{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "c8Cx-rUMVX25"
},
"source": [
"##### Copyright 2020 The TensorFlow Authors."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"cellView": "form",
"execution": {
"iopub.execute_input": "2022-12-15T00:58:12.697017Z",
"iopub.status.busy": "2022-12-15T00:58:12.696370Z",
"iopub.status.idle": "2022-12-15T00:58:12.700639Z",
"shell.execute_reply": "2022-12-15T00:58:12.700038Z"
},
"id": "I9sUhVL_VZNO"
},
"outputs": [],
"source": [
"#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6Y8E0lw5eYWm"
},
"source": [
"# int16 활성화를 사용한 훈련 후 정수 양자화"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "CGuqeuPSVNo-"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BTC1rDAuei_1"
},
"source": [
"## 개요\n",
"\n",
"[TensorFlow Lite](https://www.tensorflow.org/lite/)는 이제 TensorFlow에서 TensorFlow Lite의 플랫 버퍼 형식으로 모델을 변환하는 동안 활성화를 16bit 정수 값으로, 가중치를 8bit 정수 값으로 변환하는 작업을 지원합니다. 이 모드를 \"16x8 양자화 모드\"라고 합니다. 이 모드는 활성화가 양자화에 민감할 때 양자화된 모델의 정확성을 크게 향상시키면서도 모델 크기를 거의 3 ~ 4배 줄일 수 있습니다. 또한 이 완전 양자화된 모델은 정수 전용 하드웨어 가속기에서 사용할 수 있습니다.\n",
"\n",
"이 훈련 후 양자화 모드의 이점을 얻는 모델의 몇 가지 예는 다음과 같습니다.\n",
"\n",
"- 초고해상도\n",
"- 소음 상쇄 및 빔포밍과 같은 오디오 신호 처리\n",
"- 이미지 노이즈 제거\n",
"- 단일 이미지에서 HDR 재구성\n",
"\n",
"이 튜토리얼에서는 MNIST 모델을 처음부터 학습시키고 TensorFlow에서 정확성을 확인한 다음, 이 모드를 사용하여 모델을 Tensorflow Lite 플랫 버퍼로 변환합니다. 마지막으로, 변환된 모델의 정확성을 확인하고 원래 float32 모델과 비교합니다. 이 예제는 이 모드의 사용법을 보여주는 것이며 TensorFlow Lite에서 사용 가능한 다른 양자화 기술과 비교한 이점을 보여주지는 않습니다."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2XsEP17Zelz9"
},
"source": [
"## MNIST 모델 빌드하기"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "dDqqUIZjZjac"
},
"source": [
"### 설정"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:12.704504Z",
"iopub.status.busy": "2022-12-15T00:58:12.703980Z",
"iopub.status.idle": "2022-12-15T00:58:14.716983Z",
"shell.execute_reply": "2022-12-15T00:58:14.716206Z"
},
"id": "gyqAw1M9lyab"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-12-15 00:58:13.702544: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer.so.7'; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory\n",
"2022-12-15 00:58:13.702639: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libnvinfer_plugin.so.7'; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory\n",
"2022-12-15 00:58:13.702648: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.\n"
]
}
],
"source": [
"import logging\n",
"logging.getLogger(\"tensorflow\").setLevel(logging.DEBUG)\n",
"\n",
"import tensorflow as tf\n",
"from tensorflow import keras\n",
"import numpy as np\n",
"import pathlib"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "srTSFKjn1tMp"
},
"source": [
"16x8 양자화 모드를 사용할 수 있는지 확인합니다. "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:14.721411Z",
"iopub.status.busy": "2022-12-15T00:58:14.720732Z",
"iopub.status.idle": "2022-12-15T00:58:14.727699Z",
"shell.execute_reply": "2022-12-15T00:58:14.727077Z"
},
"id": "c6nb7OPlXs_3"
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tf.lite.OpsSet.EXPERIMENTAL_TFLITE_BUILTINS_ACTIVATIONS_INT16_WEIGHTS_INT8"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "eQ6Q0qqKZogR"
},
"source": [
"### 모델 훈련 및 내보내기"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:14.731335Z",
"iopub.status.busy": "2022-12-15T00:58:14.730840Z",
"iopub.status.idle": "2022-12-15T00:58:27.077800Z",
"shell.execute_reply": "2022-12-15T00:58:27.076788Z"
},
"id": "hWSAjQWagIHl"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\r",
" 1/1875 [..............................] - ETA: 1:17:48 - loss: 2.3468 - accuracy: 0.0312"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 20/1875 [..............................] - ETA: 5s - loss: 2.1178 - accuracy: 0.3266 "
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 40/1875 [..............................] - ETA: 4s - loss: 1.8337 - accuracy: 0.5102"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 60/1875 [..............................] - ETA: 4s - loss: 1.5593 - accuracy: 0.5974"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 81/1875 [>.............................] - ETA: 4s - loss: 1.3492 - accuracy: 0.6512"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 103/1875 [>.............................] - ETA: 4s - loss: 1.1769 - accuracy: 0.6936"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 125/1875 [=>............................] - ETA: 4s - loss: 1.0659 - accuracy: 0.7185"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 146/1875 [=>............................] - ETA: 4s - loss: 0.9814 - accuracy: 0.7369"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 168/1875 [=>............................] - ETA: 4s - loss: 0.9075 - accuracy: 0.7545"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 189/1875 [==>...........................] - ETA: 4s - loss: 0.8596 - accuracy: 0.7657"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 210/1875 [==>...........................] - ETA: 4s - loss: 0.8099 - accuracy: 0.7781"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 232/1875 [==>...........................] - ETA: 4s - loss: 0.7668 - accuracy: 0.7893"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 254/1875 [===>..........................] - ETA: 3s - loss: 0.7369 - accuracy: 0.7970"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 275/1875 [===>..........................] - ETA: 3s - loss: 0.7085 - accuracy: 0.8040"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 297/1875 [===>..........................] - ETA: 3s - loss: 0.6852 - accuracy: 0.8110"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 319/1875 [====>.........................] - ETA: 3s - loss: 0.6611 - accuracy: 0.8171"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 340/1875 [====>.........................] - ETA: 3s - loss: 0.6410 - accuracy: 0.8214"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 361/1875 [====>.........................] - ETA: 3s - loss: 0.6246 - accuracy: 0.8255"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 383/1875 [=====>........................] - ETA: 3s - loss: 0.6105 - accuracy: 0.8290"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 404/1875 [=====>........................] - ETA: 3s - loss: 0.5967 - accuracy: 0.8328"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 425/1875 [=====>........................] - ETA: 3s - loss: 0.5834 - accuracy: 0.8366"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 445/1875 [======>.......................] - ETA: 3s - loss: 0.5705 - accuracy: 0.8400"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 466/1875 [======>.......................] - ETA: 3s - loss: 0.5583 - accuracy: 0.8435"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 488/1875 [======>.......................] - ETA: 3s - loss: 0.5500 - accuracy: 0.8448"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 510/1875 [=======>......................] - ETA: 3s - loss: 0.5402 - accuracy: 0.8467"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 532/1875 [=======>......................] - ETA: 3s - loss: 0.5320 - accuracy: 0.8490"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 553/1875 [=======>......................] - ETA: 3s - loss: 0.5220 - accuracy: 0.8514"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 575/1875 [========>.....................] - ETA: 3s - loss: 0.5154 - accuracy: 0.8535"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 596/1875 [========>.....................] - ETA: 3s - loss: 0.5083 - accuracy: 0.8560"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 617/1875 [========>.....................] - ETA: 3s - loss: 0.5014 - accuracy: 0.8576"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 637/1875 [=========>....................] - ETA: 2s - loss: 0.4965 - accuracy: 0.8591"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 658/1875 [=========>....................] - ETA: 2s - loss: 0.4893 - accuracy: 0.8610"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 679/1875 [=========>....................] - ETA: 2s - loss: 0.4831 - accuracy: 0.8624"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 699/1875 [==========>...................] - ETA: 2s - loss: 0.4773 - accuracy: 0.8639"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 720/1875 [==========>...................] - ETA: 2s - loss: 0.4714 - accuracy: 0.8653"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 742/1875 [==========>...................] - ETA: 2s - loss: 0.4656 - accuracy: 0.8670"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 763/1875 [===========>..................] - ETA: 2s - loss: 0.4610 - accuracy: 0.8682"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 784/1875 [===========>..................] - ETA: 2s - loss: 0.4571 - accuracy: 0.8695"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 805/1875 [===========>..................] - ETA: 2s - loss: 0.4515 - accuracy: 0.8710"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 825/1875 [============>.................] - ETA: 2s - loss: 0.4468 - accuracy: 0.8722"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 845/1875 [============>.................] - ETA: 2s - loss: 0.4426 - accuracy: 0.8735"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 866/1875 [============>.................] - ETA: 2s - loss: 0.4383 - accuracy: 0.8747"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 887/1875 [=============>................] - ETA: 2s - loss: 0.4342 - accuracy: 0.8760"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 908/1875 [=============>................] - ETA: 2s - loss: 0.4305 - accuracy: 0.8771"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 929/1875 [=============>................] - ETA: 2s - loss: 0.4259 - accuracy: 0.8783"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 951/1875 [==============>...............] - ETA: 2s - loss: 0.4214 - accuracy: 0.8795"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 972/1875 [==============>...............] - ETA: 2s - loss: 0.4174 - accuracy: 0.8807"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
" 992/1875 [==============>...............] - ETA: 2s - loss: 0.4132 - accuracy: 0.8819"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1012/1875 [===============>..............] - ETA: 2s - loss: 0.4104 - accuracy: 0.8827"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1033/1875 [===============>..............] - ETA: 2s - loss: 0.4078 - accuracy: 0.8833"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1054/1875 [===============>..............] - ETA: 1s - loss: 0.4043 - accuracy: 0.8845"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1075/1875 [================>.............] - ETA: 1s - loss: 0.4014 - accuracy: 0.8853"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1096/1875 [================>.............] - ETA: 1s - loss: 0.3984 - accuracy: 0.8864"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1117/1875 [================>.............] - ETA: 1s - loss: 0.3959 - accuracy: 0.8870"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1138/1875 [=================>............] - ETA: 1s - loss: 0.3923 - accuracy: 0.8879"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1159/1875 [=================>............] - ETA: 1s - loss: 0.3903 - accuracy: 0.8883"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1180/1875 [=================>............] - ETA: 1s - loss: 0.3877 - accuracy: 0.8892"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1201/1875 [==================>...........] - ETA: 1s - loss: 0.3854 - accuracy: 0.8898"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1222/1875 [==================>...........] - ETA: 1s - loss: 0.3828 - accuracy: 0.8905"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1242/1875 [==================>...........] - ETA: 1s - loss: 0.3794 - accuracy: 0.8915"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1263/1875 [===================>..........] - ETA: 1s - loss: 0.3764 - accuracy: 0.8924"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1284/1875 [===================>..........] - ETA: 1s - loss: 0.3740 - accuracy: 0.8930"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1306/1875 [===================>..........] - ETA: 1s - loss: 0.3709 - accuracy: 0.8938"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1327/1875 [====================>.........] - ETA: 1s - loss: 0.3692 - accuracy: 0.8945"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1349/1875 [====================>.........] - ETA: 1s - loss: 0.3673 - accuracy: 0.8950"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1371/1875 [====================>.........] - ETA: 1s - loss: 0.3644 - accuracy: 0.8959"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1393/1875 [=====================>........] - ETA: 1s - loss: 0.3617 - accuracy: 0.8967"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1414/1875 [=====================>........] - ETA: 1s - loss: 0.3595 - accuracy: 0.8972"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1435/1875 [=====================>........] - ETA: 1s - loss: 0.3568 - accuracy: 0.8979"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1456/1875 [======================>.......] - ETA: 1s - loss: 0.3542 - accuracy: 0.8988"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1477/1875 [======================>.......] - ETA: 0s - loss: 0.3523 - accuracy: 0.8993"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1499/1875 [======================>.......] - ETA: 0s - loss: 0.3501 - accuracy: 0.8999"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1521/1875 [=======================>......] - ETA: 0s - loss: 0.3475 - accuracy: 0.9006"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1542/1875 [=======================>......] - ETA: 0s - loss: 0.3453 - accuracy: 0.9012"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1563/1875 [========================>.....] - ETA: 0s - loss: 0.3431 - accuracy: 0.9019"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1584/1875 [========================>.....] - ETA: 0s - loss: 0.3404 - accuracy: 0.9028"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1605/1875 [========================>.....] - ETA: 0s - loss: 0.3381 - accuracy: 0.9035"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1627/1875 [=========================>....] - ETA: 0s - loss: 0.3362 - accuracy: 0.9040"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1648/1875 [=========================>....] - ETA: 0s - loss: 0.3345 - accuracy: 0.9046"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1669/1875 [=========================>....] - ETA: 0s - loss: 0.3330 - accuracy: 0.9051"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1691/1875 [==========================>...] - ETA: 0s - loss: 0.3312 - accuracy: 0.9055"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1712/1875 [==========================>...] - ETA: 0s - loss: 0.3288 - accuracy: 0.9062"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1734/1875 [==========================>...] - ETA: 0s - loss: 0.3272 - accuracy: 0.9066"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1756/1875 [===========================>..] - ETA: 0s - loss: 0.3252 - accuracy: 0.9071"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1778/1875 [===========================>..] - ETA: 0s - loss: 0.3235 - accuracy: 0.9075"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1799/1875 [===========================>..] - ETA: 0s - loss: 0.3219 - accuracy: 0.9080"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1820/1875 [============================>.] - ETA: 0s - loss: 0.3205 - accuracy: 0.9083"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1841/1875 [============================>.] - ETA: 0s - loss: 0.3192 - accuracy: 0.9088"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1862/1875 [============================>.] - ETA: 0s - loss: 0.3174 - accuracy: 0.9092"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r",
"1875/1875 [==============================] - 8s 3ms/step - loss: 0.3165 - accuracy: 0.9095 - val_loss: 0.1584 - val_accuracy: 0.9553\n"
]
},
{
"data": {
"text/plain": [
""
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Load MNIST dataset\n",
"mnist = keras.datasets.mnist\n",
"(train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
"\n",
"# Normalize the input image so that each pixel value is between 0 to 1.\n",
"train_images = train_images / 255.0\n",
"test_images = test_images / 255.0\n",
"\n",
"# Define the model architecture\n",
"model = keras.Sequential([\n",
" keras.layers.InputLayer(input_shape=(28, 28)),\n",
" keras.layers.Reshape(target_shape=(28, 28, 1)),\n",
" keras.layers.Conv2D(filters=12, kernel_size=(3, 3), activation=tf.nn.relu),\n",
" keras.layers.MaxPooling2D(pool_size=(2, 2)),\n",
" keras.layers.Flatten(),\n",
" keras.layers.Dense(10)\n",
"])\n",
"\n",
"# Train the digit classification model\n",
"model.compile(optimizer='adam',\n",
" loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
" metrics=['accuracy'])\n",
"model.fit(\n",
" train_images,\n",
" train_labels,\n",
" epochs=1,\n",
" validation_data=(test_images, test_labels)\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "5NMaNZQCkW9X"
},
"source": [
"예를 들어, 단일 epoch에 대해서만 모델을 훈련시켰으므로 ~96% 정확성으로만 훈련합니다."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "xl8_fzVAZwOh"
},
"source": [
"### TensorFlow Lite 모델로 변환하기\n",
"\n",
"TensorFlow Lite [Converter](https://www.tensorflow.org/lite/models/convert)를 사용하여 이제 훈련된 모델을 TensorFlow Lite 모델로 변환할 수 있습니다.\n",
"\n",
"이제 `TFliteConverter`를 사용하여 모델을 기본 float32 형식으로 변환합니다."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:27.081780Z",
"iopub.status.busy": "2022-12-15T00:58:27.081093Z",
"iopub.status.idle": "2022-12-15T00:58:28.143103Z",
"shell.execute_reply": "2022-12-15T00:58:28.142180Z"
},
"id": "_i8B2nDZmAgQ"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op while saving (showing 1 of 1). These functions will not be directly callable after loading.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"INFO:tensorflow:Assets written to: /tmpfs/tmp/tmplvt0uukx/assets\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:tensorflow:Assets written to: /tmpfs/tmp/tmplvt0uukx/assets\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-12-15 00:58:28.021632: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
"2022-12-15 00:58:28.021671: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n"
]
}
],
"source": [
"converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
"tflite_model = converter.convert()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "F2o2ZfF0aiCx"
},
"source": [
"`.tflite` 파일에 작성합니다."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:28.147372Z",
"iopub.status.busy": "2022-12-15T00:58:28.146755Z",
"iopub.status.idle": "2022-12-15T00:58:28.150861Z",
"shell.execute_reply": "2022-12-15T00:58:28.150048Z"
},
"id": "vptWZq2xnclo"
},
"outputs": [],
"source": [
"tflite_models_dir = pathlib.Path(\"/tmp/mnist_tflite_models/\")\n",
"tflite_models_dir.mkdir(exist_ok=True, parents=True)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:28.154503Z",
"iopub.status.busy": "2022-12-15T00:58:28.153916Z",
"iopub.status.idle": "2022-12-15T00:58:28.159497Z",
"shell.execute_reply": "2022-12-15T00:58:28.158771Z"
},
"id": "Ie9pQaQrn5ue"
},
"outputs": [
{
"data": {
"text/plain": [
"84820"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tflite_model_file = tflite_models_dir/\"mnist_model.tflite\"\n",
"tflite_model_file.write_bytes(tflite_model)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "7BONhYtYocQY"
},
"source": [
"대신 모델을 16x8 양자화 모드로 양자화하려면 먼저 기본 최적화를 사용하도록 `optimizations` 플래그를 설정합니다. 그런 다음 16x8 양자화 모드가 대상 사양에서 지원되는 필수 연산임을 지정합니다."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:28.163062Z",
"iopub.status.busy": "2022-12-15T00:58:28.162504Z",
"iopub.status.idle": "2022-12-15T00:58:28.166282Z",
"shell.execute_reply": "2022-12-15T00:58:28.165367Z"
},
"id": "HEZ6ET1AHAS3"
},
"outputs": [],
"source": [
"converter.optimizations = [tf.lite.Optimize.DEFAULT]\n",
"converter.target_spec.supported_ops = [tf.lite.OpsSet.EXPERIMENTAL_TFLITE_BUILTINS_ACTIVATIONS_INT16_WEIGHTS_INT8]"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "zLxQwZq9CpN7"
},
"source": [
"int8 훈련 후 양자화의 경우와 마찬가지로, 변환기 옵션 `inference_input(output)_type`을 tf.int16으로 설정하여 완전히 정수 양자화된 모델을 생성할 수 있습니다."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "yZekFJC5-fOG"
},
"source": [
"보정 데이터를 설정합니다."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:28.170026Z",
"iopub.status.busy": "2022-12-15T00:58:28.169477Z",
"iopub.status.idle": "2022-12-15T00:58:28.659849Z",
"shell.execute_reply": "2022-12-15T00:58:28.659098Z"
},
"id": "Y3a6XFqvHbYM"
},
"outputs": [],
"source": [
"mnist_train, _ = tf.keras.datasets.mnist.load_data()\n",
"images = tf.cast(mnist_train[0], tf.float32) / 255.0\n",
"mnist_ds = tf.data.Dataset.from_tensor_slices((images)).batch(1)\n",
"def representative_data_gen():\n",
" for input_value in mnist_ds.take(100):\n",
" # Model has only one input so each data point has one element.\n",
" yield [input_value]\n",
"converter.representative_dataset = representative_data_gen"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "xW84iMYjHd9t"
},
"source": [
"마지막으로, 평소와 같이 모델을 변환합니다. 기본적으로 변환된 모델은 호출 편의를 위해 여전히 float 입력 및 출력을 사용합니다."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:28.664273Z",
"iopub.status.busy": "2022-12-15T00:58:28.663975Z",
"iopub.status.idle": "2022-12-15T00:58:30.163359Z",
"shell.execute_reply": "2022-12-15T00:58:30.162511Z"
},
"id": "yuNfl3CoHNK3"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op while saving (showing 1 of 1). These functions will not be directly callable after loading.\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"INFO:tensorflow:Assets written to: /tmpfs/tmp/tmpiqja0kzc/assets\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO:tensorflow:Assets written to: /tmpfs/tmp/tmpiqja0kzc/assets\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"2022-12-15 00:58:29.502394: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:362] Ignored output_format.\n",
"2022-12-15 00:58:29.502436: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:365] Ignored drop_control_dependency.\n"
]
},
{
"data": {
"text/plain": [
"24960"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tflite_16x8_model = converter.convert()\n",
"tflite_model_16x8_file = tflite_models_dir/\"mnist_model_quant_16x8.tflite\"\n",
"tflite_model_16x8_file.write_bytes(tflite_16x8_model)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "PhMmUTl4sbkz"
},
"source": [
"결과 파일이 약 `1/3` 크기인 것에 주목합니다."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:30.167248Z",
"iopub.status.busy": "2022-12-15T00:58:30.166580Z",
"iopub.status.idle": "2022-12-15T00:58:30.370181Z",
"shell.execute_reply": "2022-12-15T00:58:30.368995Z"
},
"id": "JExfcfLDscu4"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"total 156K\r\n",
"-rw-rw-r-- 1 kbuilder kbuilder 83K Dec 15 00:58 mnist_model.tflite\r\n",
"-rw-rw-r-- 1 kbuilder kbuilder 25K Dec 15 00:58 mnist_model_quant_16x8.tflite\r\n",
"-rw-rw-r-- 1 kbuilder kbuilder 44K Dec 15 00:56 mnist_model_quant_f16.tflite\r\n"
]
}
],
"source": [
"!ls -lh {tflite_models_dir}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "L8lQHMp_asCq"
},
"source": [
"## TensorFlow Lite 모델 실행하기"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "-5l6-ciItvX6"
},
"source": [
"Python TensorFlow Lite 인터프리터를 사용하여 TensorFlow Lite 모델을 실행합니다."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Ap_jE7QRvhPf"
},
"source": [
"### 인터프리터에 모델 로드하기"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:30.374722Z",
"iopub.status.busy": "2022-12-15T00:58:30.374439Z",
"iopub.status.idle": "2022-12-15T00:58:30.381032Z",
"shell.execute_reply": "2022-12-15T00:58:30.380405Z"
},
"id": "Jn16Rc23zTss"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"INFO: Created TensorFlow Lite XNNPACK delegate for CPU.\n"
]
}
],
"source": [
"interpreter = tf.lite.Interpreter(model_path=str(tflite_model_file))\n",
"interpreter.allocate_tensors()"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:30.384334Z",
"iopub.status.busy": "2022-12-15T00:58:30.383796Z",
"iopub.status.idle": "2022-12-15T00:58:30.387721Z",
"shell.execute_reply": "2022-12-15T00:58:30.387071Z"
},
"id": "J8Pztk1mvNVL"
},
"outputs": [],
"source": [
"interpreter_16x8 = tf.lite.Interpreter(model_path=str(tflite_model_16x8_file))\n",
"interpreter_16x8.allocate_tensors()"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2opUt_JTdyEu"
},
"source": [
"### 하나의 이미지에서 모델 테스트하기"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:30.390905Z",
"iopub.status.busy": "2022-12-15T00:58:30.390678Z",
"iopub.status.idle": "2022-12-15T00:58:30.395298Z",
"shell.execute_reply": "2022-12-15T00:58:30.394748Z"
},
"id": "AKslvo2kwWac"
},
"outputs": [],
"source": [
"test_image = np.expand_dims(test_images[0], axis=0).astype(np.float32)\n",
"\n",
"input_index = interpreter.get_input_details()[0][\"index\"]\n",
"output_index = interpreter.get_output_details()[0][\"index\"]\n",
"\n",
"interpreter.set_tensor(input_index, test_image)\n",
"interpreter.invoke()\n",
"predictions = interpreter.get_tensor(output_index)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:30.398765Z",
"iopub.status.busy": "2022-12-15T00:58:30.398160Z",
"iopub.status.idle": "2022-12-15T00:58:31.095932Z",
"shell.execute_reply": "2022-12-15T00:58:31.095249Z"
},
"id": "XZClM2vo3_bm"
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pylab as plt\n",
"\n",
"plt.imshow(test_images[0])\n",
"template = \"True:{true}, predicted:{predict}\"\n",
"_ = plt.title(template.format(true= str(test_labels[0]),\n",
" predict=str(np.argmax(predictions[0]))))\n",
"plt.grid(False)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:31.099257Z",
"iopub.status.busy": "2022-12-15T00:58:31.099004Z",
"iopub.status.idle": "2022-12-15T00:58:31.103258Z",
"shell.execute_reply": "2022-12-15T00:58:31.102674Z"
},
"id": "3gwhv4lKbYZ4"
},
"outputs": [],
"source": [
"test_image = np.expand_dims(test_images[0], axis=0).astype(np.float32)\n",
"\n",
"input_index = interpreter_16x8.get_input_details()[0][\"index\"]\n",
"output_index = interpreter_16x8.get_output_details()[0][\"index\"]\n",
"\n",
"interpreter_16x8.set_tensor(input_index, test_image)\n",
"interpreter_16x8.invoke()\n",
"predictions = interpreter_16x8.get_tensor(output_index)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:31.106327Z",
"iopub.status.busy": "2022-12-15T00:58:31.105857Z",
"iopub.status.idle": "2022-12-15T00:58:31.258756Z",
"shell.execute_reply": "2022-12-15T00:58:31.258014Z"
},
"id": "CIH7G_MwbY2x"
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.imshow(test_images[0])\n",
"template = \"True:{true}, predicted:{predict}\"\n",
"_ = plt.title(template.format(true= str(test_labels[0]),\n",
" predict=str(np.argmax(predictions[0]))))\n",
"plt.grid(False)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "LwN7uIdCd8Gw"
},
"source": [
"### 모델 평가하기"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:31.261952Z",
"iopub.status.busy": "2022-12-15T00:58:31.261710Z",
"iopub.status.idle": "2022-12-15T00:58:31.267303Z",
"shell.execute_reply": "2022-12-15T00:58:31.266628Z"
},
"id": "05aeAuWjvjPx"
},
"outputs": [],
"source": [
"# A helper function to evaluate the TF Lite model using \"test\" dataset.\n",
"def evaluate_model(interpreter):\n",
" input_index = interpreter.get_input_details()[0][\"index\"]\n",
" output_index = interpreter.get_output_details()[0][\"index\"]\n",
"\n",
" # Run predictions on every image in the \"test\" dataset.\n",
" prediction_digits = []\n",
" for test_image in test_images:\n",
" # Pre-processing: add batch dimension and convert to float32 to match with\n",
" # the model's input data format.\n",
" test_image = np.expand_dims(test_image, axis=0).astype(np.float32)\n",
" interpreter.set_tensor(input_index, test_image)\n",
"\n",
" # Run inference.\n",
" interpreter.invoke()\n",
"\n",
" # Post-processing: remove batch dimension and find the digit with highest\n",
" # probability.\n",
" output = interpreter.tensor(output_index)\n",
" digit = np.argmax(output()[0])\n",
" prediction_digits.append(digit)\n",
"\n",
" # Compare prediction results with ground truth labels to calculate accuracy.\n",
" accurate_count = 0\n",
" for index in range(len(prediction_digits)):\n",
" if prediction_digits[index] == test_labels[index]:\n",
" accurate_count += 1\n",
" accuracy = accurate_count * 1.0 / len(prediction_digits)\n",
"\n",
" return accuracy"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:31.270267Z",
"iopub.status.busy": "2022-12-15T00:58:31.269768Z",
"iopub.status.idle": "2022-12-15T00:58:32.554366Z",
"shell.execute_reply": "2022-12-15T00:58:32.553535Z"
},
"id": "T5mWkSbMcU5z"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0.9553\n"
]
}
],
"source": [
"print(evaluate_model(interpreter))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Km3cY9ry8ZlG"
},
"source": [
"16x8 양자화된 모델에 대해 평가를 반복합니다."
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"execution": {
"iopub.execute_input": "2022-12-15T00:58:32.557959Z",
"iopub.status.busy": "2022-12-15T00:58:32.557697Z",
"iopub.status.idle": "2022-12-15T00:58:36.220486Z",
"shell.execute_reply": "2022-12-15T00:58:36.219726Z"
},
"id": "-9cnwiPp6EGm"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0.9554\n"
]
}
],
"source": [
"# NOTE: This quantization mode is an experimental post-training mode,\n",
"# it does not have any optimized kernels implementations or\n",
"# specialized machine learning hardware accelerators. Therefore,\n",
"# it could be slower than the float interpreter.\n",
"print(evaluate_model(interpreter_16x8))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "L7lfxkor8pgv"
},
"source": [
"이 예에서는 정확성에 차이가 없지만 3배 축소된 크기로 모델을 16x8로 양자화했습니다.\n"
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [],
"name": "post_training_integer_quant_16x8.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"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": "3.9.16"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
![](\"https://www.tensorflow.org/images/tf_logo_32px.png\"){kind=logo}
TensorFlow.org에서 보기 ![](\"https://www.tensorflow.org/images/colab_logo_32px.png\"){kind=logo}
Google Colab에서 실행![](\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\")
GitHub에서 소스 보기 ![](\"https://www.tensorflow.org/images/download_logo_32px.png\"){kind=logo}
노트북 다운로드