Python-Mapo基于深度学习用卫星图像预测采矿前景和矿藏

# Mapo ## About Mapo AI is a virtual exploration assistant that manages your mineral exploration efforts. To learn more, check out the [Mapo Blog Post](https://blog.produvia.com/letter-to-the-mining-industry-298cf0a0c6a8) and [Mapo White Paper](https://docs.google.com/document/d/1Y5WQ_IpmxeAowbNVMdT7lOzStZruOqBy1_rQ0PUWCEk/view). # Data The `data` folder contains mineral occurence datasets for: 1. Canada > British Columbia 2. Canada > Alberta ## Canada > British Columbia ### 1. ARIS Mineral Assessment Report Index Dataset `ariasdata.csv` and `arismetadata.csv` - Number of data records: 35,898 - Date of last update: Unknown ### 2. MINFILE Mineral Occurrence Dataset `MINFILE.csv` - Number of records: 14,817 - Record last modified: 2018-11-27 To re-create `MINFILE.csv`, perform the following: 1. Go here: <http://apps.gov.bc.ca/pub/dwds/addProducts.do> 2. Search "MINFILE Mineral Occurrence Database" 3. Click `+` button to add database to order 4. Click "View Your Order" button 5. Slect "Geographic Long/Lat (dd)" under projection drop-down menu and "CSV" under format drop-down menu 6. Type your email address 7. Press "I accept the Terms and Conditions" 8. Press "Submit Order" 9. Receive email from <NRSApplications@gov.bc.ca> with the subject: "Your order XXXXXXX has been assembled" 10. Copy URL in the email body (i.e. <https://apps.gov.bc.ca/pub/dwds/initiateDownload.do?orderId=XXXXXXX>) 11. Paste URL into browser to download a ZIP file 12. Open ZIP file to extract contents 13. Open extracted folder 14. Open "MINFIL_MINERAL_FILE" folder 15. Use "MINFILE.csv" for data analysis ### 3. Mineral Resources Data System (MRDS) Dataset `mrds.csv` - Number of records: 244 - Record last modified: Unknown To re-create `mrds.csv`, perform the following: 1. Go here: <https://mrdata.usgs.gov/mrds/geo-inventory.php> 2. Click the "North America" link 3. Click the "Canada" link 4. Select "CSV" under Format menu 5. Click "Download" button 6. Look for "British Columbia" rows under the "state" column ## Canada > Alberta ### 1. Metallic Mineral Occurrence Dataset `Metallic_Mineral_Occurrence.csv` - [Website](https://geology-ags-aer.opendata.arcgis.com/datasets/metallic-mineral-occurrence) - Number of records: 385 - Record last modified: 2016-09-23 ### 2. Mineral Resources Data System (MRDS) Dataset `mrds.csv` - Number of records: 24 - Record last modified: Unknown To re-create `mrds.csv`, perform the following: 1. Go here: <https://mrdata.usgs.gov/mrds/geo-inventory.php> 2. Click the "North America" link 3. Click the "Canada" link 4. Select "CSV" under Format menu 5. Click "Download" button 6. Look for "Alberta" rows under the "state" column # Run on Local Machine 1. Create environment using pipenv with python 3.6.* ``` pipenv --python python3.6 ``` 2. Enter pipenv environment ``` pipenv shell ``` 3. Install packages (for development) ``` pipenv install -d ``` # Perform Exploratory Data Analysis (EDA) 1. To explore the existing datasets, review the `eda` folder 2. To visualize mineral occurrence data on a map, look for: Latitude, Longitude, Depth, or Elevation values. Use the guide below to get started. ## British Columbia, Canada `MINFILE.csv` - Relevant columns: `DECIMAL_LATITUDE`, `DECIMAL_LONGITUDE`, `ELEVATION`, `COMMODITY_DESCRIPTION1`, `COMMODITY_DESCRIPTION2`, `COMMODITY_DESCRIPTION3`, `COMMODITY_DESCRIPTION4`, `COMMODITY_DESCRIPTION5`, `COMMODITY_DESCRIPTION6`, `COMMODITY_DESCRIPTION7`, `COMMODITY_DESCRIPTION8` - Missing columns: year of discovery `mrds.csv` - Relevant columns: `latitude`, `longitude`, `commod1`, `commod2`, `commod3`, `disc_yr` - Missing columns: depth or elevation of discovery ## Alberta, Canada `Metallic_Mineral_Occurrence.csv` - Relevant columns: `Long_NAD83`, `Lat_NAD83`, `Depth_m`, `Comm_1`, `Comm_2`, `Location`, `Ref_AGS` `mrds.csv` - Relevant columns: `latitude`, `longitude`, `commod1`, `commod2`, `commod3`, `disc_yr` - Missing columns: depth or elevation of discovery