monai
medical
katielink commited on
Commit
3c8069c
·
1 Parent(s): 8b5b5dc

Update README Formatting

Browse files
Files changed (3) hide show
  1. README.md +20 -24
  2. configs/metadata.json +2 -1
  3. docs/README.md +20 -24
README.md CHANGED
@@ -5,14 +5,13 @@ tags:
5
  library_name: monai
6
  license: apache-2.0
7
  ---
8
- # Description
9
  A pre-trained model for segmenting nuclei cells with user clicks/interactions.
10
 
11
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/11.gif)
12
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/33.gif)
13
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/22.gif)
14
 
15
- # Model Overview
16
  This model is trained using [BasicUNet](https://docs.monai.io/en/latest/networks.html#basicunet) over [ConSeP](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet) dataset.
17
 
18
  ## Data
@@ -23,17 +22,6 @@ unzip -q consep_dataset.zip
23
  ```
24
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_dataset.jpeg)<br/>
25
 
26
- ## Training configuration
27
- The training was performed with the following:
28
-
29
- - GPU: at least 12GB of GPU memory
30
- - Actual Model Input: 5 x 128 x 128
31
- - AMP: True
32
- - Optimizer: Adam
33
- - Learning Rate: 1e-4
34
- - Loss: DiceLoss
35
-
36
-
37
  ### Preprocessing
38
  After [downloading this dataset](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet/consep_dataset.zip),
39
  python script `data_process.py` from `scripts` folder can be used to preprocess and generate the final dataset for training.
@@ -91,33 +79,45 @@ Example dataset.json
91
  }
92
  ```
93
 
 
 
 
 
 
 
 
 
 
94
 
95
- ## Input and output formats
96
- ### Input: 5 channels
 
97
  - 3 RGB channels
98
  - +ve signal channel (this nuclei)
99
  - -ve signal channel (other nuclei)
100
 
101
- ### Output: 2 channels
 
102
  - 0 = Background
103
  - 1 = Nuclei
104
 
105
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_in_out.jpeg)
106
 
107
- ## Scores
 
108
  This model achieves the following Dice score on the validation data provided as part of the dataset:
109
 
110
  - Train Dice score = 0.89
111
  - Validation Dice score = 0.85
112
 
113
 
114
- ## Training Performance
115
  A graph showing the training Loss and Dice over 50 epochs.
116
 
117
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_loss.jpeg) <br>
118
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_dice.jpeg) <br>
119
 
120
- ## Validation Performance
121
  A graph showing the validation mean Dice over 50 epochs.
122
 
123
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_val_dice.jpeg) <br>
@@ -140,8 +140,7 @@ python -m monai.bundle run --config_file configs/train.json
140
  torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config_file "['configs/train.json','configs/multi_gpu_train.json']"
141
  ```
142
 
143
- Please note that the distributed training related options depend on the actual running environment, thus you may need to remove `--standalone`, modify `--nnodes` or do some other necessary changes according to the machine you used.
144
- Please refer to [pytorch's official tutorial](https://pytorch.org/tutorials/intermediate/ddp_tutorial.html) for more details.
145
 
146
  #### Override the `train` config to execute evaluation with the trained model:
147
 
@@ -161,9 +160,6 @@ torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config
161
  python -m monai.bundle run --config_file configs/inference.json
162
  ```
163
 
164
- # Disclaimer
165
- This is an example, not to be used for diagnostic purposes.
166
-
167
  # References
168
  [1] Koohbanani, Navid Alemi, et al. "NuClick: a deep learning framework for interactive segmentation of microscopic images." Medical Image Analysis 65 (2020): 101771. https://arxiv.org/abs/2005.14511.
169
 
 
5
  library_name: monai
6
  license: apache-2.0
7
  ---
8
+ # Model Overview
9
  A pre-trained model for segmenting nuclei cells with user clicks/interactions.
10
 
11
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/11.gif)
12
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/33.gif)
13
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/22.gif)
14
 
 
15
  This model is trained using [BasicUNet](https://docs.monai.io/en/latest/networks.html#basicunet) over [ConSeP](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet) dataset.
16
 
17
  ## Data
 
22
  ```
23
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_dataset.jpeg)<br/>
24
 
 
 
 
 
 
 
 
 
 
 
 
25
  ### Preprocessing
26
  After [downloading this dataset](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet/consep_dataset.zip),
27
  python script `data_process.py` from `scripts` folder can be used to preprocess and generate the final dataset for training.
 
79
  }
80
  ```
81
 
82
+ ## Training configuration
83
+ The training was performed with the following:
84
+
85
+ - GPU: at least 12GB of GPU memory
86
+ - Actual Model Input: 5 x 128 x 128
87
+ - AMP: True
88
+ - Optimizer: Adam
89
+ - Learning Rate: 1e-4
90
+ - Loss: DiceLoss
91
 
92
+
93
+ ## Input
94
+ 5 channels
95
  - 3 RGB channels
96
  - +ve signal channel (this nuclei)
97
  - -ve signal channel (other nuclei)
98
 
99
+ ## Output
100
+ 2 channels
101
  - 0 = Background
102
  - 1 = Nuclei
103
 
104
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_in_out.jpeg)
105
 
106
+
107
+ ## Performance
108
  This model achieves the following Dice score on the validation data provided as part of the dataset:
109
 
110
  - Train Dice score = 0.89
111
  - Validation Dice score = 0.85
112
 
113
 
114
+ #### Training Loss and Dice
115
  A graph showing the training Loss and Dice over 50 epochs.
116
 
117
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_loss.jpeg) <br>
118
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_dice.jpeg) <br>
119
 
120
+ #### Validation Dice
121
  A graph showing the validation mean Dice over 50 epochs.
122
 
123
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_val_dice.jpeg) <br>
 
140
  torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config_file "['configs/train.json','configs/multi_gpu_train.json']"
141
  ```
142
 
143
+ Please note that the distributed training-related options depend on the actual running environment; thus, users may need to remove `--standalone`, modify `--nnodes`, or do some other necessary changes according to the machine used. For more details, please refer to [pytorch's official tutorial](https://pytorch.org/tutorials/intermediate/ddp_tutorial.html).
 
144
 
145
  #### Override the `train` config to execute evaluation with the trained model:
146
 
 
160
  python -m monai.bundle run --config_file configs/inference.json
161
  ```
162
 
 
 
 
163
  # References
164
  [1] Koohbanani, Navid Alemi, et al. "NuClick: a deep learning framework for interactive segmentation of microscopic images." Medical Image Analysis 65 (2020): 101771. https://arxiv.org/abs/2005.14511.
165
 
configs/metadata.json CHANGED
@@ -1,7 +1,8 @@
1
  {
2
  "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
3
- "version": "0.0.8",
4
  "changelog": {
 
5
  "0.0.8": "enable deterministic training",
6
  "0.0.7": "Update with figure links",
7
  "0.0.6": "adapt to BundleWorkflow interface",
 
1
  {
2
  "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
3
+ "version": "0.0.9",
4
  "changelog": {
5
+ "0.0.9": "Update README Formatting",
6
  "0.0.8": "enable deterministic training",
7
  "0.0.7": "Update with figure links",
8
  "0.0.6": "adapt to BundleWorkflow interface",
docs/README.md CHANGED
@@ -1,11 +1,10 @@
1
- # Description
2
  A pre-trained model for segmenting nuclei cells with user clicks/interactions.
3
 
4
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/11.gif)
5
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/33.gif)
6
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/22.gif)
7
 
8
- # Model Overview
9
  This model is trained using [BasicUNet](https://docs.monai.io/en/latest/networks.html#basicunet) over [ConSeP](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet) dataset.
10
 
11
  ## Data
@@ -16,17 +15,6 @@ unzip -q consep_dataset.zip
16
  ```
17
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_dataset.jpeg)<br/>
18
 
19
- ## Training configuration
20
- The training was performed with the following:
21
-
22
- - GPU: at least 12GB of GPU memory
23
- - Actual Model Input: 5 x 128 x 128
24
- - AMP: True
25
- - Optimizer: Adam
26
- - Learning Rate: 1e-4
27
- - Loss: DiceLoss
28
-
29
-
30
  ### Preprocessing
31
  After [downloading this dataset](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet/consep_dataset.zip),
32
  python script `data_process.py` from `scripts` folder can be used to preprocess and generate the final dataset for training.
@@ -84,33 +72,45 @@ Example dataset.json
84
  }
85
  ```
86
 
 
 
 
 
 
 
 
 
 
87
 
88
- ## Input and output formats
89
- ### Input: 5 channels
 
90
  - 3 RGB channels
91
  - +ve signal channel (this nuclei)
92
  - -ve signal channel (other nuclei)
93
 
94
- ### Output: 2 channels
 
95
  - 0 = Background
96
  - 1 = Nuclei
97
 
98
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_in_out.jpeg)
99
 
100
- ## Scores
 
101
  This model achieves the following Dice score on the validation data provided as part of the dataset:
102
 
103
  - Train Dice score = 0.89
104
  - Validation Dice score = 0.85
105
 
106
 
107
- ## Training Performance
108
  A graph showing the training Loss and Dice over 50 epochs.
109
 
110
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_loss.jpeg) <br>
111
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_dice.jpeg) <br>
112
 
113
- ## Validation Performance
114
  A graph showing the validation mean Dice over 50 epochs.
115
 
116
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_val_dice.jpeg) <br>
@@ -133,8 +133,7 @@ python -m monai.bundle run --config_file configs/train.json
133
  torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config_file "['configs/train.json','configs/multi_gpu_train.json']"
134
  ```
135
 
136
- Please note that the distributed training related options depend on the actual running environment, thus you may need to remove `--standalone`, modify `--nnodes` or do some other necessary changes according to the machine you used.
137
- Please refer to [pytorch's official tutorial](https://pytorch.org/tutorials/intermediate/ddp_tutorial.html) for more details.
138
 
139
  #### Override the `train` config to execute evaluation with the trained model:
140
 
@@ -154,9 +153,6 @@ torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config
154
  python -m monai.bundle run --config_file configs/inference.json
155
  ```
156
 
157
- # Disclaimer
158
- This is an example, not to be used for diagnostic purposes.
159
-
160
  # References
161
  [1] Koohbanani, Navid Alemi, et al. "NuClick: a deep learning framework for interactive segmentation of microscopic images." Medical Image Analysis 65 (2020): 101771. https://arxiv.org/abs/2005.14511.
162
 
 
1
+ # Model Overview
2
  A pre-trained model for segmenting nuclei cells with user clicks/interactions.
3
 
4
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/11.gif)
5
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/33.gif)
6
  ![nuclick](https://github.com/mostafajahanifar/nuclick_torch/raw/master/docs/22.gif)
7
 
 
8
  This model is trained using [BasicUNet](https://docs.monai.io/en/latest/networks.html#basicunet) over [ConSeP](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet) dataset.
9
 
10
  ## Data
 
15
  ```
16
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_dataset.jpeg)<br/>
17
 
 
 
 
 
 
 
 
 
 
 
 
18
  ### Preprocessing
19
  After [downloading this dataset](https://warwick.ac.uk/fac/cross_fac/tia/data/hovernet/consep_dataset.zip),
20
  python script `data_process.py` from `scripts` folder can be used to preprocess and generate the final dataset for training.
 
72
  }
73
  ```
74
 
75
+ ## Training configuration
76
+ The training was performed with the following:
77
+
78
+ - GPU: at least 12GB of GPU memory
79
+ - Actual Model Input: 5 x 128 x 128
80
+ - AMP: True
81
+ - Optimizer: Adam
82
+ - Learning Rate: 1e-4
83
+ - Loss: DiceLoss
84
 
85
+
86
+ ## Input
87
+ 5 channels
88
  - 3 RGB channels
89
  - +ve signal channel (this nuclei)
90
  - -ve signal channel (other nuclei)
91
 
92
+ ## Output
93
+ 2 channels
94
  - 0 = Background
95
  - 1 = Nuclei
96
 
97
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_in_out.jpeg)
98
 
99
+
100
+ ## Performance
101
  This model achieves the following Dice score on the validation data provided as part of the dataset:
102
 
103
  - Train Dice score = 0.89
104
  - Validation Dice score = 0.85
105
 
106
 
107
+ #### Training Loss and Dice
108
  A graph showing the training Loss and Dice over 50 epochs.
109
 
110
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_loss.jpeg) <br>
111
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_train_dice.jpeg) <br>
112
 
113
+ #### Validation Dice
114
  A graph showing the validation mean Dice over 50 epochs.
115
 
116
  ![](https://developer.download.nvidia.com/assets/Clara/Images/monai_pathology_nuclick_annotation_val_dice.jpeg) <br>
 
133
  torchrun --standalone --nnodes=1 --nproc_per_node=2 -m monai.bundle run --config_file "['configs/train.json','configs/multi_gpu_train.json']"
134
  ```
135
 
136
+ Please note that the distributed training-related options depend on the actual running environment; thus, users may need to remove `--standalone`, modify `--nnodes`, or do some other necessary changes according to the machine used. For more details, please refer to [pytorch's official tutorial](https://pytorch.org/tutorials/intermediate/ddp_tutorial.html).
 
137
 
138
  #### Override the `train` config to execute evaluation with the trained model:
139
 
 
153
  python -m monai.bundle run --config_file configs/inference.json
154
  ```
155
 
 
 
 
156
  # References
157
  [1] Koohbanani, Navid Alemi, et al. "NuClick: a deep learning framework for interactive segmentation of microscopic images." Medical Image Analysis 65 (2020): 101771. https://arxiv.org/abs/2005.14511.
158