dataset

Author: Biparnak Roy

README.md

@@ -16,20 +16,13 @@ If you already have the darknet built in your system then ommit ```--recurse-sub
 
 ### Libraries
 
-There are some library dependencies : opencv-python, numpy, math, argparse, imutils, requests. 
+There are some library dependencies : opencv-python, numpy, math, argparse, imutils, requests, os-sys.
 
 You can install these libraries by executing the following command **(recomended)**
 
 ```
 pip install -r requirements.txt
 ```
-else you can also use:
-```
-pip install opencv-python
-pip install numpy
-pip install math
-pip install argparse
-```
 
 ## Building Darknet
 
@@ -177,6 +170,7 @@ mobile: 63%	(left_x:  489   top_y:  239   width:   21   height:   28)
 bag:    48%	(left_x:  597   top_y:  212   width:   20   height:   25)
 ```
 - Coverting to YOLO labels format
+
 The detections stored in the text file need to be converted to the YOLO label format for this we have to configure ```yolo.sh```
 
 ```
@@ -186,7 +180,7 @@ do
 done
 
 ```
-Here in place of ```<num list>``` paste the number list generated by ```count.py```  and in places of ```data``` put your name of images.(not manditory)
+Here in place of ```<num list>``` paste the number list generated by ```count.py```  and in places of ```data``` put your name of images (only if a differenr name for the dataset was used).
 
 Now we have to run this script by running following commands:
 
@@ -198,9 +192,18 @@ After this script completes running all the labels in the YOLO format will be st
 
 Thus we successfully created the dataset!!!
 
-# Retraing the model using our dataset
+# Retraining the model using our dataset
 
 To train the model using the new dataset run the following commands.
+- To generate ```train.txt```
 
 ```
-```
+python3 generate_train.py
+```
+- To start training
+
+```
+ cd darknet
+./darknet detector train data/obj.data ../cfg/<your cfg file> ../weights/<your pretrained weight file>
+```
+You should find the checkpoints of the weights being stored in the ```new_weights``` directory.

generate_train.py

@@ -0,0 +1,13 @@
+import os
+
+image_files = []
+os.chdir("dataset")
+for filename in os.listdir(os.getcwd()):
+    if filename.endswith(".jpg"):
+        image_files.append("dataset/" + filename)
+os.chdir("..")
+with open("train.txt", "w") as outfile:
+    for image in image_files:
+        outfile.write(image)
+        outfile.write("\n")
+    outfile.close()

requirements.txt

@@ -3,4 +3,5 @@ numpy
 math
 argparse
 imutils
-requests
+requests
+os-sys