v-training good practice

This commit is contained in:
vany5921
2019-09-30 14:30:05 +08:00
parent 086187a1d3
commit 635a85a425
35 changed files with 498 additions and 6 deletions
Vendored
BIN
View File
Binary file not shown.
BIN
View File
Binary file not shown.
Binary file not shown.

After

Width:  |  Height:  |  Size: 106 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 50 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 48 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 43 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 18 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 15 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 16 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 12 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.6 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.1 KiB

BIN
View File
Binary file not shown.
+2 -1
View File
@@ -6,7 +6,8 @@
***
:memo:**[Description](#Description)**&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;:bulb:**[Quick Start](en/quick_start/m5stickv/m5stickv_quick_start)**&nbsp;&nbsp;&nbsp;:electric_plug:**[Schematic](#Schematic)**&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;🛒**[Purchase](https://m5stack.com/collections/m5-core/products/stickv)**&nbsp;&nbsp;<img src="https://m5stack.oss-cn-shenzhen.aliyuncs.com/image/EasyLoader_logo-min.jpg">**[EasyLoader](#EasyLoader)**
:memo:**[Description](#Description)**&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;:bulb:**[Quick Start](en/quick_start/m5stickv/m5stickv_quick_start)**&nbsp;&nbsp;&nbsp;:electric_plug:**[Schematic](#Schematic)**&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;🛒**[Purchase](https://m5stack.com/collections/m5-core/products/stickv)**&nbsp;&nbsp;<img src="https://m5stack.oss-cn-shenzhen.aliyuncs.com/image/EasyLoader_logo-min.jpg">**[EasyLoader](#EasyLoader)**&nbsp;&nbsp;&nbsp;&nbsp;:camera:**[V-Training](en/related_documents/v-training)**
## Description
+1 -1
View File
@@ -119,7 +119,7 @@
One issue commonly exists on M5Stick-C, which happens when the battery is at a low level. In this case, some of the devices will have problem powering on, this is probably caused by a chaos timing sequence of power-on of ESP32, some component is involved, such as AXP192, 552, ESP32
There is one way to bring the device back to live: 1, Connect G0 to 3V3. 2. Plug in the USB cable. 3, The screen will light up and leave the USB to charge the device.
There is one way to bring the device back to live: 1. Connect G0 and 3V3 with a 680 ohm resisto. 2. Plug in the USB cable. 3. The screen will light up and leave the USB to charge the device.
</div>
</div>
BIN
View File
Binary file not shown.
+71
View File
@@ -0,0 +1,71 @@
# BeetleC {docsify-ignore-all}
<img src="assets\img\product_pics\hat\bugc_hat\bugc_hat_01.jpg" width="30%" height="30%"> <img src="assets\img\product_pics\hat\bugc_hat\bugc_hat_02.jpg" width="30%" height="30%">
:memo:**[Description](#Description)**&nbsp;&nbsp;&nbsp;&nbsp;<img src="https://m5stack.oss-cn-shenzhen.aliyuncs.com/image/EasyLoader_M5StickC_logo_min.png">**[EasyLoader](#EasyLoader)**&nbsp;&nbsp;&nbsp;&nbsp;🛒**[Purchase](https://m5stack.com/products/bugc-w-o-m5stickc)**
## Description
**BugC** is a programmable robot base compatible with the M5StickC. This has four DC motors, motor driver, two RGB LEDs, battery holder and a switch.
The Bugc base needs to be used in conjunction with the M5StickC controller. The base comes with an STM32F030F4 micro controller which controls all the motors and LEDs and this is controlled through I2C protocol by the M5StickC which sits on top of the base.
<img src="assets\img\product_pics\hat\beetlec_hat\beetlec_hat_03.jpg" width="30%" height="30%">
## Product Features
- remote control
- Programmable
- Four-way motor driver
- 4xGeared motors
- 2xRGB LEDs
- Simple and compact design
- Equipped with a battery holder
- Flexible movement in all directions
## Weight and Size
- Size: 55mm * 40mm * 25mm
- Motor shaft: ⌀0.81mm
## Applications
- RC motor
- Robot control
- Smart and cognitive toys
## Package Includes
- 1x BugC base
- 1x 16340 Battery
## EasyLoader
<img src="https://m5stack.oss-cn-shenzhen.aliyuncs.com/image/EasyLoader_M5StickC_logo.png" width="100px" style="margin-top:20px">
<a href="https://m5stack.oss-cn-shenzhen.aliyuncs.com/EasyLoader/HAT/BeetleC/EasyLoader_BugC.exe"><button type="button" class="btn btn-primary">click to download EasyLoader</button></a>
>EasyLoader is a simple and fast program burner. Every product page in EasyLoader provides a product-related case program. This can be burned to the M5 device through simple steps, and a series of function verifications can be performed.
>2. After downloading the software, double-click to run the application, connect the M5 device to the computer through the data cable, select the port parameters, click **"Burn"** to burn the program (**For M5StickC, set the baud rate to 115200 or 750000**)
## Code
### 1. Arduino IDE
*To get complete code, please click [here](https://github.com/m5stack/M5-ProductExampleCodes/tree/master/Hat/bugC/stickC/bugC).*
## Video
<video width="500" height="500" controls>
<source src="https://m5stack.oss-cn-shenzhen.aliyuncs.com/video/Product_example_video/HAT/bugC.mp4" type="video/mp4">
</video>
+1 -1
View File
@@ -8,7 +8,7 @@
## Description
**SERVO** is made to implement the bestest, easiest way to drive Servo motors. This M5 module will make quick work of your next Servo project! It is able to drive mutiple Servo motors, up to 12 channals. We also added an DC input for power supplement. Through M-BUS the DC in can automatically power the M5 core at top.
**SERVO** is made to implement the bestest, easiest way to drive Servo motors. This M5 module will make quick work of your next Servo project! It is able to drive mutiple Servo motors, up to 12 channels(Max load :14W) . We also added an DC input for power supplement. Through M-BUS the DC in can automatically power the M5 core at top.
Servo is powered by MEGA328 communicate via IIC(0x53).
+55
View File
@@ -15,6 +15,8 @@
**[6. Run Recognition Program](#Run-Recognition-Program)**
**[7. Good Practice](#Good-Practice)**
## Burner Firmware
@@ -133,3 +135,56 @@
}
</style>
## Good Practice
>1. If your Loss line looks like below, there is something wrong with your dataset, you need to either clear it up or add more pictures. If everything is OK, and adding more pictures does not help, our network structure may not good for solve your problem.
<img src="assets\img\related_documents\v-training\12.jpg" >
<img src="assets\img\related_documents\v-training\13.jpg" >
>2. If your Loss line looks like below, it seems the Convolutional Neural Network does not converge at all. There might be some serious problems in your data-set, check your data-set. If everything is OK, and adding more pictures does not help, our network structure may not good for solve your problem.
<img src="assets\img\related_documents\v-training\14.jpg" >
>3. If your Loss line looks like below, but recognition results are still not good, you might need to add more pictures and try again;
If everything is OK, and adding more pictures does not help, our network structure may not good for solve your problem.
<img src="assets\img\related_documents\v-training\15.jpg" >
>4. If the results you get is not very ideal, you might want to try to add few more pictures and train again. Network may converge better this time.
### Common Errors(FAQ):
>1. “CONTENT: Number of Classes presented in Train and Vaild dataset is not equal.”
Or
“CONTENT: Train or Valid folder not found. If you are using the M5StickV software, make sure you reach enough image counts of 35 per class.”
Or
“CONTENT: Number of Classes presented in Train and Vaild dataset is not equal.”
Answer: Your zip file should looks like this:
Inside your zip file, you have two folder named train and valid(or vaild, is also ok)
<img src="assets\img\related_documents\v-training\16.jpg" >
Inside each of the folder, You have several folders start named by class (class should only be a number from 1 to 10). The number of folders in your train and valid folder should be the same. And the name of the folder should follow each other and starting at 1.
<img src="assets\img\related_documents\v-training\17.jpg" >
>2. “CONTENT: Lake of Enough Valid Dataset, Only 16 pictures found, but you need 20 in total.”
Or
“CNTENT: Lake of Enough Train Dataset, Only 43 pictures found, but you need 45 in total.”
Answer: You dont have enough pictures in your train or valid folder. You need to add more photos to them. Or you accidentally add an extra class.
>3. “CONTENT: Number of Classes should larger or equal to two.”
Answer: Sorry, we dont support single class, you need to at least have two or more class folders in your valid and train folder.
>4. “CONTENT: Unexpected error happened during checking dataset, cannot identify image file 'dataset_tmp/xxxxxxxx_dataset/train/2/1.jpg'”
Answer: The system cannot read the image while processing it. You might need to replace this picture.
+77
View File
@@ -587,3 +587,80 @@ __________________________
>import Library
><img src="/image/Advanced module/Execute_user.gif" width="50%">
# SDCard
__________________________
#### Function Description
>SD card read and write operations
><img src="/image/Advanced module/SDCard.png" width="40%">
* __open sdcard file mode__
Open the specified file and perform a read or write operation. This file must exist in the r and r+ states, otherwise an error is reported. The a, w, and w+ modes are automatically created if no files exist.
* __file read all__
Read all the bytes of the file
* __file read bytes__
Read the given number of bytes
* __file read line__
Read a line of content
* __file write__
Write bytes to the file
* __file set seek__
Set the offset position of the read
* __file get seek__
Get the current offset position
* __sdcard listdir__
List the specified catalog file
* __file mkdir__
Set the offset position of the read
* __sdcard remove__
Delete specified file
* __sdcard rmdir__
Delete the specified directory
* __sdcard rename__
File rename
#### Instructions
>Create a test folder, create a file named TEST.text, write helloM5Stack, and read M5Stack
><img src="/image/Advanced module/SDCard_user.gif" width="50%">
# Http
__________________________
#### Function Description
>Send HTTP protocol
><img src="/image/Advanced module/Http.png" width="40%">
* __Http Request__
Create an Http access request by GET, POST, DELET, PUT, PATCH, the URL is the full HTTP connection address, create the header Headers, Data data in the form of a dictionary, send successfully to execute Success, and send failure to execute Fail.
* __Get Status Code__
Return status code
* __Get Data__
Return request data
#### Instructions
>Send a GET request to Baidu, return data and print
><img src="/image/Advanced module/Http_user.gif" width="50%">

Some files were not shown because too many files have changed in this diff Show More