From 161a934aac9e4346d151de07ed1a2fbcfc0645d1 Mon Sep 17 00:00:00 2001 From: Glenn Jocher Date: Thu, 1 Aug 2019 22:36:28 +0200 Subject: [PATCH 1/4] Update README.md --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index f57cfcf8..77255b54 100755 --- a/README.md +++ b/README.md @@ -53,10 +53,10 @@ Our Jupyter [notebook](https://colab.research.google.com/github/ultralytics/yolo Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set. Default training settings produce loss plots below, with **training speed of 0.25 s/batch on a V100 GPU (almost 50 COCO epochs/day)**. -Here we see training results from `coco_1img.data`, `coco_10img.data` and `coco_100img.data`, 3 example files available in the `data/` folder, which train and test on the first 1, 10 and 100 images of the coco2014 trainval dataset. +Here we see training results from `coco_16img.data`, `coco_64img.data`, 2 example datasets available in the `data/` folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset. `from utils import utils; utils.plot_results()` -![results](https://user-images.githubusercontent.com/26833433/56207787-ec9e7000-604f-11e9-94dd-e1fcc374270f.png) +![results](https://user-images.githubusercontent.com/26833433/62325526-1fa82a80-b4ac-11e9-958e-2a263bf15ab0.png) ## Image Augmentation From 3f1a0d63e814d8aec23849a15da159b11670ba09 Mon Sep 17 00:00:00 2001 From: Glenn Jocher Date: Thu, 1 Aug 2019 22:38:35 +0200 Subject: [PATCH 2/4] Update README.md --- README.md | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/README.md b/README.md index 77255b54..a4d25e3b 100755 --- a/README.md +++ b/README.md @@ -51,9 +51,7 @@ Our Jupyter [notebook](https://colab.research.google.com/github/ultralytics/yolo **Resume Training:** `python3 train.py --resume` to resume training from `weights/last.pt`. -Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set. Default training settings produce loss plots below, with **training speed of 0.25 s/batch on a V100 GPU (almost 50 COCO epochs/day)**. - -Here we see training results from `coco_16img.data`, `coco_64img.data`, 2 example datasets available in the `data/` folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset. +Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set, though we can train on smaller datasets to get started. Here we see training results from `coco_16img.data`, `coco_64img.data`, 2 example datasets available in the `data/` folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset. `from utils import utils; utils.plot_results()` ![results](https://user-images.githubusercontent.com/26833433/62325526-1fa82a80-b4ac-11e9-958e-2a263bf15ab0.png) From 02b7f2c7d6aab9099637e43377e0c6ec09c364e8 Mon Sep 17 00:00:00 2001 From: Glenn Jocher Date: Thu, 1 Aug 2019 22:40:18 +0200 Subject: [PATCH 3/4] Update README.md --- README.md | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index a4d25e3b..36a6063a 100755 --- a/README.md +++ b/README.md @@ -47,13 +47,11 @@ Our Jupyter [notebook](https://colab.research.google.com/github/ultralytics/yolo # Training -**Start Training:** `python3 train.py` to begin training after downloading COCO data with `data/get_coco_dataset.sh`. +**Start Training:** `python3 train.py` to begin training after downloading COCO data with `data/get_coco_dataset.sh`. Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set. **Resume Training:** `python3 train.py --resume` to resume training from `weights/last.pt`. -Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set, though we can train on smaller datasets to get started. Here we see training results from `coco_16img.data`, `coco_64img.data`, 2 example datasets available in the `data/` folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset. - -`from utils import utils; utils.plot_results()` +**Plot Training:** `from utils import utils; utils.plot_results()` plots training results from `coco_16img.data`, `coco_64img.data`, 2 example datasets available in the `data/` folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset. ![results](https://user-images.githubusercontent.com/26833433/62325526-1fa82a80-b4ac-11e9-958e-2a263bf15ab0.png) ## Image Augmentation From ae577f51e27e3fc2bab3ca2924671f700282bb4e Mon Sep 17 00:00:00 2001 From: Glenn Jocher Date: Thu, 1 Aug 2019 22:44:22 +0200 Subject: [PATCH 4/4] Update examples.ipynb --- examples.ipynb | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/examples.ipynb b/examples.ipynb index 45a96848..64c49f34 100644 --- a/examples.ipynb +++ b/examples.ipynb @@ -46,11 +46,8 @@ "import glob\n", "import torch\n", "import os\n", - "import matplotlib.pyplot as plt\n", - "%matplotlib inline\n", "\n", - "from IPython.display import Image \n", - "from IPython.display import clear_output\n", + "from IPython.display import Image, clear_output\n", "print('PyTorch %s %s' % (torch.__version__, torch.cuda.get_device_properties(0) if torch.cuda.is_available() else 'CPU'))" ], "execution_count": 39, @@ -337,4 +334,4 @@ "outputs": [] } ] -} \ No newline at end of file +}