Commit
•
1b1681b
1
Parent(s):
9d7c778
Created using Colaboratory
Browse files- tutorial.ipynb +8 -8
tutorial.ipynb
CHANGED
@@ -74,7 +74,7 @@
|
|
74 |
"clear_output()\n",
|
75 |
"print('Setup complete. Using torch %s %s' % (torch.__version__, torch.cuda.get_device_properties(0) if torch.cuda.is_available() else 'CPU'))"
|
76 |
],
|
77 |
-
"execution_count":
|
78 |
"outputs": [
|
79 |
{
|
80 |
"output_type": "stream",
|
@@ -212,7 +212,7 @@
|
|
212 |
"gdrive_download('1Y6Kou6kEB0ZEMCCpJSKStCor4KAReE43','coco2017val.zip') # val2017 dataset\n",
|
213 |
"!mv ./coco ../ # move folder alongside /yolov5"
|
214 |
],
|
215 |
-
"execution_count":
|
216 |
"outputs": [
|
217 |
{
|
218 |
"output_type": "stream",
|
@@ -238,7 +238,7 @@
|
|
238 |
"# Run YOLOv5x on COCO val2017\n",
|
239 |
"!python test.py --weights yolov5x.pt --data coco.yaml --img 672"
|
240 |
],
|
241 |
-
"execution_count":
|
242 |
"outputs": [
|
243 |
{
|
244 |
"output_type": "stream",
|
@@ -352,7 +352,7 @@
|
|
352 |
"gdrive_download('1n_oKgR81BJtqk75b00eAjdv03qVCQn2f','coco128.zip') # coco128 dataset\n",
|
353 |
"!mv ./coco128 ../ # move folder alongside /yolov5"
|
354 |
],
|
355 |
-
"execution_count":
|
356 |
"outputs": [
|
357 |
{
|
358 |
"output_type": "stream",
|
@@ -405,7 +405,7 @@
|
|
405 |
"# Train YOLOv5s on coco128 for 3 epochs\n",
|
406 |
"!python train.py --img 640 --batch 16 --epochs 3 --data coco128.yaml --cfg yolov5s.yaml --weights yolov5s.pt --nosave --cache"
|
407 |
],
|
408 |
-
"execution_count":
|
409 |
"outputs": [
|
410 |
{
|
411 |
"output_type": "stream",
|
@@ -622,7 +622,7 @@
|
|
622 |
"colab_type": "text"
|
623 |
},
|
624 |
"source": [
|
625 |
-
"Training losses and performance metrics are saved to Tensorboard and also to a `runs/exp0/results.txt` logfile. `results.txt` is plotted as `results.png` after training completes. Partially completed `results.txt` files can be plotted with `from utils.utils import plot_results; plot_results()`. Here we show YOLOv5s trained on coco128 to 300 epochs, starting from scratch (
|
626 |
]
|
627 |
},
|
628 |
{
|
@@ -639,7 +639,7 @@
|
|
639 |
"source": [
|
640 |
"from utils.utils import plot_results; plot_results() # plot results.txt files as results.png"
|
641 |
],
|
642 |
-
"execution_count":
|
643 |
"outputs": [
|
644 |
{
|
645 |
"output_type": "execute_result",
|
@@ -701,7 +701,7 @@
|
|
701 |
"!rm -rf yolov5 && git clone https://github.com/ultralytics/yolov5\n",
|
702 |
"%cd yolov5"
|
703 |
],
|
704 |
-
"execution_count":
|
705 |
"outputs": []
|
706 |
},
|
707 |
{
|
|
|
74 |
"clear_output()\n",
|
75 |
"print('Setup complete. Using torch %s %s' % (torch.__version__, torch.cuda.get_device_properties(0) if torch.cuda.is_available() else 'CPU'))"
|
76 |
],
|
77 |
+
"execution_count": null,
|
78 |
"outputs": [
|
79 |
{
|
80 |
"output_type": "stream",
|
|
|
212 |
"gdrive_download('1Y6Kou6kEB0ZEMCCpJSKStCor4KAReE43','coco2017val.zip') # val2017 dataset\n",
|
213 |
"!mv ./coco ../ # move folder alongside /yolov5"
|
214 |
],
|
215 |
+
"execution_count": null,
|
216 |
"outputs": [
|
217 |
{
|
218 |
"output_type": "stream",
|
|
|
238 |
"# Run YOLOv5x on COCO val2017\n",
|
239 |
"!python test.py --weights yolov5x.pt --data coco.yaml --img 672"
|
240 |
],
|
241 |
+
"execution_count": null,
|
242 |
"outputs": [
|
243 |
{
|
244 |
"output_type": "stream",
|
|
|
352 |
"gdrive_download('1n_oKgR81BJtqk75b00eAjdv03qVCQn2f','coco128.zip') # coco128 dataset\n",
|
353 |
"!mv ./coco128 ../ # move folder alongside /yolov5"
|
354 |
],
|
355 |
+
"execution_count": null,
|
356 |
"outputs": [
|
357 |
{
|
358 |
"output_type": "stream",
|
|
|
405 |
"# Train YOLOv5s on coco128 for 3 epochs\n",
|
406 |
"!python train.py --img 640 --batch 16 --epochs 3 --data coco128.yaml --cfg yolov5s.yaml --weights yolov5s.pt --nosave --cache"
|
407 |
],
|
408 |
+
"execution_count": null,
|
409 |
"outputs": [
|
410 |
{
|
411 |
"output_type": "stream",
|
|
|
622 |
"colab_type": "text"
|
623 |
},
|
624 |
"source": [
|
625 |
+
"Training losses and performance metrics are saved to Tensorboard and also to a `runs/exp0/results.txt` logfile. `results.txt` is plotted as `results.png` after training completes. Partially completed `results.txt` files can be plotted with `from utils.utils import plot_results; plot_results()`. Here we show YOLOv5s trained on coco128 to 300 epochs, starting from scratch (blue), and from pretrained `yolov5s.pt` (orange)."
|
626 |
]
|
627 |
},
|
628 |
{
|
|
|
639 |
"source": [
|
640 |
"from utils.utils import plot_results; plot_results() # plot results.txt files as results.png"
|
641 |
],
|
642 |
+
"execution_count": null,
|
643 |
"outputs": [
|
644 |
{
|
645 |
"output_type": "execute_result",
|
|
|
701 |
"!rm -rf yolov5 && git clone https://github.com/ultralytics/yolov5\n",
|
702 |
"%cd yolov5"
|
703 |
],
|
704 |
+
"execution_count": null,
|
705 |
"outputs": []
|
706 |
},
|
707 |
{
|