pybaseutils 2.3.5__tar.gz → 2.3.6__tar.gz

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pybaseutils
-Version: 2.3.5
+Version: 2.3.6
 Summary: pybaseutils
 Home-page: https://github.com/PanJinquan/base-utils
 Author: PanJinquan

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils/__init__.py

@@ -8,4 +8,4 @@
 # --------------------------------------------------------
 """
 
-__version__ = '2.3.5'
+__version__ = '2.3.6'

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils/audio/audio_utils.py

@@ -32,7 +32,7 @@ def load_pcm(filename):
 
 def float32toint16(data_f32):
     omin = -32768
-    omax = 32767
+    omax = 32768
     imin = -1.0
     imax = 1.0
     # 将数据float32从[-1.0,1.0]映射到int16
@@ -43,7 +43,7 @@ def float32toint16(data_f32):
 
 def int16tofloat32(data_int16):
     imin = -32768
-    imax = 32767
+    imax = 32768
     omin = -1.0
     omax = 1.0
     audio_data = np.array(data_int16, dtype=np.float32)
@@ -148,7 +148,7 @@ def save_pcm_from_numpy(audio_data: np.ndarray, pcm_file, data_type=np.int16):
     :return:
     """
     omin = -32768
-    omax = 32767
+    omax = 32768
     imin = -1.0
     imax = 1.0
     # 将数据float32转换到int16
@@ -159,7 +159,7 @@ def save_pcm_from_numpy(audio_data: np.ndarray, pcm_file, data_type=np.int16):
     return bytes
 
 
-def librosa_load(audio_file, sr=None, mono=True):
+def librosa_load(audio_file, sr=None, mono=True, dtype="float32"):
     """
     默认将多声道音频文件转换为单声道，并返回一维数组；
     如果你需要处理多声道音频文件，可以使用 mono=False,参数来保留所有声道，并返回二维数组。
@@ -168,47 +168,55 @@ def librosa_load(audio_file, sr=None, mono=True):
     :param mono: 设置为true是单通道，否则是双通道
     :return:
     """
-    audio_data, sr = librosa.load(audio_file, sr=sr, mono=mono)  # 非常耗时
+    # TODO 已经对齐soundfile_load的效果
+    audio_data, sr = librosa.load(audio_file, sr=sr, mono=mono)  # 非常耗时，返回是float32的数据
+    if dtype == "int16":  # 将 float32 [-1.0, 1.0] 转换为 int16,
+        audio_data = float32toint16(audio_data)
     return audio_data, sr
 
 
-def soundfile_load(audio_file, sr=None, mono=True):
+def soundfile_load(audio_file, sr=None, mono=True, dtype="float32"):
     """
     默认会将多声道音频文件的每个声道分别存储在二维数组的不同列中，因此返回的是一个二维数组
+
     :param audio_file:
     :param sr: sampling rate 当设置的采样率与音频原始采样率不一致时，会进行重采样，导致非常耗时
     :param mono: 设置为true是单通道，否则是双通道
     :return:
     """
+    # TODO 默认情况下，soundfile会将数据归一化到 [-1.0, 1.0] 的浮点数范围内
+    # audio_data, orig_sr = sf.read(audio_file, samplerate=None, dtype="int16")  # int16不支持重采样
     audio_data, orig_sr = sf.read(audio_file, samplerate=None, dtype="float32")  # 不能直接修改samplerate
     if mono and audio_data.ndim == 2:
         audio_data = np.mean(audio_data, axis=-1)
     if sr is None or orig_sr == sr:
         sr = orig_sr
-    else:
+    else:  # TODO librosa.resample 要求输入是浮点数float32
         audio_data = librosa.resample(y=audio_data, orig_sr=orig_sr, target_sr=sr)  # 使用librosa进行重采样至目标采样率
+    if dtype == "int16":  # 将 float32 [-1.0, 1.0] 转换为 int16,
+        audio_data = float32toint16(audio_data)
     return audio_data, sr
 
 
 def wavfile_load(audio_file):
     sr, wav_data = wavfile.read(audio_file)  # int16类型
-    # 转为float64类型
-    wav_data = wav_data / (32768)
-    # wav_data:[-0.03305054 -0.03561401 -0.038114697]
     return wav_data, sr
 
 
-def read_audio(audio_file, sr=16000, mono=True):
+def read_audio(audio_file, sr=16000, mono=True, dtype="float32", use_sf=True):
     """
     默认将多声道音频文件转换为单声道，并返回一维数组；
     如果你需要处理多声道音频文件，可以使用 mono=False,参数来保留所有声道，并返回二维数组。
     :param audio_file:
     :param sr: sampling rate
     :param mono: 设置为true是单通道，否则是双通道
+    :param dtype: 数据类型，int16, float32
     :return:
     """
-    audio_data, sr = librosa_load(audio_file, sr=sr, mono=mono)  # 巨慢
-    # audio_data, sr = soundfile_load(audio_file, sr=sr, mono=mono)  # 巨慢
+    if use_sf:
+        audio_data, sr = soundfile_load(audio_file, sr=sr, mono=mono, dtype=dtype)  # 巨慢
+    else:
+        audio_data, sr = librosa_load(audio_file, sr=sr, mono=mono, dtype=dtype)  # 巨慢
     return audio_data, sr
 
 
@@ -216,10 +224,11 @@ def save_audio(audio_file, audio_data, sr=16000):
     """
     save audio file
     :param audio_file:
-    :param audio_data:
+    :param audio_data: float32
     :param sr: sampling rate
     :return:
     """
+    assert audio_data.dtype == np.float32, "input data must be float32"
     sf.write(audio_file, audio_data, samplerate=sr)
 
 

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils/dict_uils.py

@@ -26,7 +26,7 @@ def str2dict(data: str):
     return None
 
 
-def dict_slice(data: Dict, slice=()):
+def dict_slice(data: Dict, slice=[]):
     """
     字典像列表一样切片
     :param data:

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils/image_utils.py

@@ -136,16 +136,19 @@ def show_batch_image(title, batch_images, index=0):
         cv_show_image(title, image)
 
 
-def show_image_plt(title, image):
+def show_image_plt(title, image, use_rgb=True):
     """
     use matplotlib to show image
     调用matplotlib显示RGB图片
     :param title: 图像标题
-    :param image: 图像的数据
+    :param image: 图像的数据 BGR格式
+    :param use_rgb: True:输入image是RGB的图像, False:返输入image是BGR格式的图像
     :return:
     """
     # plt.figure("show_image")
     # print(image.dtype)
+    if image.shape[-1] == 3 and (not use_rgb):
+        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)  # 将BGR转为RGB
     channel = len(image.shape)
     if channel == 3:
         plt.imshow(image)
@@ -154,6 +157,7 @@ def show_image_plt(title, image):
     plt.axis('on')  # 关掉坐标轴为 off
     plt.title(title)  # 图像题目
     plt.show()
+    return image
 
 
 plot_image = show_image_plt
@@ -537,7 +541,7 @@ def requests_url(url, timeout=None):
     return stream
 
 
-def read_images_url(url: str, size=None, norm=False, use_rgb=False, timeout=5):
+def read_image_url(url: str, size=None, norm=False, use_rgb=False, timeout=5):
     """
     根据url或者图片路径，读取图片
     :param url:
@@ -570,6 +574,16 @@ def read_images_url(url: str, size=None, norm=False, use_rgb=False, timeout=5):
     return image
 
 
+def read_images_url(url: str or list, size=None, norm=False, use_rgb=False, timeout=5):
+    if isinstance(url, str): return read_image_url(url, size=size, norm=norm, use_rgb=use_rgb, timeout=timeout)
+    batch = []
+    for path in url:
+        image = read_image_url(path, size=size, norm=norm, use_rgb=use_rgb, timeout=timeout)
+        if image is None:  print("no image:{}".format(path))
+        batch.append(image)
+    return batch
+
+
 def read_image_batch(image_list):
     """
     批量读取图片

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pybaseutils
-Version: 2.3.5
+Version: 2.3.6
 Summary: pybaseutils
 Home-page: https://github.com/PanJinquan/base-utils
 Author: PanJinquan

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/pybaseutils.egg-info/SOURCES.txt

@@ -187,7 +187,9 @@ test_py/gradio_app.py
 test_py/gradio_app_v2.py
 test_py/kafka_worker.py
 test_py/men_tracemalloc.py
+test_py/model_process.py
 test_py/performance.py
+test_py/slowfastnet.py
 test_py/Image_enhance/__init__.py
 test_py/Image_enhance/dmeo01.py
 test_py/WebCrawler/__init__.py

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/test_py/aije/demo_labelme_crop.py

@@ -98,11 +98,12 @@ if __name__ == '__main__':
     对labelme的数据目标进行裁剪，用于制作分类数据集
     """
     # 接地引线'
-    names = ['保护屏标示牌', '显示屏', '通道切换开关', '信号复归按钮', '红色圆形标签,黄色圆形标签','地极端子','地极标签',
+    names = ['保护屏标示牌', '显示屏', '通道切换开关', '信号复归按钮', '红色圆形标签,黄色圆形标签', '地极端子',
+             '地极标签',
              '红色压板开关合上,红色压板开关断开,白色压板开关合上,白色压板开关断开', '重合闸把手停用,重合闸把手单重',
-             '本体端子箱关闭,本体端子箱打开', '主变端子箱关闭,主变端子箱打开','分接开关箱关闭,分接开关箱打开',
-             '主变压器标示牌','端子箱标示牌','油温表','绕组温度表','油位指示标示牌','本体阀门','呼吸器油杯',
-             '沉降观察点标示牌','冷却器控制箱打开,冷却器控制箱关闭','冷却器','风扇','潜油泵','油流指示器','喷淋启动装置'
+             '本体端子箱关闭,本体端子箱打开', '主变端子箱关闭,主变端子箱打开', '分接开关箱关闭,分接开关箱打开',
+             '主变压器标示牌', '端子箱标示牌', '油温表', '绕组温度表', '油位指示标示牌', '本体阀门', '呼吸器油杯',
+             '沉降观察点标示牌', '冷却器控制箱打开,冷却器控制箱关闭', '冷却器', '风扇', '潜油泵', '油流指示器', '喷淋启动装置'
              ]
     target_name = ['身穿工作服', '未穿工作服',
                    '绝缘鞋', '脚穿绝缘鞋', '长筒靴', '脚穿长筒靴', '其他鞋', '脚穿其他鞋',
@@ -118,13 +119,20 @@ if __name__ == '__main__':
                    "红色圆形标签", "黄色圆形标签",
                    "白色压板开关合上", "白色压板开关断开", "红色压板开关合上", "红色压板开关断开",
                    "重合闸把手停用", "重合闸把手单重",
-                   '本体端子箱关闭','本体端子箱打开', '主变端子箱关闭','主变端子箱打开', '分接开关箱关闭','分接开关箱打开',
-                   '冷却器控制箱打开','冷却器控制箱关闭'
+                   '本体端子箱关闭', '本体端子箱打开',
+                   '主变端子箱关闭', '主变端子箱打开',
+                   '分接开关箱关闭', '分接开关箱打开',
+                   '冷却器控制箱打开', '冷却器控制箱关闭',
+                   '万用表线头', '万用表红色线头', '万用表黑色线头'
                    ]
     # 还差核相的数据
     datasets = [
         # TODO 训练集
-        '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det/02-东莞-220KV主变压器日常巡视/dataset-v02/images',
+        '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det/02-东莞-220KV主变压器日常巡视/dataset-test01/images',
+        '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det/02-东莞-220KV主变压器日常巡视/dataset-test02/images',
+        '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det/01-东莞-投退重合闸操作/dataset-test01/images',
+        '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det/01-东莞-投退重合闸操作/dataset-test02/images',
+
     ]
     # dataroot = '/home/PKing/nasdata/dataset-dmai/AIJE/dataset/aije-v2-det'
     # datasets = get_sub_dataset(dataroot)

pybaseutils-2.3.6/test_py/demo3.py

@@ -0,0 +1,48 @@
+# -*- coding: utf-8 -*-
+"""
+# --------------------------------------------------------
+# @Author : Pan
+# @E-mail :
+# @Date   : 2025-07-08 14:10:15
+# @Brief  :
+# --------------------------------------------------------
+"""
+import cv2
+import numpy as np
+from pybaseutils import file_utils, image_utils
+
+import cv2
+
+# 打开默认摄像头（0表示第一个摄像头）
+rtsp_url = "rtsp://username:password@192.168.1.64:554/stream1"
+# cap = cv2.VideoCapture("192.168.2.52livestream")
+# cap = cv2.VideoCapture("rtsp://0.0.0.0:8554/")
+# cap = cv2.VideoCapture("rtsp://192.168.2.52:8554/")
+cap = cv2.VideoCapture("rtsp://admin:dm-ai12345@192.168.2.222")
+
+# 检查摄像头是否成功打开
+if not cap.isOpened():
+    print("无法打开摄像头")
+    exit()
+
+print("按 'q' 键退出")
+
+while True:
+    # 读取一帧
+    ret, frame = cap.read()
+
+    # 如果读取成功，ret为True
+    if not ret:
+        print("无法读取帧")
+        break
+
+    # 显示帧
+    cv2.imshow('Camera', frame)
+
+    # 按'q'键退出
+    if cv2.waitKey(1) & 0xFF == ord('q'):
+        break
+
+# 释放资源
+cap.release()
+cv2.destroyAllWindows()

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/test_py/demo_gif_video.py

@@ -38,8 +38,8 @@ def android_gif():
 
 
 def python_gif():
-    video_file = "/home/PKing/nasdata/Project/3D/Car-Distance-Speed-Estimation/data/output/video0_004.mp4"
-    video_utils.video2gif(video_file, interval=6, func=resize_fun, fps=4, end=6, use_pil=True, vis=True)
+    video_file = "/home/PKing/nasdata/Project/3D/Car-Distance-Speed-Estimation/data/output/video1_003.mp4"
+    video_utils.video2gif(video_file, interval=6, func=resize_fun, fps=4,start=3,end=10, use_pil=True, vis=True)
 
 
 def image_gif():

pybaseutils-2.3.6/test_py/model_process.py

@@ -0,0 +1,30 @@
+# -*- coding:utf-8 -*-
+"""
+提供模型相关方法
+"""
+import logging
+import threading
+import time
+import asyncio
+from pybaseutils import json_utils
+
+
+class EvaluationMethod():
+    """
+    Evaluation核心实现方法类
+    """
+
+    def __init__(self):
+        self.msg = "ok"
+
+    async def calligraphy_hard_evaluation(self, params):
+        results = {"result": params, "msg": self.msg}
+        return results
+
+
+if __name__ == '__main__':
+    w = EvaluationMethod()
+    params = {"data": []}
+    # 使用 asyncio.run() 来运行异步方法
+    result = asyncio.run(w.calligraphy_hard_evaluation(params))
+    print(result)

pybaseutils-2.3.6/test_py/slowfastnet.py

@@ -0,0 +1,217 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from torch.autograd import Variable
+
+__all__ = ['resnet50', 'resnet101','resnet152', 'resnet200']
+
+
+
+class Bottleneck(nn.Module):
+    expansion = 4
+
+    def __init__(self, inplanes, planes, stride=1, downsample=None, head_conv=1):
+        super(Bottleneck, self).__init__()
+        if head_conv == 1:
+            self.conv1 = nn.Conv3d(inplanes, planes, kernel_size=1, bias=False)
+            self.bn1 = nn.BatchNorm3d(planes)
+        elif head_conv == 3:
+            self.conv1 = nn.Conv3d(inplanes, planes, kernel_size=(3, 1, 1), bias=False, padding=(1, 0, 0))
+            self.bn1 = nn.BatchNorm3d(planes)
+        else:
+            raise ValueError("Unsupported head_conv!")
+        self.conv2 = nn.Conv3d(
+            planes, planes, kernel_size=(1, 3, 3), stride=(1,stride,stride), padding=(0, 1, 1), bias=False)
+        self.bn2 = nn.BatchNorm3d(planes)
+        self.conv3 = nn.Conv3d(planes, planes * 4, kernel_size=1, bias=False)
+        self.bn3 = nn.BatchNorm3d(planes * 4)
+        self.relu = nn.ReLU(inplace=True)
+        self.downsample = downsample
+        self.stride = stride
+
+    def forward(self, x):
+        residual = x
+
+        out = self.conv1(x)
+        out = self.bn1(out)
+        out = self.relu(out)
+
+        out = self.conv2(out)
+        out = self.bn2(out)
+        out = self.relu(out)
+
+        out = self.conv3(out)
+        out = self.bn3(out)
+
+        if self.downsample is not None:
+            residual = self.downsample(x)
+        out += residual
+        out = self.relu(out)
+
+        return out
+
+
+class SlowFast(nn.Module):
+    def __init__(self, block=Bottleneck, layers=[3, 4, 6, 3], class_num=10, dropout=0.5 ):
+        super(SlowFast, self).__init__()
+
+        self.fast_inplanes = 8
+        self.fast_conv1 = nn.Conv3d(3, 8, kernel_size=(5, 7, 7), stride=(1, 2, 2), padding=(2, 3, 3), bias=False)
+        self.fast_bn1 = nn.BatchNorm3d(8)
+        self.fast_relu = nn.ReLU(inplace=True)
+        self.fast_maxpool = nn.MaxPool3d(kernel_size=(1, 3, 3), stride=(1, 2, 2), padding=(0, 1, 1))
+        self.fast_res2 = self._make_layer_fast(block, 8, layers[0], head_conv=3)
+        self.fast_res3 = self._make_layer_fast(
+            block, 16, layers[1], stride=2, head_conv=3)
+        self.fast_res4 = self._make_layer_fast(
+            block, 32, layers[2], stride=2, head_conv=3)
+        self.fast_res5 = self._make_layer_fast(
+            block, 64, layers[3], stride=2, head_conv=3)
+        
+        self.lateral_p1 = nn.Conv3d(8, 8*2, kernel_size=(5, 1, 1), stride=(8, 1 ,1), bias=False, padding=(2, 0, 0))
+        self.lateral_res2 = nn.Conv3d(32,32*2, kernel_size=(5, 1, 1), stride=(8, 1 ,1), bias=False, padding=(2, 0, 0))
+        self.lateral_res3 = nn.Conv3d(64,64*2, kernel_size=(5, 1, 1), stride=(8, 1 ,1), bias=False, padding=(2, 0, 0))
+        self.lateral_res4 = nn.Conv3d(128,128*2, kernel_size=(5, 1, 1), stride=(8, 1 ,1), bias=False, padding=(2, 0, 0))
+
+        self.slow_inplanes = 64+64//8*2
+        self.slow_conv1 = nn.Conv3d(3, 64, kernel_size=(1, 7, 7), stride=(1, 2, 2), padding=(0, 3, 3), bias=False)
+        self.slow_bn1 = nn.BatchNorm3d(64)
+        self.slow_relu = nn.ReLU(inplace=True)
+        self.slow_maxpool = nn.MaxPool3d(kernel_size=(1, 3, 3), stride=(1, 2, 2), padding=(0, 1, 1))
+        self.slow_res2 = self._make_layer_slow(block, 64, layers[0], head_conv=1)
+        self.slow_res3 = self._make_layer_slow(
+            block, 128, layers[1], stride=2, head_conv=1)
+        self.slow_res4 = self._make_layer_slow(
+            block, 256, layers[2], stride=2, head_conv=3)
+        self.slow_res5 = self._make_layer_slow(
+            block, 512, layers[3], stride=2, head_conv=3)
+        self.dp = nn.Dropout(dropout)
+        self.fc = nn.Linear(self.fast_inplanes+2048, class_num, bias=False)
+    def forward(self, input):
+        fast, lateral = self.FastPath(input[:, :, ::2, :, :])
+        slow = self.SlowPath(input[:, :, ::16, :, :], lateral)
+        x = torch.cat([slow, fast], dim=1)
+        x = self.dp(x)
+        x = self.fc(x)
+        return x
+
+
+
+    def SlowPath(self, input, lateral):
+        x = self.slow_conv1(input)
+        x = self.slow_bn1(x)
+        x = self.slow_relu(x)
+        x = self.slow_maxpool(x)
+        x = torch.cat([x, lateral[0]],dim=1)
+        x = self.slow_res2(x)
+        x = torch.cat([x, lateral[1]],dim=1)
+        x = self.slow_res3(x)
+        x = torch.cat([x, lateral[2]],dim=1)
+        x = self.slow_res4(x)
+        x = torch.cat([x, lateral[3]],dim=1)
+        x = self.slow_res5(x)
+        x = nn.AdaptiveAvgPool3d(1)(x)
+        x = x.view(-1, x.size(1))
+        return x
+
+    def FastPath(self, input):
+        lateral = []
+        x = self.fast_conv1(input)
+        x = self.fast_bn1(x)
+        x = self.fast_relu(x)
+        pool1 = self.fast_maxpool(x)
+        lateral_p = self.lateral_p1(pool1)
+        lateral.append(lateral_p)
+
+        res2 = self.fast_res2(pool1)
+        lateral_res2 = self.lateral_res2(res2)
+        lateral.append(lateral_res2)
+        
+        res3 = self.fast_res3(res2)
+        lateral_res3 = self.lateral_res3(res3)
+        lateral.append(lateral_res3)
+
+        res4 = self.fast_res4(res3)
+        lateral_res4 = self.lateral_res4(res4)
+        lateral.append(lateral_res4)
+
+        res5 = self.fast_res5(res4)
+        x = nn.AdaptiveAvgPool3d(1)(res5)
+        x = x.view(-1, x.size(1))
+
+        return x, lateral
+
+    def _make_layer_fast(self, block, planes, blocks, stride=1, head_conv=1):
+        downsample = None
+        if stride != 1 or self.fast_inplanes != planes * block.expansion:
+            downsample = nn.Sequential(
+                nn.Conv3d(
+                    self.fast_inplanes,
+                    planes * block.expansion,
+                    kernel_size=1,
+                    stride=(1,stride,stride),
+                    bias=False), nn.BatchNorm3d(planes * block.expansion))
+
+        layers = []
+        layers.append(block(self.fast_inplanes, planes, stride, downsample, head_conv=head_conv))
+        self.fast_inplanes = planes * block.expansion
+        for i in range(1, blocks):
+            layers.append(block(self.fast_inplanes, planes, head_conv=head_conv))
+        return nn.Sequential(*layers)
+
+    def _make_layer_slow(self, block, planes, blocks, stride=1, head_conv=1):
+        downsample = None
+        if stride != 1 or self.slow_inplanes != planes * block.expansion:
+            downsample = nn.Sequential(
+                nn.Conv3d(
+                    self.slow_inplanes,
+                    planes * block.expansion,
+                    kernel_size=1,
+                    stride=(1,stride,stride),
+                    bias=False), nn.BatchNorm3d(planes * block.expansion))
+
+        layers = []
+        layers.append(block(self.slow_inplanes, planes, stride, downsample, head_conv=head_conv))
+        self.slow_inplanes = planes * block.expansion
+        for i in range(1, blocks):
+            layers.append(block(self.slow_inplanes, planes, head_conv=head_conv))
+  
+        self.slow_inplanes = planes * block.expansion + planes * block.expansion//8*2
+        return nn.Sequential(*layers)
+
+
+
+
+def resnet50(**kwargs):
+    """Constructs a ResNet-50 model.
+    """
+    model = SlowFast(Bottleneck, [3, 4, 6, 3], **kwargs)
+    return model
+
+
+def resnet101(**kwargs):
+    """Constructs a ResNet-101 model.
+    """
+    model = SlowFast(Bottleneck, [3, 4, 23, 3], **kwargs)
+    return model
+
+
+def resnet152(**kwargs):
+    """Constructs a ResNet-101 model.
+    """
+    model = SlowFast(Bottleneck, [3, 8, 36, 3], **kwargs)
+    return model
+
+
+def resnet200(**kwargs):
+    """Constructs a ResNet-101 model.
+    """
+    model = SlowFast(Bottleneck, [3, 24, 36, 3], **kwargs)
+    return model
+
+if __name__ == "__main__":
+    num_classes = 101
+    input_tensor = torch.autograd.Variable(torch.rand(1, 3, 64, 224, 224))
+    model = resnet50(class_num=num_classes)
+    output = model(input_tensor)
+    print(output.size())

{pybaseutils-2.3.5 → pybaseutils-2.3.6}/test_py/web_socket/asr_service.py

@@ -64,7 +64,7 @@ class MainSocketHandler(tornado.websocket.WebSocketHandler):
 
     def on_message(self, data):
         self.time = time.time()  # 更新最后活跃时间
-        print(f"收到来自{self.cid}的消息")
+        print(f"收到来自{self.cid}的消息,data={data[:10]}")
         # 使用FunASR进行流式识别
         result = model.generate(input=data, cache=cache, is_final=False, chunk_size=chunk_size,
                                 encoder_chunk_look_back=encoder_chunk_look_back,

pybaseutils-2.3.5/test_py/demo3.py

@@ -1,41 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-# --------------------------------------------------------
-# @Author : Pan
-# @E-mail : 
-# @Date   : 2025-07-08 14:10:15
-# @Brief  :
-# --------------------------------------------------------
-"""
-import cv2
-import numpy as np
-from pybaseutils import file_utils, image_utils
-
-import os
-from pybaseutils.dataloader import parser_coco_kps
-
-import os
-from pybaseutils.dataloader import parser_voc
-
-if __name__ == "__main__":
-    # 修改为自己数据集的路径
-    data_root = "/home/PKing/nasdata/tmp/tmp/pen/笔尖指尖标注方法"
-    class_name = []
-    dataset = parser_voc.VOCDataset(filename=None,
-                                    data_root=data_root,
-                                    anno_dir=None,
-                                    image_dir=None,
-                                    class_name=class_name,
-                                    transform=None,
-                                    use_rgb=False,
-                                    check=False,
-                                    shuffle=False)
-    print("have num:{}".format(len(dataset)))
-    class_name = dataset.class_name
-    for i in range(len(dataset)):
-        data = dataset.__getitem__(i)
-        image, targets, image_id = data["image"], data["target"], data["image_id"]
-        print(image_id)
-        bboxes, labels = targets[:, 0:4], targets[:, 4:5]
-        parser_voc.show_target_image(image, bboxes, labels, normal=False, transpose=False,
-                                     class_name=class_name, use_rgb=False, thickness=3, fontScale=1.2)