{ "cells": [ { "cell_type": "code", "execution_count": 61, "id": "7c02ae76-b540-4b16-9492-e9210b3b9249", "metadata": {}, "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", "import random\n", "\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "\n", "import numpy as np\n", "from omegaconf import OmegaConf\n", "\n", "%load_ext autoreload\n", "%autoreload 2\n", "\n", "from importlib.util import find_spec\n", "if find_spec(\"text_recognizer\") is None:\n", " import sys\n", " sys.path.append('..')" ] }, { "cell_type": "code", "execution_count": 62, "id": "ccdb6dde-47e5-429a-88f2-0764fb7e259a", "metadata": {}, "outputs": [], "source": [ "from hydra import compose, initialize\n", "from omegaconf import OmegaConf\n", "from hydra.utils import instantiate" ] }, { "cell_type": "code", "execution_count": 63, "id": "3cf50475-39f2-4642-a7d1-5bcbc0a036f7", "metadata": {}, "outputs": [], "source": [ "path = \"../training/conf/network/convnext.yaml\"" ] }, { "cell_type": "code", "execution_count": 64, "id": "e52ecb01-c975-4e55-925d-1182c7aea473", "metadata": {}, "outputs": [], "source": [ "with open(path, \"rb\") as f:\n", " cfg = OmegaConf.load(f)" ] }, { "cell_type": "code", "execution_count": 65, "id": "f939aa37-7b1d-45cc-885c-323c4540bda1", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'_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]]}" ] }, "execution_count": 65, "metadata": {}, "output_type": "execute_result" } ], "source": [ "cfg" ] }, { "cell_type": "code", "execution_count": 66, "id": "aaeab329-aeb0-4a1b-aa35-5a2aab81b1d0", "metadata": {}, "outputs": [], "source": [ "net = instantiate(cfg)" ] }, { "cell_type": "code", "execution_count": 67, "id": "618b997c-e6a6-4487-b70c-9d260cb556d3", "metadata": {}, "outputs": [], "source": [ "from torchinfo import summary" ] }, { "cell_type": "code", "execution_count": 68, "id": "25759b7b-8deb-4163-b75d-a1357c9fe88f", "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/plain": [ "====================================================================================================\n", "Layer (type:depth-idx) Output Shape Param #\n", "====================================================================================================\n", "ConvNext [2, 128, 7, 128] --\n", "├─Conv2d: 1-1 [2, 16, 56, 1024] 800\n", "├─ModuleList: 1-2 -- --\n", "│ └─ModuleList: 2-1 -- --\n", "│ │ └─ConvNextBlock: 3-1 [2, 16, 56, 1024] --\n", "│ │ │ └─Conv2d: 4-1 [2, 16, 56, 1024] 800\n", "│ │ │ └─Sequential: 4-2 [2, 16, 56, 1024] 9,280\n", "│ │ │ └─Identity: 4-3 [2, 16, 56, 1024] --\n", "│ │ └─ModuleList: 3-2 -- --\n", "│ │ │ └─ConvNextBlock: 4-4 [2, 16, 56, 1024] 10,080\n", "│ │ │ └─ConvNextBlock: 4-5 [2, 16, 56, 1024] 10,080\n", "│ │ │ └─ConvNextBlock: 4-6 [2, 16, 56, 1024] 10,080\n", "│ │ └─Downsample: 3-3 [2, 32, 28, 512] --\n", "│ │ │ └─Sequential: 4-7 [2, 32, 28, 512] 2,080\n", "│ └─ModuleList: 2-2 -- --\n", "│ │ └─ConvNextBlock: 3-4 [2, 32, 28, 512] --\n", "│ │ │ └─Conv2d: 4-8 [2, 32, 28, 512] 1,600\n", "│ │ │ └─Sequential: 4-9 [2, 32, 28, 512] 36,992\n", "│ │ │ └─Identity: 4-10 [2, 32, 28, 512] --\n", "│ │ └─ModuleList: 3-5 -- --\n", "│ │ │ └─ConvNextBlock: 4-11 [2, 32, 28, 512] 38,592\n", "│ │ │ └─ConvNextBlock: 4-12 [2, 32, 28, 512] 38,592\n", "│ │ │ └─ConvNextBlock: 4-13 [2, 32, 28, 512] 38,592\n", "│ │ └─Downsample: 3-6 [2, 64, 14, 256] --\n", "│ │ │ └─Sequential: 4-14 [2, 64, 14, 256] 8,256\n", "│ └─ModuleList: 2-3 -- --\n", "│ │ └─ConvNextBlock: 3-7 [2, 64, 14, 256] --\n", "│ │ │ └─Conv2d: 4-15 [2, 64, 14, 256] 3,200\n", "│ │ │ └─Sequential: 4-16 [2, 64, 14, 256] 147,712\n", "│ │ │ └─Identity: 4-17 [2, 64, 14, 256] --\n", "│ │ └─ModuleList: 3-8 -- --\n", "│ │ │ └─ConvNextBlock: 4-18 [2, 64, 14, 256] 150,912\n", "│ │ │ └─ConvNextBlock: 4-19 [2, 64, 14, 256] 150,912\n", "│ │ │ └─ConvNextBlock: 4-20 [2, 64, 14, 256] 150,912\n", "│ │ │ └─ConvNextBlock: 4-21 [2, 64, 14, 256] 150,912\n", "│ │ │ └─ConvNextBlock: 4-22 [2, 64, 14, 256] 150,912\n", "│ │ │ └─ConvNextBlock: 4-23 [2, 64, 14, 256] 150,912\n", "│ │ └─Downsample: 3-9 [2, 128, 7, 128] --\n", "│ │ │ └─Sequential: 4-24 [2, 128, 7, 128] 32,896\n", "├─LayerNorm: 1-3 [2, 128, 7, 128] 128\n", "====================================================================================================\n", "Total params: 1,295,232\n", "Trainable params: 1,295,232\n", "Non-trainable params: 0\n", "Total mult-adds (G): 16.88\n", "====================================================================================================\n", "Input size (MB): 0.46\n", "Forward/backward pass size (MB): 598.21\n", "Params size (MB): 5.18\n", "Estimated Total Size (MB): 603.85\n", "====================================================================================================" ] }, "execution_count": 68, "metadata": {}, "output_type": "execute_result" } ], "source": [ "summary(net, (2, 1, 56, 1024), device=\"cpu\", depth=4)" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.4" } }, "nbformat": 4, "nbformat_minor": 5 }