summaryrefslogtreecommitdiff
path: root/notebooks/04-convnext.ipynb
diff options
context:
space:
mode:
Diffstat (limited to 'notebooks/04-convnext.ipynb')
-rw-r--r--notebooks/04-convnext.ipynb169
1 files changed, 83 insertions, 86 deletions
diff --git a/notebooks/04-convnext.ipynb b/notebooks/04-convnext.ipynb
index 1779d13..b8d4e56 100644
--- a/notebooks/04-convnext.ipynb
+++ b/notebooks/04-convnext.ipynb
@@ -2,10 +2,19 @@
"cells": [
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": 14,
"id": "7c02ae76-b540-4b16-9492-e9210b3b9249",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The autoreload extension is already loaded. To reload it, use:\n",
+ " %reload_ext autoreload\n"
+ ]
+ }
+ ],
"source": [
"import os\n",
"os.environ['CUDA_VISIBLE_DEVICE'] = ''\n",
@@ -28,7 +37,7 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 15,
"id": "ccdb6dde-47e5-429a-88f2-0764fb7e259a",
"metadata": {},
"outputs": [],
@@ -40,7 +49,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 16,
"id": "3cf50475-39f2-4642-a7d1-5bcbc0a036f7",
"metadata": {},
"outputs": [],
@@ -50,7 +59,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 37,
"id": "e52ecb01-c975-4e55-925d-1182c7aea473",
"metadata": {},
"outputs": [],
@@ -61,7 +70,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 38,
"id": "f939aa37-7b1d-45cc-885c-323c4540bda1",
"metadata": {},
"outputs": [
@@ -71,7 +80,7 @@
"{'_target_': 'text_recognizer.networks.convnext.ConvNext', 'dim': 16, 'dim_mults': [2, 4, 8], 'depths': [3, 3, 6], 'downsampling_factors': [[2, 2], [2, 2], [2, 2]], 'attn': {'_target_': 'text_recognizer.networks.convnext.TransformerBlock', 'attn': {'_target_': 'text_recognizer.networks.convnext.Attention', 'dim': 128, 'heads': 4, 'dim_head': 64, 'scale': 8}, 'ff': {'_target_': 'text_recognizer.networks.convnext.FeedForward', 'dim': 128, 'mult': 4}}}"
]
},
- "execution_count": 5,
+ "execution_count": 38,
"metadata": {},
"output_type": "execute_result"
}
@@ -82,19 +91,19 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 19,
"id": "a2b420c1",
"metadata": {},
"outputs": [],
"source": [
- "cfg.dim_mults = [2, 4, 8, 8, 8]\n",
- "cfg.depths = [3, 3, 3, 3, 6]\n",
- "cfg.downsampling_factors = [[2, 2], [2, 2], [2, 2], [2, 1], [2, 1]]"
+ "cfg.dim_mults = [2, 4, 8, 8]\n",
+ "cfg.depths = [3, 3, 6, 6]\n",
+ "cfg.downsampling_factors = [[2, 2], [2, 2], [2, 2], [2, 1]]"
]
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 39,
"id": "c9589350",
"metadata": {},
"outputs": [],
@@ -104,7 +113,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 40,
"id": "618b997c-e6a6-4487-b70c-9d260cb556d3",
"metadata": {},
"outputs": [],
@@ -114,7 +123,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 41,
"id": "25759b7b-8deb-4163-b75d-a1357c9fe88f",
"metadata": {
"scrolled": false
@@ -126,90 +135,78 @@
"====================================================================================================\n",
"Layer (type:depth-idx) Output Shape Param #\n",
"====================================================================================================\n",
- "ConvNext [2, 128, 18, 80] 5,969,376\n",
- "├─Conv2d: 1-1 [2, 16, 576, 640] 800\n",
- "├─ModuleList: 1 -- --\n",
- "│ └─ModuleList: 2 -- --\n",
- "│ │ └─ConvNextBlock: 3-1 [2, 16, 576, 640] --\n",
- "│ │ │ └─Conv2d: 4-1 [2, 16, 576, 640] 800\n",
- "│ │ │ └─Sequential: 4-2 [2, 16, 576, 640] 9,280\n",
- "│ │ │ └─Identity: 4-3 [2, 16, 576, 640] --\n",
- "│ │ └─ModuleList: 3 -- --\n",
- "│ │ │ └─ConvNextBlock: 4-4 [2, 16, 576, 640] 10,080\n",
+ "ConvNext [2, 128, 72, 80] 165,408\n",
+ "├─TransformerBlock: 1-5 [2, 128, 72, 80] (recursive)\n",
+ "│ └─Attention: 2-6 [2, 128, 72, 80] (recursive)\n",
+ "│ │ └─LayerNorm: 3-13 [2, 128, 72, 80] (recursive)\n",
+ "├─Conv2d: 1-3 [2, 16, 576, 640] (recursive)\n",
+ "├─TransformerBlock: 1 -- --\n",
+ "│ └─Attention: 2 -- --\n",
+ "│ │ └─Conv2d: 3-15 [2, 128, 72, 80] (recursive)\n",
+ "│ └─FeedForward: 2-7 [2, 128, 72, 80] (recursive)\n",
+ "│ │ └─Residual: 3-16 [2, 128, 72, 80] (recursive)\n",
+ "│ │ │ └─Sequential: 4-26 [2, 128, 72, 80] (recursive)\n",
+ "├─Conv2d: 1-3 [2, 16, 576, 640] (recursive)\n",
+ "├─ModuleList: 1-4 -- --\n",
+ "│ └─ModuleList: 2-3 -- --\n",
+ "│ │ └─ConvNextBlock: 3-4 [2, 16, 576, 640] --\n",
+ "│ │ │ └─Conv2d: 4-2 [2, 16, 576, 640] 800\n",
+ "│ │ │ └─Sequential: 4-3 [2, 16, 576, 640] 9,280\n",
+ "│ │ │ └─Identity: 4-4 [2, 16, 576, 640] --\n",
+ "│ │ └─ModuleList: 3-5 -- --\n",
"│ │ │ └─ConvNextBlock: 4-5 [2, 16, 576, 640] 10,080\n",
"│ │ │ └─ConvNextBlock: 4-6 [2, 16, 576, 640] 10,080\n",
- "│ │ └─Downsample: 3-2 [2, 32, 288, 320] --\n",
- "│ │ │ └─Sequential: 4-7 [2, 32, 288, 320] 2,080\n",
- "│ └─ModuleList: 2 -- --\n",
- "│ │ └─ConvNextBlock: 3-3 [2, 32, 288, 320] --\n",
- "│ │ │ └─Conv2d: 4-8 [2, 32, 288, 320] 1,600\n",
- "│ │ │ └─Sequential: 4-9 [2, 32, 288, 320] 36,992\n",
- "│ │ │ └─Identity: 4-10 [2, 32, 288, 320] --\n",
- "│ │ └─ModuleList: 3 -- --\n",
- "│ │ │ └─ConvNextBlock: 4-11 [2, 32, 288, 320] 38,592\n",
+ "│ │ │ └─ConvNextBlock: 4-7 [2, 16, 576, 640] 10,080\n",
+ "│ │ └─Downsample: 3-6 [2, 32, 288, 320] --\n",
+ "│ │ │ └─Sequential: 4-8 [2, 32, 288, 320] 2,080\n",
+ "│ └─ModuleList: 2-4 -- --\n",
+ "│ │ └─ConvNextBlock: 3-7 [2, 32, 288, 320] --\n",
+ "│ │ │ └─Conv2d: 4-9 [2, 32, 288, 320] 1,600\n",
+ "│ │ │ └─Sequential: 4-10 [2, 32, 288, 320] 36,992\n",
+ "│ │ │ └─Identity: 4-11 [2, 32, 288, 320] --\n",
+ "│ │ └─ModuleList: 3-8 -- --\n",
"│ │ │ └─ConvNextBlock: 4-12 [2, 32, 288, 320] 38,592\n",
"│ │ │ └─ConvNextBlock: 4-13 [2, 32, 288, 320] 38,592\n",
- "│ │ └─Downsample: 3-4 [2, 64, 144, 160] --\n",
- "│ │ │ └─Sequential: 4-14 [2, 64, 144, 160] 8,256\n",
- "│ └─ModuleList: 2 -- --\n",
- "│ │ └─ConvNextBlock: 3-5 [2, 64, 144, 160] --\n",
- "│ │ │ └─Conv2d: 4-15 [2, 64, 144, 160] 3,200\n",
- "│ │ │ └─Sequential: 4-16 [2, 64, 144, 160] 147,712\n",
- "│ │ │ └─Identity: 4-17 [2, 64, 144, 160] --\n",
- "│ │ └─ModuleList: 3 -- --\n",
- "│ │ │ └─ConvNextBlock: 4-18 [2, 64, 144, 160] 150,912\n",
+ "│ │ │ └─ConvNextBlock: 4-14 [2, 32, 288, 320] 38,592\n",
+ "│ │ └─Downsample: 3-9 [2, 64, 144, 160] --\n",
+ "│ │ │ └─Sequential: 4-15 [2, 64, 144, 160] 8,256\n",
+ "│ └─ModuleList: 2-5 -- --\n",
+ "│ │ └─ConvNextBlock: 3-10 [2, 64, 144, 160] --\n",
+ "│ │ │ └─Conv2d: 4-16 [2, 64, 144, 160] 3,200\n",
+ "│ │ │ └─Sequential: 4-17 [2, 64, 144, 160] 147,712\n",
+ "│ │ │ └─Identity: 4-18 [2, 64, 144, 160] --\n",
+ "│ │ └─ModuleList: 3-11 -- --\n",
"│ │ │ └─ConvNextBlock: 4-19 [2, 64, 144, 160] 150,912\n",
"│ │ │ └─ConvNextBlock: 4-20 [2, 64, 144, 160] 150,912\n",
- "│ │ └─Downsample: 3-6 [2, 128, 72, 80] --\n",
- "│ │ │ └─Sequential: 4-21 [2, 128, 72, 80] 32,896\n",
- "│ └─ModuleList: 2 -- --\n",
- "│ │ └─ConvNextBlock: 3-7 [2, 128, 72, 80] --\n",
- "│ │ │ └─Conv2d: 4-22 [2, 128, 72, 80] 6,400\n",
- "│ │ │ └─Sequential: 4-23 [2, 128, 72, 80] 590,336\n",
- "│ │ │ └─Identity: 4-24 [2, 128, 72, 80] --\n",
- "│ │ └─ModuleList: 3 -- --\n",
- "│ │ │ └─ConvNextBlock: 4-25 [2, 128, 72, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-26 [2, 128, 72, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-27 [2, 128, 72, 80] 596,736\n",
- "│ │ └─Downsample: 3-8 [2, 128, 36, 80] --\n",
- "│ │ │ └─Sequential: 4-28 [2, 128, 36, 80] 32,896\n",
- "│ └─ModuleList: 2 -- --\n",
- "│ │ └─ConvNextBlock: 3-9 [2, 128, 36, 80] --\n",
- "│ │ │ └─Conv2d: 4-29 [2, 128, 36, 80] 6,400\n",
- "│ │ │ └─Sequential: 4-30 [2, 128, 36, 80] 590,336\n",
- "│ │ │ └─Identity: 4-31 [2, 128, 36, 80] --\n",
- "│ │ └─ModuleList: 3 -- --\n",
- "│ │ │ └─ConvNextBlock: 4-32 [2, 128, 36, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-33 [2, 128, 36, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-34 [2, 128, 36, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-35 [2, 128, 36, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-36 [2, 128, 36, 80] 596,736\n",
- "│ │ │ └─ConvNextBlock: 4-37 [2, 128, 36, 80] 596,736\n",
- "│ │ └─Downsample: 3-10 [2, 128, 18, 80] --\n",
- "│ │ │ └─Sequential: 4-38 [2, 128, 18, 80] 32,896\n",
- "├─TransformerBlock: 1-2 [2, 128, 18, 80] --\n",
- "│ └─Attention: 2-1 [2, 128, 18, 80] --\n",
- "│ │ └─LayerNorm: 3-11 [2, 128, 18, 80] 128\n",
- "│ │ └─Conv2d: 3-12 [2, 768, 18, 80] 98,304\n",
- "│ │ └─Conv2d: 3-13 [2, 128, 18, 80] 32,768\n",
- "│ └─FeedForward: 2-2 [2, 128, 18, 80] --\n",
- "│ │ └─Residual: 3-14 [2, 128, 18, 80] --\n",
- "│ │ │ └─Sequential: 4-39 [2, 128, 18, 80] 131,712\n",
- "├─LayerNorm: 1-3 [2, 128, 18, 80] 128\n",
+ "│ │ │ └─ConvNextBlock: 4-21 [2, 64, 144, 160] 150,912\n",
+ "│ │ │ └─ConvNextBlock: 4-22 [2, 64, 144, 160] 150,912\n",
+ "│ │ │ └─ConvNextBlock: 4-23 [2, 64, 144, 160] 150,912\n",
+ "│ │ │ └─ConvNextBlock: 4-24 [2, 64, 144, 160] 150,912\n",
+ "│ │ └─Downsample: 3-12 [2, 128, 72, 80] --\n",
+ "│ │ │ └─Sequential: 4-25 [2, 128, 72, 80] 32,896\n",
+ "├─TransformerBlock: 1-5 [2, 128, 72, 80] (recursive)\n",
+ "│ └─Attention: 2-6 [2, 128, 72, 80] (recursive)\n",
+ "│ │ └─LayerNorm: 3-13 [2, 128, 72, 80] (recursive)\n",
+ "│ │ └─Conv2d: 3-14 [2, 768, 72, 80] 98,304\n",
+ "│ │ └─Conv2d: 3-15 [2, 128, 72, 80] (recursive)\n",
+ "│ └─FeedForward: 2-7 [2, 128, 72, 80] (recursive)\n",
+ "│ │ └─Residual: 3-16 [2, 128, 72, 80] (recursive)\n",
+ "│ │ │ └─Sequential: 4-26 [2, 128, 72, 80] (recursive)\n",
+ "├─LayerNorm: 1-6 [2, 128, 72, 80] 128\n",
"====================================================================================================\n",
- "Total params: 7,735,296\n",
- "Trainable params: 7,735,296\n",
+ "Total params: 1,558,144\n",
+ "Trainable params: 1,558,144\n",
"Non-trainable params: 0\n",
- "Total mult-adds (G): 140.23\n",
+ "Total mult-adds (G): 114.00\n",
"====================================================================================================\n",
"Input size (MB): 2.95\n",
- "Forward/backward pass size (MB): 3987.21\n",
- "Params size (MB): 30.94\n",
- "Estimated Total Size (MB): 4021.10\n",
+ "Forward/backward pass size (MB): 3822.06\n",
+ "Params size (MB): 5.57\n",
+ "Estimated Total Size (MB): 3830.58\n",
"===================================================================================================="
]
},
- "execution_count": 13,
+ "execution_count": 41,
"metadata": {},
"output_type": "execute_result"
}
@@ -221,7 +218,7 @@
{
"cell_type": "code",
"execution_count": null,
- "id": "7a7e27f4",
+ "id": "05c1d499",
"metadata": {},
"outputs": [],
"source": []
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2024-09-20 02:37:21 +0000