Commit 2b3e88d3 authored by Christoph Sommer's avatar Christoph Sommer

update notebooks

parent 60ad6578
......@@ -10,23 +10,19 @@
},
{
"cell_type": "markdown",
"metadata": {
"jupyter": {
"source_hidden": true
}
},
"metadata": {},
"source": [
"#### Usage\n",
"#### Usage:\n",
"\n",
"A **code cell** will either:\n",
"* import required modules\n",
"* produce GUI elements for setting parameters or input files\n",
"* execute a CARE processing step\n",
"* an execution is finished, when the *****-symbol left of the code cell disappears\n",
"* an execution is finished, when the *-symbol left of the code cell disappears\n",
"\n",
"Use **SHIFT+ENTER** to execute the selected code cell and step through the notebook\n",
"\n",
"#### Background\n",
"#### Note:\n",
"* CARE requires perfectly alligned pairs of low and high quality images for training. The low quality image can have a by a factor of 2 lower resoltion. \n",
"* After training, new, low quality images can be predicted with the trained model.\n",
"* Input images will be loaded with the Bioformats library (supports e. g. tif, czi, lsm, etc.)\n",
......@@ -49,7 +45,8 @@
"metadata": {},
"source": [
"## 0. Create or load project file\n",
"___"
"___\n",
"if project has been trained already, you can jump to section **4. Prediction** after loading the project file"
]
},
{
......@@ -73,7 +70,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"if project had been trained already, you can jump to section **4. Prediction** after loading the project file"
"if project has been trained already, you can jump to section **4. Prediction** after loading the project file"
]
},
{
......@@ -164,7 +161,7 @@
"metadata": {},
"outputs": [],
"source": [
"care.CareTrainer().predict_multiple(predict_file.value, n_tiles=(1,4,4))"
"care.CareTrainer().predict_multiple( predict_file.value, n_tiles=(1,4,4) )"
]
},
{
......
......@@ -10,23 +10,19 @@
},
{
"cell_type": "markdown",
"metadata": {
"jupyter": {
"source_hidden": true
}
},
"metadata": {},
"source": [
"#### Usage\n",
"#### Usage:\n",
"\n",
"A **code cell** will either:\n",
"* import required modules\n",
"* produce GUI elements for setting parameters or input files\n",
"* execute a N2V processing step\n",
"* an execution is finished, when the *****-symbol left of the code cell disappears\n",
"* an execution is finished, when the *-symbol left of the code cell disappears\n",
"\n",
"Use **SHIFT+ENTER** to execute the selected code cell and step through the notebook\n",
"\n",
"#### Background\n",
"#### Note:\n",
"* Input images will be loaded with the Bioformats library (supports e. g. tif, czi, lsm, etc.)\n",
"* 2D, 3D and movies are supported.\n",
"* Each channel will be processed independently\n",
......@@ -51,7 +47,8 @@
"metadata": {},
"source": [
"## 0. Create or load project file\n",
"___"
"___\n",
"if project has been trained already, you can jump to section **4. Predict only** after loading the project file"
]
},
{
......@@ -119,19 +116,9 @@
"metadata": {},
"source": [
"## 3. Denoise (train and predict)\n",
"---"
]
},
{
"cell_type": "markdown",
"metadata": {
"jupyter": {
"source_hidden": true
}
},
"source": [
"* Images will be tiled according to the **n_tiles** (z,y,x) parameter. (usefull for big images)\n",
"* For advanced training parameters type and execute *bif_n2v.train_predict?* in an empty code cell"
"---\n",
"* Images will be tiled according to the **n_tiles** = **(z, y, x)** parameter. (required for big images)\n",
"* For advanced training parameters type and execute *n2v.train_predict?* in an empty code cell"
]
},
{
......@@ -145,7 +132,7 @@
"outputs": [],
"source": [
"n2v.params.save()\n",
"n2v.train_predict(n_tiles=(1,1,1))"
"n2v.train_predict( n_tiles=(1,2,2) )"
]
},
{
......@@ -155,7 +142,8 @@
"## 4. Predict only\n",
"---\n",
"\n",
"**requires loaded and trained project**"
"**requires a loaded and trained project**\n",
"* Images will be tiled according to the **n_tiles** (z,y,x) parameter. (usefull for big images)"
]
},
{
......@@ -173,30 +161,7 @@
"metadata": {},
"outputs": [],
"source": [
"n2v.predict(files.value, n_tiles=(1,1,1))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"tensorflow.__version__"
"n2v.predict( files.value, n_tiles=(1,2,2) )"
]
},
{
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment