mirror of
https://github.com/blakeblackshear/frigate.git
synced 2026-05-01 19:17:41 +03:00
Create utility to train mobilenet classification models
This commit is contained in:
parent
2bd6fa53fe
commit
34a85b238b
@ -11,6 +11,8 @@ joserfc == 1.0.*
|
||||
pathvalidate == 3.2.*
|
||||
markupsafe == 3.0.*
|
||||
python-multipart == 0.0.12
|
||||
# Classification Model
|
||||
tensorflow == 2.19.*
|
||||
# General
|
||||
mypy == 1.6.1
|
||||
onvif-zeep-async == 3.1.*
|
||||
|
||||
98
frigate/util/classification.py
Normal file
98
frigate/util/classification.py
Normal file
@ -0,0 +1,98 @@
|
||||
"""Util for classification models."""
|
||||
|
||||
import os
|
||||
|
||||
import cv2
|
||||
import numpy as np
|
||||
import tensorflow as tf
|
||||
from tensorflow.keras import layers, models, optimizers
|
||||
from tensorflow.keras.applications import MobileNetV2
|
||||
from tensorflow.keras.preprocessing.image import ImageDataGenerator
|
||||
|
||||
BATCH_SIZE = 16
|
||||
EPOCHS = 50
|
||||
LEARNING_RATE = 0.001
|
||||
|
||||
|
||||
@staticmethod
|
||||
def generate_representative_dataset(train_dir: str):
|
||||
image_paths = []
|
||||
for root, dirs, files in os.walk("train"):
|
||||
for file in files:
|
||||
if file.lower().endswith((".jpg", ".jpeg", ".png")):
|
||||
image_paths.append(os.path.join(root, file))
|
||||
|
||||
for path in image_paths[:300]:
|
||||
img = cv2.imread(path)
|
||||
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
||||
img = cv2.resize(img, (224, 224))
|
||||
img_array = np.array(img, dtype=np.float32) / 255.0
|
||||
img_array = img_array[None, ...]
|
||||
yield [img_array]
|
||||
|
||||
|
||||
@staticmethod
|
||||
def train_classification_model(train_dir: str) -> bool:
|
||||
"""Train a classification model."""
|
||||
num_classes = len(
|
||||
[d for d in os.listdir(train_dir) if os.path.isdir(os.path.join(train_dir, d))]
|
||||
)
|
||||
|
||||
# Start with imagenet base model with 35% of channels in each layer
|
||||
base_model = MobileNetV2(
|
||||
input_shape=(224, 224, 3),
|
||||
include_top=False,
|
||||
weights="imagenet",
|
||||
alpha=0.35,
|
||||
)
|
||||
base_model.trainable = False # Freeze pre-trained layers
|
||||
|
||||
model = models.Sequential(
|
||||
[
|
||||
base_model,
|
||||
layers.GlobalAveragePooling2D(),
|
||||
layers.Dense(128, activation="relu"),
|
||||
layers.Dropout(0.3),
|
||||
layers.Dense(num_classes, activation="softmax"),
|
||||
]
|
||||
)
|
||||
|
||||
model.compile(
|
||||
optimizer=optimizers.Adam(learning_rate=LEARNING_RATE),
|
||||
loss="categorical_crossentropy",
|
||||
metrics=["accuracy"],
|
||||
)
|
||||
|
||||
# create training set
|
||||
datagen = ImageDataGenerator(rescale=1.0 / 255, validation_split=0.2)
|
||||
train_gen = datagen.flow_from_directory(
|
||||
"train",
|
||||
target_size=(224, 224),
|
||||
batch_size=BATCH_SIZE,
|
||||
class_mode="categorical",
|
||||
subset="training",
|
||||
)
|
||||
|
||||
# write labelmap
|
||||
class_indices = train_gen.class_indices
|
||||
index_to_class = {v: k for k, v in class_indices.items()}
|
||||
sorted_classes = [index_to_class[i] for i in range(len(index_to_class))]
|
||||
with open(os.path.join(train_dir, "labelmap.txt"), "w") as f:
|
||||
for class_name in sorted_classes:
|
||||
f.write(f"{class_name}\n")
|
||||
|
||||
# train the model
|
||||
model.fit(train_gen, epochs=EPOCHS)
|
||||
|
||||
# convert model to tflite
|
||||
converter = tf.lite.TFLiteConverter.from_keras_model(model)
|
||||
converter.optimizations = [tf.lite.Optimize.DEFAULT]
|
||||
converter.representative_dataset = generate_representative_dataset
|
||||
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
|
||||
converter.inference_input_type = tf.uint8
|
||||
converter.inference_output_type = tf.uint8
|
||||
tflite_model = converter.convert()
|
||||
|
||||
# write model
|
||||
with open(os.path.join(train_dir, "model.tflite"), "wb") as f:
|
||||
f.write(tflite_model)
|
||||
Loading…
Reference in New Issue
Block a user