yolov4 3.2.0

YOLOv4: Optimal Speed and Accuracy of Object Detection

tensorflow-yolov4

YOLOv4 Implemented in Tensorflow 2.

>=v3, :exclamation: Do not try to train. Not yet stable.

Docs: https://wiki.loliot.net/docs/lang/python/libraries/yolov4/python-yolov4-about

Installation

python3 -m pip install yolov4

Changelog

Ref: CHANGELOG

tensorflow-yolov4 (3.2.0) unstable; urgency=medium

• config: add yolov4-tiny-relu-new_coords.cfg
• c_src: layers: add yolo_tpu_layer_new_coords
• c_src, common, tf, tflite, mAP: add prob_thresh
• config: add yolov4-tiny-relu-new_coords-tpu.cfg
• common: base_class: modify text

-- Hyeonki Hong [e-mail] Mon, 22 Feb 2021 01:30:53 +0900

tensorflow-yolov4 (3.1.0) unstable; urgency=medium

• tf: layers: yolo_layer: remove some operations
• common: metalayer: yolo_layer: add new_coords
• tf: layers: yolo_layer: implement _coords_1 for new_coords
• tf: layers: convolutional_layer: add logistic activation
• c_src: py_box: add get_yolo_*_detections
• config: add yolov4-csp.cfg and yolov4x-mish.cfg
• c_src: layers: modify yolo_tpu_layer
• tf, tflite: reflect v3.1.0 changes

-- Hyeonki Hong [e-mail] Sun, 21 Feb 2021 02:37:18 +0900

tensorflow-yolov4 (3.0.1) unstable; urgency=medium

• pypi: add 'pyproject.toml' and update 'setup.*'
• common: parser: update error message
• c_src: modify extended operator synonyms
• c_src: change c-header to cpp-header

-- Hyeonki Hong [e-mail] Thu, 18 Feb 2021 19:02:01 +0900

tensorflow-yolov4 (3.0.0) unstable; urgency=medium

• common: parser: add parse_cfg()
• common: config: implement YOLOConfig
• common: metalayer: implement metalayers
• common: metalayer: add YoloTpuLayer
• tf: layers: implement layers
• tf: layers: redefine BatchNormalization for transfer learning
• tf: layers: add YoloTpuLayer
• tf: dataset: refactor YOLODataset
• tf: dataset: add '_augmentation_cache'
• tf: training: refactor YOLOv4Loss
• tf: training: implement YOLOCallbackAtEachStep
• tf: training: refactor SaveWeightsCallback
• tf: utils: move weights.py from tf
• tf: utils: move mAP part from tf
• tf: utils: move tflite part from tf and fix converting issue
• c_src: py_box: add 'yolo_diou_nms'
• c_src: py_box: add convert_dataset_to_ground_truth
• c_src: layers: implement 'yolo_layer'
• tf, tflite: reflect v3.0.0 changes

-- Hyeonki Hong [e-mail] Mon, 15 Feb 2021 00:15:08 +0900

tensorflow-yolov4 (2.1.0) unstable; urgency=medium

• tflite: add edgetpu_lib argument to load_tflite()

-- Hyeonki Hong [e-mail] Wed, 27 Jan 2021 11:09:02 +0900

tensorflow-yolov4 (2.0.3) unstable; urgency=medium

• common: allow media_path to be video stream id
• tflite: fix problem of not finding load_delegate()

-- Hyeonki Hong [e-mail] Thu, 07 Jan 2021 21:34:35 +0900

tensorflow-yolov4 (2.0.2) ustable; urgency=medium

• tf: set converter.experimental_new_converter to False
• tf: set converter.inference_in/output_type to tf.uint8
• tflite: fix problem of not finding load_delegate()

-- Hyeonki Hong [e-mail] Sun, 20 Dec 2020 13:06:08 +0900

tensorflow-yolov4 (2.0.1) unstable; urgency=medium

• tf: set memory growth to True when using GPU
• model: head: correct line ending
• tf: add **kwargs parameter to compile() and fit()
• tf: fix strange ground truth values issue

-- Hyeonki Hong [e-mail] Sat, 05 Dec 2020 18:26:08 +0900

tensorflow-yolov4 (2.0.0) unstable; urgency=medium

• tf:dataset: add data format check
• common: base_class: change input_size from integer to (width, height)
• common: base_class: set default arguments for candidates_to_pred_bboxes
• tf: dataset: modify variable initialization according to input_size type change
• common: predict: reflect input_size type change
• common: media: reflect input_size type change to resize_image
• tf:dataset: reflect input_size type change to data augmentation
• common: base_class: reflect predict.fit_pred_bboxes_to_original change
• common: predict: fix problem that occurred because image size change was not considered
• tf: reflect input_size type change
• model: head: reflect input_size type change
• tf: train: reflect input_size type change
• common: base_class: change strides to immutable constant
• common: base_class: add cap.release() to release resources
• tflite: reflect input_size type change
• tf: optimize predict() using @tf.function
• common: predict: fix problem caused by incorrect iou calculation

-- Hyeonki Hong [e-mail] Tue, 27 Oct 2020 00:11:59 +0900

tensorflow-yolov4 (1.2.1) unstable; urgency=medium

• tf: set num_sample to the number of dataset, if not specified
• tf: modify to bring images in order on post-training
• common: base_class: add iou/score_threshold argument
• tf: add images_optional argument in save_dataset_for_mAP()
• common: base_class: modify to call cv2.namedWindow once
• common: media: modify to create colors once
• tf: dataset: update docs

-- Hyeonki Hong [e-mail] Fri, 04 Sep 2020 20:21:56 +0900

tensorflow-yolov4 (1.2.0) unstable; urgency=medium

• model: remove tpu argument
• tf: reflect model changes
• tflite: remove tpu_hair
• tf, tflite: move inference into BaseClass
• model: yolov4: remove unused import
• pylint: update .pylintrc
• common: base_class: modify inference() to be able to use cam
• common: base_class: add fps display to screen

-- Hyeonki Hong [e-mail] Tue, 11 Aug 2020 21:43:29 +0900

tensorflow-yolov4 (1.1.0) unstable; urgency=medium

• model: add kernel_regularizer argument
• tf: remove subdivision
• tf: dataset: fix issue where coordinates become 0 due to repeated division
• tf: dataset: modify to shuffle when augmentation
• tf: train: modify order of declaration
• tf: train: give xiou_loss to more weight
• tf: train: use keras.losses.BinaryCrossentropy to avoid nan problem
• tf: train: add verbose argument
• tf: dataset: change converted_coco format
• tf: add image_path_prefix argument to load_dataset()
• tf: add loss_verbose argument to compile()
• tf: reflect dataset format change to save_dataset_for_mAP()

-- Hyeonki Hong [e-mail] Mon, 10 Aug 2020 19:09:15 +0900

tensorflow-yolov4 (1.0.0) unstable; urgency=medium

• tf: dataset: add label smooting argument
• tf: add label_smooting argument to load_dataset()
• tf: train: set truth_conf threshold to 0.5 because of label smoothing

-- Hyeonki Hong [e-mail] Wed, 05 Aug 2020 13:05:02 +0900

tensorflow-yolov4 (0.24.0) unstable; urgency=medium

• tf: set default num of sample for mAP to 1000
• tf: fix error handling when image is empty

-- Hyeonki Hong [e-mail] Tue, 04 Aug 2020 12:40:25 +0900

tensorflow-yolov4 (0.23.0) unstable; urgency=medium

• tf: dataset: add error handling when bboxes are empty
• common: media: replace space of class name to underbar
• tf: implement save_dataset_for_mAP()
• tf: dataset: add error handling when image is empty

-- Hyeonki Hong [e-mail] Mon, 03 Aug 2020 20:49:51 +0900

tensorflow-yolov4 (0.22.0) unstable; urgency=medium

• tf: add TFLITE_BUILTINS, SELECT_TF_OPS to default supported_ops
• tf: dataset: modify to check if image exists when creating a dataset
• tf: dataset: remove preprocess_dataset
• tf: dataset: add cut_out
• tf: dataset: add _next_random_augmentation_data()
• tf: dataset: add mix_up
• tf: dataset: implement mosaic

-- Hyeonki Hong [e-mail] Sun, 02 Aug 2020 17:22:43 +0900

tensorflow-yolov4 (0.21.0) unstable; urgency=medium

• tf: train: set epsilon used for division to 1e-8
• tf: train: add giou and iou to iou_type
• tf: train: fix loss function of bbox_probabilities

-- Hyeonki Hong [e-mail] Thu, 30 Jul 2020 05:18:12 +0900

tensorflow-yolov4 (0.20.0) unstable; urgency=medium

• tf: train: modify epsilon to 1e-9
• tf: train: remove weight for conf_noobj_loss
• tf: add arguments to fit()
• tf: dataset: fix problem of not finding images
• tf: add SaveWeightsCallback

-- Hyeonki Hong [e-mail] Wed, 29 Jul 2020 05:06:25 +0900

tensorflow-yolov4 (0.19.0) unstable; urgency=medium

• tf: weights: modify 'set' to 'load'
• tf: weights: implement *_save_weights
• tf: add save_weights()

-- Hyeonki Hong [e-mail] Fri, 24 Jul 2020 06:22:14 +0900

tensorflow-yolov4 (0.18.0) unstable; urgency=medium

• yolov4: clarify batch number
• tf: remove by_name in load_weights
• tf: dataset: simplify code
• tf: dataset: fix problem of making a batch with same image

-- Hyeonki Hong [e-mail] Thu, 23 Jul 2020 20:30:23 +0900

tensorflow-yolov4 (0.17.0) unstable; urgency=medium

• tf: dataset: fix index range for Tiny
• tf: remove expect_partial() in load_weights()
• tflite: fix issue with the number of outputs by model
• model: add tpu argument in Tiny
• tflite: add tpu argument
• utility: rename to common
• common: implement BaseClass
• tflite: add tensorflow.lite
• model: head: change dimension from 4D to 3D in Tiny
• tflite: add tpu_hair
• common: media: add type cast in resize_image
• tf: add num_calibration_steps argument in save_as_tflite
• common: base_class: move strides property from tf
• tf: set by_name to True in load_weights

Thanks to @RealHandy

-- Hyeonki Hong [e-mail] Thu, 23 Jul 2020 15:43:08 +0900

tensorflow-yolov4 (0.16.0) unstable; urgency=medium

• tf: add tiny argument to init and remove from others

-- Hyeonki Hong [e-mail] Wed, 15 Jul 2020 13:21:12 +0900

tensorflow-yolov4 (0.15.0) unstable; urgency=medium

• tf: add quantization and data_set args to save_as_tflite
• utility: media: add string length check
• model: add activation args
• model: head: remove for loop
• model: backbone: implement CSPDarknet53Tiny
• model: neck: implement PANetTiny
• model: head: implement YOLOv3HeadTiny
• model: yolov4: implement YOLOv4Tiny
• tf: weights: implement tiny funcs
• tf: reflect YOLOv4Tiny

-- Hyeonki Hong [e-mail] Wed, 15 Jul 2020 03:43:03 +0900

tensorflow-yolov4 (0.14.0) unstable; urgency=medium

• github: add python publish action
• model: neck: use bilinear in UpSampling2D
• test: update script
• tflite: refactor YOLOv4
• utility: predict: fix according to pylint warning
• pylint: update .pylintrc
• model: head: use tf.constant to avoid broadcasting
• github: add python lint action

-- Hyeonki Hong [e-mail] Tue, 14 Jul 2020 02:14:12 +0900

tensorflow-yolov4 (0.13.0) unstable; urgency=medium

• tf: remove tensorboard callback
• tf: modify compile() and fit() to be similar to model
• yolov4: rename 'data' to 'test'
• utility: media: update docs and variable name
• utility: predict: remove batch_size
• utility: media: rename funcs
• tf: add utility funcs to YOLOv4 member funcs
• test: add test script

-- Hyeonki Hong [e-mail] Mon, 13 Jul 2020 13:36:32 +0900

tensorflow-yolov4 (0.12.0) unstable; urgency=medium

• utility: train: refactor bbox_*iou and remove duplicate funcs
• yolov4: rename parameters
• yolov4: remove utils
• tf: add FileNotFoundError in YOLOv4.inference()
• utility: predict: add dimension for batch size
• pylint: update .pylintrc
• tf: add YOLOv4.save_as_tflite()
• model: clean up
• utility: weights: move to tf.weights
• utility: train: move to tf.train
• model: neck: implement PANet
• mdel: head: implement YOLOv3Head
• model: yolov4: Apply neck and head class
• yolov4: reflect model changes
• tf: dataset: add batch_size
• tf: train: implement YOLOv4Loss
• utility: media: modify rectangle thickness
• model: common: use softplus instead of ln(1+exp(x))
• tf: train: use epsilon instead of tf.math.divide_no_nan
• tf: refactor YOLOv4.fit() and .compile()

-- Hyeonki Hong [e-mail] Sun, 12 Jul 2020 03:50:50 +0900

tensorflow-yolov4 (0.11.0) unstable; urgency=medium

• tf: remove utils.draw_bbox in predict()
• yolov4: rename files and functions and change order
• utility: utils: remove get_anchors()
• utility: media: impelment resize(), draw_bbox()
• utility: utils: implement DIoU_NMS
• utility: utils: fix dimensional calculation problems
• utility: refactor dataset
• tf: remove train
• utility: train: implement make_compiled_loss()
• utility: media: fix bug that could not resize some images
• utility: train: remove problem of division by zero

-- Hyeonki Hong [e-mail] Mon, 29 Jun 2020 21:05:39 +0900

tensorflow-yolov4 (0.10.0) unstable; urgency=medium

• core: yolov4: refactor decode()
• core: utils: remove sigmoid in postprocess_bbboxe()
• tf: apply YOLOv4 changes to make_model()
• core: yolov4: move decode_train() to tf.YOLOv4.train()

-- Hyeonki Hong [e-mail] Thu, 25 Jun 2020 00:48:44 +0900

tensorflow-yolov4 (0.9.0) unstable; urgency=medium

• tf: modify hyperparameters as properties
• tf: add weights_type argument to load_weights()
• core: utils: implement _np_fromfile()
• core: utils: implement a way to partially load weights
• tf: train: move learning_rate_* to argument
• core: move YOLOConv2D to common
• core: common: remove bn argument of YOLOConv2D
• core: utils: refactor yolo_conv2d_set_weights
• core: yolov4: refactor YOLOv4
• core: utils: refactor load_weights
• tf: refactor make_model
• yolov4: change YoloV4 to YOLOv4

-- Hyeonki Hong [e-mail] Wed, 24 Jun 2020 02:58:27 +0900

tensorflow-yolov4 (0.8.0) unstable; urgency=medium

• core: use tf.keras.layers.UpSampling2D
• core: refactor Mish
• core: common: remove residual_block
• core: remove sequential in _ResBlock
• core: backbone: Set LeakyReLU's alpha to 0.1

-- Hyeonki Hong [e-mail] Tue, 23 Jun 2020 02:21:01 +0900

tensorflow-yolov4 (0.7.0) unstable; urgency=medium

• tf: fix to proceed to the next step even if an error occurs
• tf: modify video_interval_ms to cv_waitKey_delay
• core: backbone: refactor CSPDarknet53
• core: utils: implement csp_darknet53_set_weights()

-- Hyeonki Hong [e-mail] Mon, 22 Jun 2020 23:01:32 +0900

tensorflow-yolov4 (0.6.0) unstable; urgency=medium

• tf: set first_step_epochs according to the weight usage
• tf: fix syntax error

-- Hyeonki Hong [e-mail] Fri, 19 Jun 2020 17:09:57 +0900

tensorflow-yolov4 (0.5.0) unstable; urgency=medium

• core: dataset: add yolo type
• tf: add dataset_type parameter to YoloV4.train
• tf: add epochs parameter to train
• tf: add save_interval parameter to train

-- Hyeonki Hong [e-mail] Fri, 19 Jun 2020 14:30:50 +0900

tensorflow-yolov4 (0.4.0) unstable; urgency=medium

• core: dataset: remove cfg module
• tf: implement YoloV4.train

-- Hyeonki Hong [e-mail] Thu, 11 Jun 2020 17:45:44 +0900

tensorflow-yolov4 (0.3.0) unstable; urgency=medium

• core: utils: use numpy instead of tensorflow
• pypi: remove install_requires and change to manual installation
• yolov4: add video_interval_ms

-- Hyeonki Hong [e-mail] Mon, 08 Jun 2020 23:59:41 +0900

tensorflow-yolov4 (0.2.0) unstable; urgency=medium

• pylint: create .pylintrc and run black
• core: remove config.py
• yolov4: change tfyolov4 to yolov4
• yolov4: remove detect**.py and implement YoloV4.inference

-- Hyeonki Hong [e-mail] Mon, 08 Jun 2020 02:20:49 +0900

tensorflow-yolov4 (0.1.0) unstable; urgency=medium

• yolov4: fork from 'hunglc007/tensorflow-yolov4-tflite'

-- Hyeonki Hong [e-mail] Fri, 05 Jun 2020 20:17:45 +0900