,
         return_embeddings: bool = False,
         **kwargs: Any
     ) -> Tensor:
         b, n, device = *x.shape, x.device
-        x += self.token_emb(x)
+        x = self.token_emb(x)
         if self.pos_emb is not None:
             x += self.pos_emb(x)
         x = self.emb_dropout(x)
diff --git a/text_recognizer/networks/util.py b/text_recognizer/networks/util.py
index 9c6b151..05b10a8 100644
--- a/text_recognizer/networks/util.py
+++ b/text_recognizer/networks/util.py
@@ -22,42 +22,3 @@ def activation_function(activation: str) -> Type[nn.Module]:
         ]
     )
     return activation_fns[activation.lower()]
-
-
-# def configure_backbone(backbone: Union[OmegaConf, NamedTuple]) -> Type[nn.Module]:
-#     """Loads a backbone network."""
-#     network_module = importlib.import_module("text_recognizer.networks")
-#     backbone_class = getattr(network_module, backbone.type)
-#
-#     if "pretrained" in backbone.args:
-#         logger.info("Loading pretrained backbone.")
-#         checkpoint_file = Path(__file__).resolve().parents[2] / backbone.args.pop(
-#             "pretrained"
-#         )
-#
-#         # Loading state directory.
-#         state_dict = torch.load(checkpoint_file)
-#         network_args = state_dict["network_args"]
-#         weights = state_dict["model_state"]
-#
-#         freeze = False
-#         if "freeze" in backbone.args and backbone.args["freeze"] is True:
-#             backbone.args.pop("freeze")
-#             freeze = True
-#
-#         # Initializes the network with trained weights.
-#         backbone_ = backbone_(**backbone.args)
-#         backbone_.load_state_dict(weights)
-#         if freeze:
-#             for params in backbone_.parameters():
-#                 params.requires_grad = False
-#     else:
-#         backbone_ = getattr(network_module, backbone.type)
-#         backbone_ = backbone_(**backbone.args)
-#
-#     if "remove_layers" in backbone_args and backbone_args["remove_layers"] is not None:
-#         backbone = nn.Sequential(
-#             *list(backbone.children())[:][: -backbone_args["remove_layers"]]
-#         )
-#
-#     return backbone