This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
pipeline:window:cryolo:picking_general_refine [2019/09/18 10:30] twagner [Picking particles - Using the general model refined for your data] |
pipeline:window:cryolo:picking_general_refine [2020/06/05 09:05] twagner |
||
---|---|---|---|
Line 1: | Line 1: | ||
+ | <note important> | ||
+ | |||
+ | **DOCUMENTATION OUTDATED** | ||
+ | |||
+ | The documentation has moved to https:// | ||
+ | |||
+ | </ | ||
+ | |||
===== Picking particles - Using the general model refined for your data ===== | ===== Picking particles - Using the general model refined for your data ===== | ||
Line 6: | Line 14: | ||
What does // | What does // | ||
- | The general model was trained on a lot of particles with a variety of shapes and therefore learned a good set of generic features. The last layers, however, learn a fairly abstract representation of the particles and it might be that they do not perfectly fit for your particle at hand. Fine-tuning only trains | + | The general model was trained on a lot of particles with a variety of shapes and therefore learned a robust |
Why should I // | Why should I // | ||
- | - From theory, using fine-tuning should reduce the risk of overfitting ((Overfitting means, that the model works good on the training micrographs, | + | - From theory, using fine-tuning should reduce the risk of overfitting ((Overfitting means, that the model works good on the training micrographs, |
- The training is much faster, as not all layers have to be trained. | - The training is much faster, as not all layers have to be trained. | ||
- The training will need less GPU memory ((We are testing crYOLO with its default configuration on graphic cards with >= 8 GB memory. Using the fine tune mode, it should also work with GPUs with 4 GB memory)) and therefore is usable with NVIDIA cards with less memory. | - The training will need less GPU memory ((We are testing crYOLO with its default configuration on graphic cards with >= 8 GB memory. Using the fine tune mode, it should also work with GPUs with 4 GB memory)) and therefore is usable with NVIDIA cards with less memory. | ||
- | However, the fine tune mode is still somewhat experimental and we will update this section | + | <note important> |
+ | The fine tune mode is still somewhat experimental and we will update this section | ||
+ | </ | ||
If you followed the installation instructions, | If you followed the installation instructions, | ||
Line 39: | Line 48: | ||
</ | </ | ||
- | <hidden **Create the configuration file using the command line:**> | + | <html> |
+ | <div style=" | ||
+ | < | ||
+ | </ | ||
+ | </ | ||
+ | <hidden> | ||
I assume your box files for training are in the folder '' | I assume your box files for training are in the folder '' | ||
< | < | ||
- | cryoloo.py config config_cryolo.json 160 --train_image_folder train_image --train_annot_folder train_annot --pretrained_weights gmodel_phosnet_20190516.h5 | + | cryolo_gui.py config config_cryolo.json 160 --train_image_folder train_image --train_annot_folder train_annot --pretrained_weights gmodel_phosnet_20190516.h5 |
</ | </ | ||
To get a full description of all available options type: | To get a full description of all available options type: | ||
< | < | ||
- | cryoloo.py config -h | + | cryolo_gui.py config -h |
</ | </ | ||
Line 55: | Line 69: | ||
< | < | ||
- | cryoloo.py config config_cryolo.json 160 --train_image_folder train_image --train_annot_folder train_annot --pretrained_weights gmodel_phosnet_20190516.h5 --valid_image_folder valid_img --valid_annot_folder valid_annot | + | cryolo_gui.py config config_cryolo.json 160 --train_image_folder train_image --train_annot_folder train_annot --pretrained_weights gmodel_phosnet_20190516.h5 --valid_image_folder valid_img --valid_annot_folder valid_annot |
</ | </ | ||
</ | </ | ||
+ | < | ||
+ | <div style=" | ||
+ | <b> </ | ||
+ | </ | ||
+ | </ | ||
==== 4. Training ==== | ==== 4. Training ==== | ||
Line 71: | Line 89: | ||
In the GUI choose the action //train//. In the //" | In the GUI choose the action //train//. In the //" | ||
- | {{ : | + | |
+ | {{ : | ||
In the //" | In the //" | ||
{{ : | {{ : | ||
<note important> | <note important> | ||
- | The number of layers to fine tune (specified by layers_fine_tune in the //" | + | **Adjust the number of layers to train** |
+ | |||
+ | The number of layers to fine tune (specified by layers_fine_tune in the //" | ||
</ | </ | ||
Line 93: | Line 114: | ||
</ | </ | ||
- | + | < | |
- | <hidden **Run training with the command line**> | + | <div style=" |
+ | < | ||
+ | </ | ||
+ | </ | ||
+ | <hidden> | ||
In comparison to the training from scratch, you can skip the warm up training ( -w 0 ). Moreover you have to add the // | In comparison to the training from scratch, you can skip the warm up training ( -w 0 ). Moreover you have to add the // | ||
Line 101: | Line 126: | ||
</ | </ | ||
</ | </ | ||
+ | < | ||
+ | <div style=" | ||
+ | <b> </ | ||
+ | </ | ||
+ | </ | ||
==== 5. Picking ==== | ==== 5. Picking ==== | ||
{{page> | {{page> |