当前位置: 首页 > news >正文

Mac平台M1PRO芯片MiniCPM-V-2.6网页部署跑通

Mac平台M1PRO芯片MiniCPM-V-2.6网页部署跑通

契机

2.6的小钢炮可以输入视频了,我必须拉到本地跑跑。主要解决2.6版本默认绑定flash_atten问题,pip install flash_attn也无法安装,因为强制依赖cuda。主要解决的就是这个问题,还有 BFloat16 is not supported on MPS问题解决。

环境

  • macos版本:版本15.0 Beta版(24A5279h) || 版本15.1 Beta版(24B5009l)
  • 芯片:m1 pro
  • 代码仓库:https://github.com/OpenBMB/MiniCPM-V.git
  • 分支:main
  • 代码版本:b0125d8a yiranyyu 2606375857@qq.com on 2024/8/9 at 10:25
  • python版本:3.9

解决问题

#拉下这个仓库
git clone [https://github.com/OpenBMB/MiniCPM-V.git](https://github.com/OpenBMB/MiniCPM-V.git) #把requirements.txt安装下
#modelscope需要手动安装
pip install http://thunlp.oss-cn-qingdao.aliyuncs.com/multi_modal/never_delete/modelscope_studio-0.4.0.9-py3-none-any.whl
#dcord如果安装有问题,参考我LAVIS博客#找到根目录web_demo_2.6.py运行
#首先添加环境变量,mps参数,见下图
--device mps
PYTORCH_ENABLE_MPS_FALLBACK=1

请添加图片描述


#第一次运行web_demo_2.6.py报错如下
ImportError: This modeling file requires the following packages that were not found in your environment: flash_attn. Run `pip install flash_attn`#直接修改代码
from typing import Union
from transformers.dynamic_module_utils import get_imports
from unittest.mock import patch
# fix the imports
def fixed_get_imports(filename: Union[str, os.PathLike]) -> list[str]:imports = get_imports(filename)if not torch.cuda.is_available() and "flash_attn" in imports:imports.remove("flash_attn")return imports#79行左右修改为
with patch("transformers.dynamic_module_utils.get_imports", fixed_get_imports):model = AutoModel.from_pretrained(model_path, trust_remote_code=True, torch_dtype=torch.bfloat16)model = model.to(device=device)

完整代码如下

#!/usr/bin/env python
# encoding: utf-8
import torch
import argparse
from transformers import AutoModel, AutoTokenizer
import gradio as gr
from PIL import Image
from decord import VideoReader, cpu
import io
import os
import copy
import requests
import base64
import json
import traceback
import re
import modelscope_studio as mgr
from typing import Union
from transformers.dynamic_module_utils import get_imports
from unittest.mock import patch# README, How to run demo on different devices# For Nvidia GPUs.
# python web_demo_2.6.py --device cuda# For Mac with MPS (Apple silicon or AMD GPUs).
# PYTORCH_ENABLE_MPS_FALLBACK=1 python web_demo_2.6.py --device mps# Argparser
parser = argparse.ArgumentParser(description='demo')
parser.add_argument('--device', type=str, default='cuda', help='cuda or mps')
parser.add_argument('--multi-gpus', action='store_true', default=False, help='use multi-gpus')
args = parser.parse_args()
device = args.device
assert device in ['cuda', 'mps']# fix the imports
def fixed_get_imports(filename: Union[str, os.PathLike]) -> list[str]:imports = get_imports(filename)if not torch.cuda.is_available() and "flash_attn" in imports:imports.remove("flash_attn")return imports# Load model
model_path = 'openbmb/MiniCPM-V-2_6'
if 'int4' in model_path:if device == 'mps':print('Error: running int4 model with bitsandbytes on Mac is not supported right now.')exit()model = AutoModel.from_pretrained(model_path, trust_remote_code=True)
else:if args.multi_gpus:from accelerate import load_checkpoint_and_dispatch, init_empty_weights, infer_auto_device_mapwith init_empty_weights():model = AutoModel.from_pretrained(model_path, trust_remote_code=True, attn_implementation='sdpa', torch_dtype=torch.bfloat16)device_map = infer_auto_device_map(model, max_memory={0: "10GB", 1: "10GB"},no_split_module_classes=['SiglipVisionTransformer', 'Qwen2DecoderLayer'])device_id = device_map["llm.model.embed_tokens"]device_map["llm.lm_head"] = device_id # firtt and last layer should be in same devicedevice_map["vpm"] = device_iddevice_map["resampler"] = device_iddevice_id2 = device_map["llm.model.layers.26"]device_map["llm.model.layers.8"] = device_id2device_map["llm.model.layers.9"] = device_id2device_map["llm.model.layers.10"] = device_id2device_map["llm.model.layers.11"] = device_id2device_map["llm.model.layers.12"] = device_id2device_map["llm.model.layers.13"] = device_id2device_map["llm.model.layers.14"] = device_id2device_map["llm.model.layers.15"] = device_id2device_map["llm.model.layers.16"] = device_id2#print(device_map)model = load_checkpoint_and_dispatch(model, model_path, dtype=torch.bfloat16, device_map=device_map)else:with patch("transformers.dynamic_module_utils.get_imports", fixed_get_imports):model = AutoModel.from_pretrained(model_path, trust_remote_code=True, torch_dtype=torch.bfloat16)model = model.to(device=device)
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
model.eval()ERROR_MSG = "Error, please retry"
model_name = 'MiniCPM-V 2.6'
MAX_NUM_FRAMES = 64
IMAGE_EXTENSIONS = {'.jpg', '.jpeg', '.png', '.bmp', '.tiff', '.webp'}
VIDEO_EXTENSIONS = {'.mp4', '.mkv', '.mov', '.avi', '.flv', '.wmv', '.webm', '.m4v'}def get_file_extension(filename):return os.path.splitext(filename)[1].lower()def is_image(filename):return get_file_extension(filename) in IMAGE_EXTENSIONSdef is_video(filename):return get_file_extension(filename) in VIDEO_EXTENSIONSform_radio = {'choices': ['Beam Search', 'Sampling'],#'value': 'Beam Search','value': 'Sampling','interactive': True,'label': 'Decode Type'
}def create_component(params, comp='Slider'):if comp == 'Slider':return gr.Slider(minimum=params['minimum'],maximum=params['maximum'],value=params['value'],step=params['step'],interactive=params['interactive'],label=params['label'])elif comp == 'Radio':return gr.Radio(choices=params['choices'],value=params['value'],interactive=params['interactive'],label=params['label'])elif comp == 'Button':return gr.Button(value=params['value'],interactive=True)def create_multimodal_input(upload_image_disabled=False, upload_video_disabled=False):return mgr.MultimodalInput(upload_image_button_props={'label': 'Upload Image', 'disabled': upload_image_disabled, 'file_count': 'multiple'},upload_video_button_props={'label': 'Upload Video', 'disabled': upload_video_disabled, 'file_count': 'single'},submit_button_props={'label': 'Submit'})def chat(img, msgs, ctx, params=None, vision_hidden_states=None):try:print('msgs:', msgs)answer = model.chat(image=None,msgs=msgs,tokenizer=tokenizer,**params)res = re.sub(r'(<box>.*</box>)', '', answer)res = res.replace('<ref>', '')res = res.replace('</ref>', '')res = res.replace('<box>', '')answer = res.replace('</box>', '')print('answer:', answer)return 0, answer, None, Noneexcept Exception as e:print(e)traceback.print_exc()return -1, ERROR_MSG, None, Nonedef encode_image(image):if not isinstance(image, Image.Image):if hasattr(image, 'path'):image = Image.open(image.path).convert("RGB")else:image = Image.open(image.file.path).convert("RGB")# resize to max_sizemax_size = 448*16if max(image.size) > max_size:w,h = image.sizeif w > h:new_w = max_sizenew_h = int(h * max_size / w)else:new_h = max_sizenew_w = int(w * max_size / h)image = image.resize((new_w, new_h), resample=Image.BICUBIC)return image## save by BytesIO and convert to base64#buffered = io.BytesIO()#image.save(buffered, format="png")#im_b64 = base64.b64encode(buffered.getvalue()).decode()#return {"type": "image", "pairs": im_b64}def encode_video(video):def uniform_sample(l, n):gap = len(l) / nidxs = [int(i * gap + gap / 2) for i in range(n)]return [l[i] for i in idxs]if hasattr(video, 'path'):vr = VideoReader(video.path, ctx=cpu(0))else:vr = VideoReader(video.file.path, ctx=cpu(0))sample_fps = round(vr.get_avg_fps() / 1)  # FPSframe_idx = [i for i in range(0, len(vr), sample_fps)]if len(frame_idx)>MAX_NUM_FRAMES:frame_idx = uniform_sample(frame_idx, MAX_NUM_FRAMES)video = vr.get_batch(frame_idx).asnumpy()video = [Image.fromarray(v.astype('uint8')) for v in video]video = [encode_image(v) for v in video]print('video frames:', len(video))return videodef check_mm_type(mm_file):if hasattr(mm_file, 'path'):path = mm_file.pathelse:path = mm_file.file.pathif is_image(path):return "image"if is_video(path):return "video"return Nonedef encode_mm_file(mm_file):if check_mm_type(mm_file) == 'image':return [encode_image(mm_file)]if check_mm_type(mm_file) == 'video':return encode_video(mm_file)return Nonedef make_text(text):#return {"type": "text", "pairs": text} # # For remote callreturn textdef encode_message(_question):files = _question.filesquestion = _question.textpattern = r"\[mm_media\]\d+\[/mm_media\]"matches = re.split(pattern, question)message = []if len(matches) != len(files) + 1:gr.Warning("Number of Images not match the placeholder in text, please refresh the page to restart!")assert len(matches) == len(files) + 1text = matches[0].strip()if text:message.append(make_text(text))for i in range(len(files)):message += encode_mm_file(files[i])text = matches[i + 1].strip()if text:message.append(make_text(text))return messagedef check_has_videos(_question):images_cnt = 0videos_cnt = 0for file in _question.files:if check_mm_type(file) == "image":images_cnt += 1else:videos_cnt += 1return images_cnt, videos_cntdef count_video_frames(_context):num_frames = 0for message in _context:for item in message["content"]:#if item["type"] == "image": # For remote callif isinstance(item, Image.Image):num_frames += 1return num_framesdef respond(_question, _chat_bot, _app_cfg, params_form):_context = _app_cfg['ctx'].copy()_context.append({'role': 'user', 'content': encode_message(_question)})images_cnt = _app_cfg['images_cnt']videos_cnt = _app_cfg['videos_cnt']files_cnts = check_has_videos(_question)if files_cnts[1] + videos_cnt > 1 or (files_cnts[1] + videos_cnt == 1 and files_cnts[0] + images_cnt > 0):gr.Warning("Only supports single video file input right now!")return _question, _chat_bot, _app_cfgif params_form == 'Beam Search':params = {'sampling': False,'num_beams': 3,'repetition_penalty': 1.2,"max_new_tokens": 2048}else:params = {'sampling': True,'top_p': 0.8,'top_k': 100,'temperature': 0.7,'repetition_penalty': 1.05,"max_new_tokens": 2048}if files_cnts[1] + videos_cnt > 0:params["max_inp_length"] = 4352 # 4096+256params["use_image_id"] = Falseparams["max_slice_nums"] = 1 if count_video_frames(_context) > 16 else 2code, _answer, _, sts = chat("", _context, None, params)images_cnt += files_cnts[0]videos_cnt += files_cnts[1]_context.append({"role": "assistant", "content": [make_text(_answer)]})_chat_bot.append((_question, _answer))if code == 0:_app_cfg['ctx']=_context_app_cfg['sts']=sts_app_cfg['images_cnt'] = images_cnt_app_cfg['videos_cnt'] = videos_cntupload_image_disabled = videos_cnt > 0upload_video_disabled = videos_cnt > 0 or images_cnt > 0return create_multimodal_input(upload_image_disabled, upload_video_disabled), _chat_bot, _app_cfgdef fewshot_add_demonstration(_image, _user_message, _assistant_message, _chat_bot, _app_cfg):ctx = _app_cfg["ctx"]message_item = []if _image is not None:image = Image.open(_image).convert("RGB")ctx.append({"role": "user", "content": [encode_image(image), make_text(_user_message)]})message_item.append({"text": "[mm_media]1[/mm_media]" + _user_message, "files": [_image]})else:if _user_message:ctx.append({"role": "user", "content": [make_text(_user_message)]})message_item.append({"text": _user_message, "files": []})else:message_item.append(None)if _assistant_message:ctx.append({"role": "assistant", "content": [make_text(_assistant_message)]})message_item.append({"text": _assistant_message, "files": []})else:message_item.append(None)_chat_bot.append(message_item)return None, "", "", _chat_bot, _app_cfgdef fewshot_respond(_image, _user_message, _chat_bot, _app_cfg, params_form):user_message_contents = []_context = _app_cfg["ctx"].copy()if _image:image = Image.open(_image).convert("RGB")user_message_contents += [encode_image(image)]if _user_message:user_message_contents += [make_text(_user_message)]if user_message_contents:_context.append({"role": "user", "content": user_message_contents})if params_form == 'Beam Search':params = {'sampling': False,'num_beams': 3,'repetition_penalty': 1.2,"max_new_tokens": 2048}else:params = {'sampling': True,'top_p': 0.8,'top_k': 100,'temperature': 0.7,'repetition_penalty': 1.05,"max_new_tokens": 2048}code, _answer, _, sts = chat("", _context, None, params)_context.append({"role": "assistant", "content": [make_text(_answer)]})if _image:_chat_bot.append([{"text": "[mm_media]1[/mm_media]" + _user_message, "files": [_image]},{"text": _answer, "files": []}])else:_chat_bot.append([{"text": _user_message, "files": [_image]},{"text": _answer, "files": []}])if code == 0:_app_cfg['ctx']=_context_app_cfg['sts']=stsreturn None, '', '', _chat_bot, _app_cfgdef regenerate_button_clicked(_question, _image, _user_message, _assistant_message, _chat_bot, _app_cfg, params_form):if len(_chat_bot) <= 1 or not _chat_bot[-1][1]:gr.Warning('No question for regeneration.')return '', _image, _user_message, _assistant_message, _chat_bot, _app_cfgif _app_cfg["chat_type"] == "Chat":images_cnt = _app_cfg['images_cnt']videos_cnt = _app_cfg['videos_cnt']_question = _chat_bot[-1][0]_chat_bot = _chat_bot[:-1]_app_cfg['ctx'] = _app_cfg['ctx'][:-2]files_cnts = check_has_videos(_question)images_cnt -= files_cnts[0]videos_cnt -= files_cnts[1]_app_cfg['images_cnt'] = images_cnt_app_cfg['videos_cnt'] = videos_cntupload_image_disabled = videos_cnt > 0upload_video_disabled = videos_cnt > 0 or images_cnt > 0_question, _chat_bot, _app_cfg = respond(_question, _chat_bot, _app_cfg, params_form)return _question, _image, _user_message, _assistant_message, _chat_bot, _app_cfgelse:last_message = _chat_bot[-1][0]last_image = Nonelast_user_message = ''if last_message.text:last_user_message = last_message.textif last_message.files:last_image = last_message.files[0].file.path_chat_bot = _chat_bot[:-1]_app_cfg['ctx'] = _app_cfg['ctx'][:-2]_image, _user_message, _assistant_message, _chat_bot, _app_cfg = fewshot_respond(last_image, last_user_message, _chat_bot, _app_cfg, params_form)return _question, _image, _user_message, _assistant_message, _chat_bot, _app_cfgdef flushed():return gr.update(interactive=True)def clear(txt_message, chat_bot, app_session):txt_message.files.clear()txt_message.text = ''chat_bot = copy.deepcopy(init_conversation)app_session['sts'] = Noneapp_session['ctx'] = []app_session['images_cnt'] = 0app_session['videos_cnt'] = 0return create_multimodal_input(), chat_bot, app_session, None, '', ''def select_chat_type(_tab, _app_cfg):_app_cfg["chat_type"] = _tabreturn _app_cfginit_conversation = [[None,{# The first message of bot closes the typewriter."text": "You can talk to me now","flushing": False}],
]css = """
video { height: auto !important; }
.example label { font-size: 16px;}
"""introduction = """## Features:
1. Chat with single image
2. Chat with multiple images
3. Chat with video
4. In-context few-shot learningClick `How to use` tab to see examples.
"""with gr.Blocks(css=css) as demo:with gr.Tab(model_name):with gr.Row():with gr.Column(scale=1, min_width=300):gr.Markdown(value=introduction)params_form = create_component(form_radio, comp='Radio')regenerate = create_component({'value': 'Regenerate'}, comp='Button')clear_button = create_component({'value': 'Clear History'}, comp='Button')with gr.Column(scale=3, min_width=500):app_session = gr.State({'sts':None,'ctx':[], 'images_cnt': 0, 'videos_cnt': 0, 'chat_type': 'Chat'})chat_bot = mgr.Chatbot(label=f"Chat with {model_name}", value=copy.deepcopy(init_conversation), height=600, flushing=False, bubble_full_width=False)with gr.Tab("Chat") as chat_tab:txt_message = create_multimodal_input()chat_tab_label = gr.Textbox(value="Chat", interactive=False, visible=False)txt_message.submit(respond,[txt_message, chat_bot, app_session, params_form],[txt_message, chat_bot, app_session])with gr.Tab("Few Shot") as fewshot_tab:fewshot_tab_label = gr.Textbox(value="Few Shot", interactive=False, visible=False)with gr.Row():with gr.Column(scale=1):image_input = gr.Image(type="filepath", sources=["upload"])with gr.Column(scale=3):user_message = gr.Textbox(label="User")assistant_message = gr.Textbox(label="Assistant")with gr.Row():add_demonstration_button = gr.Button("Add Example")generate_button = gr.Button(value="Generate", variant="primary")add_demonstration_button.click(fewshot_add_demonstration,[image_input, user_message, assistant_message, chat_bot, app_session],[image_input, user_message, assistant_message, chat_bot, app_session])generate_button.click(fewshot_respond,[image_input, user_message, chat_bot, app_session, params_form],[image_input, user_message, assistant_message, chat_bot, app_session])chat_tab.select(select_chat_type,[chat_tab_label, app_session],[app_session])chat_tab.select( # do clearclear,[txt_message, chat_bot, app_session],[txt_message, chat_bot, app_session, image_input, user_message, assistant_message])fewshot_tab.select(select_chat_type,[fewshot_tab_label, app_session],[app_session])fewshot_tab.select( # do clearclear,[txt_message, chat_bot, app_session],[txt_message, chat_bot, app_session, image_input, user_message, assistant_message])chat_bot.flushed(flushed,outputs=[txt_message])regenerate.click(regenerate_button_clicked,[txt_message, image_input, user_message, assistant_message, chat_bot, app_session, params_form],[txt_message, image_input, user_message, assistant_message, chat_bot, app_session])clear_button.click(clear,[txt_message, chat_bot, app_session],[txt_message, chat_bot, app_session, image_input, user_message, assistant_message])with gr.Tab("How to use"):with gr.Column():with gr.Row():image_example = gr.Image(value="http://thunlp.oss-cn-qingdao.aliyuncs.com/multi_modal/never_delete/m_bear2.gif", label='1. Chat with single or multiple images', interactive=False, width=400, elem_classes="example")example2 = gr.Image(value="http://thunlp.oss-cn-qingdao.aliyuncs.com/multi_modal/never_delete/video2.gif", label='2. Chat with video', interactive=False, width=400, elem_classes="example")example3 = gr.Image(value="http://thunlp.oss-cn-qingdao.aliyuncs.com/multi_modal/never_delete/fshot.gif", label='3. Few shot', interactive=False, width=400, elem_classes="example")# launch
demo.launch(share=False, debug=True, show_api=False, server_port=8885, server_name="0.0.0.0")
#第一次运行web_demo_2.6.py报错如下
File "/Usxxxxxxxckages/torch/nn/modules/module.py", line 1158, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
TypeError: BFloat16 is not supported on MPS#重装依赖
pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cpu#再次运行就没问题了
#这里下载模型20g可能会等一段时间,最后借助魔法下载,看这网速在疯狂跑就没问题
#成功运行输出如下
Loading checkpoint shards: 100%|██████████| 4/4 [00:21<00:00,  5.33s/it]
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Running on local URL:  http://0.0.0.0:8885To create a public link, set `share=True` in `launch()`.
IMPORTANT: You are using gradio version 4.22.0, however version 4.29.0 is available, please upgrade.
--------

效果展示

图片理解

Sampling解码

请添加图片描述

Beam Search解码

请添加图片描述

视频理解

Sampling解码

请添加图片描述

Beam Search解码

请添加图片描述

系统占用

请添加图片描述

总结

  • 解决flash_attn强制依赖问题
  • 解决bfloat16在mps无法使用问题
  • 看系统占用是没走mps,添加的环境变量也可以看出
  • Sampling瞎回答,Beam Search回答很惊喜
  • Beam Search处理视频4秒,在m1pro下,当前代码中需要230s左右
  • ollama部署还在研究中…

写到最后

请添加图片描述

相关文章:

Mac平台M1PRO芯片MiniCPM-V-2.6网页部署跑通

Mac平台M1PRO芯片MiniCPM-V-2.6网页部署跑通 契机 ⚙ 2.6的小钢炮可以输入视频了&#xff0c;我必须拉到本地跑跑。主要解决2.6版本默认绑定flash_atten问题&#xff0c;pip install flash_attn也无法安装&#xff0c;因为强制依赖cuda。主要解决的就是这个问题&#xff0c;还…...

MyBatis:Maven,Git,TortoiseGit,Gradle

1&#xff0c;Maven Maven是一个非常优秀的项目管理工具&#xff0c;采用一种“约定优于配置&#xff08;CoC&#xff09;”的策略来管理项目。使用Maven不仅可以把源代码构建成可发布的项目&#xff08;包括编译、打包、测试和分发&#xff09;&#xff0c;还可以生成报告、生…...

获取链表中间位置的两种方法方法

方法一&#xff1a; 我们可以计算链表节点的数量&#xff0c;然后遍历链表找到前半部分的尾节点。 方法二: 我们也可以使用快慢指针在一次遍历中找到&#xff1a;慢指针一次走一步&#xff0c;快指针一次走两步&#xff0c;快慢指针同时出发。当快指针移动到链表的末尾时&am…...

第二十天的学习(2024.8.8)Vue拓展

昨天的笔记中&#xff0c;我们进行的项目已经可以在网页上显示查询到数据库中的数据&#xff0c;今天的笔记中将会完成在网页上进行增删改查的操作 1.删除表中数据 现在网页上只能呈现出数据库中的数据&#xff0c;我们首先添加一个删除按钮&#xff0c;使其可以对数据库数据…...

微信小程序教程011:全局配置:Window

文章目录 1、window1.1、`window`-小程序窗口的组成部分1.2、了解 window 节点常用的配置项1.3、设置导航栏的标题1.4、设置导航栏的背景色1.5、设置导航栏的标题颜色1.6、全局开启下拉刷新功能1.7、设置下拉刷新时窗口的背景色1.8、设置下拉刷新时 loading 的样式1.9、设置上拉…...

Tomcat服务器和Web项目的部署

目录 一、概述和作用 二、安装 1.进入官网 2.Download下面选择想要下载的版本 3.点击Which version查看版本所需要的JRE版本 4.返回上一页下载和电脑和操作系统匹配的Tomcat 5. 安装完成后&#xff0c;点击bin目录下的startup.bat&#xff08;linux系统下就运行startup.sh&…...

PCIe学习笔记(22)

Transaction Ordering Transaction Ordering Rules 表2-40定义了PCI Express Transactions的排序要求。该表中定义的规则统一适用于PCI Express上所有类型的事务&#xff0c;包括内存、I/O、配置和消息。该表中定义的排序规则适用于单个流量类(TC)。不同TC标签的事务之间没有…...

Vue3 依赖注入Provide / Inject

在实际开发中&#xff0c;我们经常需要从父组件向子组件传递数据&#xff0c;一般情况下&#xff0c;我们使用 props。但有时候会遇到深度嵌套的组件&#xff0c;而深层的子组件只需要父组件的部分内容。在这种情况下&#xff0c;如果仍然将 prop 沿着组件链逐级传递下去&#…...

Python | Leetcode Python题解之第332题重新安排行程

题目&#xff1a; 题解&#xff1a; class Solution:def findItinerary(self, tickets: List[List[str]]) -> List[str]:def dfs(curr: str):while vec[curr]:tmp heapq.heappop(vec[curr])dfs(tmp)stack.append(curr)vec collections.defaultdict(list)for depart, arri…...

React状态管理:react-redux和redux-saga(适合由vue转到react的同学)

注意&#xff1a;本文不会把所有知识点都写一遍&#xff0c;并不适合纯新手阅读 首先Redux是一种状态管理方案&#xff0c;本身和react并没有什么联系&#xff0c;redux也可以结合其他框架来用。 react-redux是基于react的一种状态管理实现&#xff0c;他不像vuex那样直接内置在…...

刷题技巧:双指针法的核心思想总结+例题整合+力扣接雨水双指针c++实现

双指针法的核心思想是通过同时操作两个指针来遍历数据结构&#xff0c;通常是数组或链表&#xff0c;以达到优化算法性能的目的。具体来说&#xff0c;双指针法能够减少时间复杂度、空间复杂度&#xff0c;或者简化逻辑结构。以下是双指针法的几个核心思想&#xff1a; ps 下面…...

什么是前端微服务,有何优势

随着互联网技术的发展&#xff0c;传统的单体应用架构已经无法满足复杂业务场景的需求。微服务架构的兴起为后端应用的开发和部署提供了灵活性和可扩展性。与此同时&#xff0c;前端开发也经历了类似的演变&#xff0c;前端微服务作为一种新兴的架构模式应运而生。 一、前端微服…...

小论文写作——02:编故事

一篇论文&#xff0c;可以发水刊&#xff0c;也可以发顶刊顶会&#xff0c;这两者的区别就是一个故事编的好不好。 你的论文ABC&#xff0c;但不能之说有ABC。创新就是看你故事编的怎么样&#xff1f;创新是编出来的。 我们要说&#xff1a;我发现了问题&#xff0c;然后准备…...

GIT企业开发使用介绍

0.认识git git就是一个版本控制器&#xff0c;记录每次的修改以及版本迭代的一个管理系统 至于为什么会有git的出现&#xff0c;主要是为了解决一份代码改了又改&#xff0c;但最后还是要第一版的情况 git 可以控制电脑上所有格式的文档 1.安装git sudo yum install git -y…...

文件上传-前端验证

查看源代码&#xff08;找验证代码&#xff09; 1、源代码直接找到验证代码 示例&#xff1a; function checkFileExt(filename){var flag false; //状态var arr ["jpg","png","gif"]; //允许上传的文件//取出上传文件的扩展名var index f…...

ROT加密算法login-RESERVE

ROT算法(字母轮换加密) 也称为Caesar加密&#xff0c;是一种简单的字母替换加密算法。它通过将字母表中的每个字母向后&#xff08;或向前&#xff09;移动固定的位置来加密文本。 加密步骤&#xff1a; 选择一个固定的偏移量&#xff08;通常是1到25之间的整数&#xff09;&…...

C++ 新特性 | C++20 常用新特性介绍

目录 1、模块(Modules) 2、协程(Coroutines) 3、概念(Concepts) 4、范围(Ranges) 5、三向比较符&#xff08;three-way comparison&#xff09; C软件异常排查从入门到精通系列教程&#xff08;专栏文章列表&#xff0c;欢迎订阅&#xff0c;持续更新...&#xff09;https…...

Java设计模式之策略模式实践

1、策略接口 /*** 策略接口*/ public interface DemoStrategy {Result execute(); } 2、策略工厂 /*** 策略工厂*/ Component public class DemoFactory {Resourceprivate final Map<String, DemoStrategy> demoStrategy new ConcurrentHashMap<>();public Demo…...

C语言——结构体数组、结构体指针、结构体函数与二级指针

C语言中的结构体&#xff08;struct&#xff09;是一种用户自定义的数据类型&#xff0c;它允许你将不同类型的数据项组合成一个单一的类型。结构体数组则是一种特殊的数组&#xff0c;其元素为结构体类型。这意味着你可以在一个数组中存储多个具有相同结构的记录。 定义结构体…...

【4】策略模式

如上图所示&#xff0c;如果要加入一个新的货币&#xff0c;那么就需要对类中的Calculate函数进行修改&#xff0c;这违背了封闭开放原则。 上图中的方式更加合适&#xff0c;搞一个抽象类&#xff08;方法中可以用多态调用&#xff09;&#xff0c;然后每个货币自己是一个类&a…...

【算法训练营Day07】字符串part1

文章目录 反转字符串反转字符串II替换数字 反转字符串 题目链接&#xff1a;344. 反转字符串 双指针法&#xff0c;两个指针的元素直接调转即可 class Solution {public void reverseString(char[] s) {int head 0;int end s.length - 1;while(head < end) {char temp …...

Nginx server_name 配置说明

Nginx 是一个高性能的反向代理和负载均衡服务器&#xff0c;其核心配置之一是 server 块中的 server_name 指令。server_name 决定了 Nginx 如何根据客户端请求的 Host 头匹配对应的虚拟主机&#xff08;Virtual Host&#xff09;。 1. 简介 Nginx 使用 server_name 指令来确定…...

【Go】3、Go语言进阶与依赖管理

前言 本系列文章参考自稀土掘金上的 【字节内部课】公开课&#xff0c;做自我学习总结整理。 Go语言并发编程 Go语言原生支持并发编程&#xff0c;它的核心机制是 Goroutine 协程、Channel 通道&#xff0c;并基于CSP&#xff08;Communicating Sequential Processes&#xff0…...

如何理解 IP 数据报中的 TTL?

目录 前言理解 前言 面试灵魂一问&#xff1a;说说对 IP 数据报中 TTL 的理解&#xff1f;我们都知道&#xff0c;IP 数据报由首部和数据两部分组成&#xff0c;首部又分为两部分&#xff1a;固定部分和可变部分&#xff0c;共占 20 字节&#xff0c;而即将讨论的 TTL 就位于首…...

.Net Framework 4/C# 关键字(非常用,持续更新...)

一、is 关键字 is 关键字用于检查对象是否于给定类型兼容,如果兼容将返回 true,如果不兼容则返回 false,在进行类型转换前,可以先使用 is 关键字判断对象是否与指定类型兼容,如果兼容才进行转换,这样的转换是安全的。 例如有:首先创建一个字符串对象,然后将字符串对象隐…...

PAN/FPN

import torch import torch.nn as nn import torch.nn.functional as F import mathclass LowResQueryHighResKVAttention(nn.Module):"""方案 1: 低分辨率特征 (Query) 查询高分辨率特征 (Key, Value).输出分辨率与低分辨率输入相同。"""def __…...

2025年渗透测试面试题总结-腾讯[实习]科恩实验室-安全工程师(题目+回答)

安全领域各种资源&#xff0c;学习文档&#xff0c;以及工具分享、前沿信息分享、POC、EXP分享。不定期分享各种好玩的项目及好用的工具&#xff0c;欢迎关注。 目录 腾讯[实习]科恩实验室-安全工程师 一、网络与协议 1. TCP三次握手 2. SYN扫描原理 3. HTTPS证书机制 二…...

Web中间件--tomcat学习

Web中间件–tomcat Java虚拟机详解 什么是JAVA虚拟机 Java虚拟机是一个抽象的计算机&#xff0c;它可以执行Java字节码。Java虚拟机是Java平台的一部分&#xff0c;Java平台由Java语言、Java API和Java虚拟机组成。Java虚拟机的主要作用是将Java字节码转换为机器代码&#x…...

WebRTC从入门到实践 - 零基础教程

WebRTC从入门到实践 - 零基础教程 目录 WebRTC简介 基础概念 工作原理 开发环境搭建 基础实践 三个实战案例 常见问题解答 1. WebRTC简介 1.1 什么是WebRTC&#xff1f; WebRTC&#xff08;Web Real-Time Communication&#xff09;是一个支持网页浏览器进行实时语音…...

毫米波雷达基础理论(3D+4D)

3D、4D毫米波雷达基础知识及厂商选型 PreView : https://mp.weixin.qq.com/s/bQkju4r6med7I3TBGJI_bQ 1. FMCW毫米波雷达基础知识 主要参考博文&#xff1a; 一文入门汽车毫米波雷达基本原理 &#xff1a;https://mp.weixin.qq.com/s/_EN7A5lKcz2Eh8dLnjE19w 毫米波雷达基础…...