Merge pull request #14 from FlyingQianMM/develop_qh

change ins_seg demo datasets

docs/apis/load_model.md

@@ -16,21 +16,21 @@ paddlex.load_model(model_dir)
 * **paddlex.cv.models**, 模型类。
 
 ### 示例
-> 1. [点击下载](https://bj.bcebos.com/paddlex/models/garbage_epoch_12.tar.gz)PaddleX在垃圾分拣数据上训练的MaskRCNN模型
-> 2. [点击下载](https://bj.bcebos.com/paddlex/datasets/garbage_ins_det.tar.gz)垃圾分拣数据集
+> 1. [点击下载](https://bj.bcebos.com/paddlex/models/xiaoduxiong_epoch_12.tar.gz)PaddleX在小度熊分拣数据上训练的MaskRCNN模型
+> 2. [点击下载](https://bj.bcebos.com/paddlex/datasets/xiaoduxiong_ins_det.tar.gz)小度熊分拣数据集
 
 ```
 import paddlex as pdx
 
-model_dir = './garbage_epoch_12'
-data_dir = './garbage_ins_det/JPEGImages'
-ann_file = './garbage_ins_det/val.json'
+model_dir = './xiaoduxiong_epoch_12'
+data_dir = './xiaoduxiong_ins_det/JPEGImages'
+ann_file = './xiaoduxiong_ins_det/val.json'
 
 # 加载垃圾分拣模型
 model = pdx.load_model(model_dir)
 
 # 预测
-pred_result = model.predict('./garbage_ins_det/JPEGImages/000114.bmp')
+pred_result = model.predict('./xiaoduxiong_ins_det/JPEGImages/WechatIMG114.jpeg')
 
 # 在验证集上进行评估
 eval_reader = pdx.cv.datasets.CocoDetection(data_dir=data_dir,

docs/apis/visualize.md

@@ -14,13 +14,13 @@ paddlex.det.visualize(image, result, threshold=0.5, save_dir=None)
 > * **save_dir**(str): 可视化结果保存路径。若为None，则表示不保存，该函数将可视化的结果以np.ndarray的形式返回；若设为目录路径，则将可视化结果保存至该目录下
 
 ### 使用示例
-> 点击下载如下示例中的[模型](https://bj.bcebos.com/paddlex/models/garbage_epoch_12.tar.gz)和[测试图片](https://bj.bcebos.com/paddlex/datasets/garbage.bmp)
+> 点击下载如下示例中的[模型](https://bj.bcebos.com/paddlex/models/xiaoduxiong_epoch_12.tar.gz)和[测试图片](https://bj.bcebos.com/paddlex/datasets/xiaoduxiong.jpeg)
 ```
 import paddlex as pdx
-model = pdx.load_model('garbage_epoch_12')
-result = model.predict('garbage.bmp')
-pdx.det.visualize('garbage.bmp', result, save_dir='./')
-# 预测结果保存在./visualize_garbage.bmp
+model = pdx.load_model('xiaoduxiong_epoch_12')
+result = model.predict('xiaoduxiong.jpeg')
+pdx.det.visualize('xiaoduxiong.jpeg', result, save_dir='./')
+# 预测结果保存在./visualize_xiaoduxiong.jpeg
 ```
 
 ## 语义分割预测结果可视化

docs/tutorials/train/detection.md

@@ -2,7 +2,7 @@
 
 ------
 
-更多检测模型在VOC数据集或COCO数据集上的训练代码可参考[代码tutorials/train/detection/faster_rcnn_r50_fpn.py](https://github.com/PaddlePaddle/PaddleX/blob/develop/tutorials/train/detection/faster_rcnn_r50_fpn.py)、[代码tutorials/train/detection/yolov3_mobilenetv1.py](https://github.com/PaddlePaddle/PaddleX/blob/develop/tutorials/train/detection/yolov3_mobilenetv1.py)。
+更多检测模型在VOC数据集或COCO数据集上的训练代码可参考[代码tutorials/train/detection/faster_rcnn_r50_fpn.py](https://github.com/PaddlePaddle/PaddleX/blob/develop/tutorials/train/detection/faster_rcnn_r50_fpn.py)、[代码tutorials/train/detection/yolov3_darknet53.py](https://github.com/PaddlePaddle/PaddleX/blob/develop/tutorials/train/detection/yolov3_darknet53.py)。
 
 **1.下载并解压训练所需的数据集**
 
@@ -116,4 +116,4 @@ predict_result = model.predict('./insect_det/JPEGImages/1968.jpg')
 pdx.det.visualize('./insect_det/JPEGImages/1968.jpg', predict_result, threshold=0.5, save_dir='./output/faster_rcnn_r50_fpn')
 ```
 
-![](../images/visualized_fasterrcnn.jpg)
+![](../../images/visualized_fasterrcnn.jpg)

docs/tutorials/train/instance_segmentation.md

@@ -14,11 +14,11 @@ os.environ['CUDA_VISIBLE_DEVICES'] = '0'
 import paddlex as pdx
 ```
 
-> 这里使用垃圾分拣数据集，训练集、验证集和测试共包含283个样本，6个类别。
+> 这里使用小度熊分拣数据集，训练集、验证集和测试共包含21个样本，1个类别。
 
 ```python
-garbage_dataset = 'https://bj.bcebos.com/paddlex/datasets/garbage_ins_det.tar.gz'
-pdx.utils.download_and_decompress(garbage_dataset, path='./')
+xiaoduxiong_dataset = 'https://bj.bcebos.com/paddlex/datasets/xiaoduxiong_ins_det.tar.gz'
+pdx.utils.download_and_decompress(xiaoduxiong_dataset, path='./')
 ```
 
 **2.定义训练和验证过程中的数据处理和增强操作**
@@ -47,19 +47,19 @@ eval_transforms = transforms.Compose([
 
 ```python
 train_dataset = pdx.datasets.CocoDetection(
-    data_dir='garbage_ins_det/JPEGImages',
-    ann_file='garbage_ins_det/train.json',
+    data_dir='xiaoduxiong_ins_det/JPEGImages',
+    ann_file='xiaoduxiong_ins_det/train.json',
     transforms=train_transforms,
     shuffle=True)
 eval_dataset = pdx.datasets.CocoDetection(
-    data_dir='garbage_ins_det/JPEGImages',
-    ann_file='garbage_ins_det/val.json',
+    data_dir='xiaoduxiong_ins_det/JPEGImages',
+    ann_file='xiaoduxiong_ins_det/val.json',
     transforms=eval_transforms)
 ```
 
 **4.创建Mask RCNN模型，并进行训练**
 
-> 创建带FPN结构的Mask RCNN模型，`num_classes` 需要设置为包含背景类的类别数，即: 目标类别数量(6) + 1。
+> 创建带FPN结构的Mask RCNN模型，`num_classes` 需要设置为包含背景类的类别数，即: 目标类别数量(1) + 1。
 
 ```python
 num_classes = len(train_dataset.labels)
@@ -75,6 +75,7 @@ model.train(
     train_batch_size=1,
     eval_dataset=eval_dataset,
     learning_rate=0.00125,
+    warmup_steps=10,
     lr_decay_epochs=[8, 11],
     save_dir='output/mask_rcnn_r50_fpn',
     use_vdl=True)
@@ -98,19 +99,19 @@ print("eval_metrics:", eval_metrics)
 > 结果输出：
 
 ```python
-eval_metrics: {'bbox_mmap': 0.858306, 'segm_mmap': 0.864278}
+eval_metrics: OrderedDict([('bbox_mmap', 0.5038283828382838), ('segm_mmap', 0.7025202520252025)])
 
 ```
 
 > 训练完用模型对图片进行测试。
 
 ```python
-predict_result = model.predict('./garbage_ins_det/JPEGImages/000114.bmp')
+predict_result = model.predict('./xiaoduxiong_ins_det/JPEGImages/WechatIMG114.jpeg')
 ```
 
 > 可视化测试结果：
 
 ```python
-pdx.det.visualize('./garbage_ins_det/JPEGImages/000114.bmp', predict_result, threshold=0.7, save_dir='./output/mask_rcnn_r50_fpn')
+pdx.det.visualize('./xiaoduxiong_ins_det/JPEGImages/WechatIMG114.jpeg', predict_result, threshold=0.7, save_dir='./output/mask_rcnn_r50_fpn')
 ```
-![](../../images/visualized_maskrcnn.bmp)
+![](../../images/visualized_maskrcnn.jpeg)

tutorials/train/detection/mask_rcnn_r50_fpn.py

@@ -5,9 +5,9 @@ os.environ['CUDA_VISIBLE_DEVICES'] = '0'
 from paddlex.det import transforms
 import paddlex as pdx
 
-# 下载和解压垃圾分拣数据集
-garbage_dataset = 'https://bj.bcebos.com/paddlex/datasets/garbage_ins_det.tar.gz'
-pdx.utils.download_and_decompress(garbage_dataset, path='./')
+# 下载和解压小度熊分拣数据集
+xiaoduxiong_dataset = 'https://bj.bcebos.com/paddlex/datasets/xiaoduxiong_ins_det.tar.gz'
+pdx.utils.download_and_decompress(xiaoduxiong_dataset, path='./')
 
 # 定义训练和验证时的transforms
 train_transforms = transforms.Compose([
@@ -25,13 +25,13 @@ eval_transforms = transforms.Compose([
 
 # 定义训练和验证所用的数据集
 train_dataset = pdx.datasets.CocoDetection(
-    data_dir='garbage_ins_det/JPEGImages',
-    ann_file='garbage_ins_det/train.json',
+    data_dir='xiaoduxiong_ins_det/JPEGImages',
+    ann_file='xiaoduxiong_ins_det/train.json',
     transforms=train_transforms,
     shuffle=True)
 eval_dataset = pdx.datasets.CocoDetection(
-    data_dir='garbage_ins_det/JPEGImages',
-    ann_file='garbage_ins_det/val.json',
+    data_dir='xiaoduxiong_ins_det/JPEGImages',
+    ann_file='xiaoduxiong_ins_det/val.json',
     transforms=eval_transforms)
 
 # 初始化模型，并进行训练
@@ -48,6 +48,7 @@ model.train(
     train_batch_size=1,
     eval_dataset=eval_dataset,
     learning_rate=0.00125,
+    warmup_steps=10,
     lr_decay_epochs=[8, 11],
     save_dir='output/mask_rcnn_r50_fpn',
     use_vdl=True)

tutorials/train/detection/yolov3_mobilenetv1.py → tutorials/train/detection/yolov3_darknet53.py

@@ -44,7 +44,7 @@ eval_dataset = pdx.datasets.VOCDetection(
 # 浏览器打开 https://0.0.0.0:8001即可
 # 其中0.0.0.0为本机访问，如为远程服务, 改成相应机器IP
 num_classes = len(train_dataset.labels)
-model = pdx.det.YOLOv3(num_classes=num_classes)
+model = pdx.det.YOLOv3(num_classes=num_classes, backbone='DarkNet53')
 model.train(
     num_epochs=270,
     train_dataset=train_dataset,
@@ -52,5 +52,5 @@ model.train(
     eval_dataset=eval_dataset,
     learning_rate=0.000125,
     lr_decay_epochs=[210, 240],
-    save_dir='output/yolov3_mobilenetv1',
+    save_dir='output/yolov3_darknet53',
     use_vdl=True)