You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* Release Candidate announcement 1.4.1rc2
* Moved 1.4.1 into the prime position, 1.4.0 to previous stable versions.
* update download sizes
* update link to oanes paper
Copy file name to clipboardexpand all lines: download.html
+80-80
Original file line number
Diff line number
Diff line change
@@ -13,57 +13,56 @@ <h2>Binaries</h2>
13
13
<br>
14
14
</p>
15
15
16
-
<h3id="stable">Version 1.4.0.post1 (10. November 2023)</h3>
16
+
<h3id="stable">Release Candidate 1.4.1rc2 (27. February 2025)</h3>
17
17
18
18
<p><strong>Release Highlights:</strong></p>
19
19
<ul>
20
-
<li><ahref="{{site.baseurl}}/documentation/nn/nn.html">Neural Network Workflow</a>: run pre-trained neural networks from <ahref="https://bioimage.io">Bioimage.io Model Zoo</a> in ilastik
21
-
<ul>
22
-
<li>Networks can run on CPU, GPU (choose the GPU build for Windows, or Linux), or on a remote server</li>
23
-
<li>No Neural Network training, yet - we’re working on it.</li>
24
-
</ul>
25
-
</li>
26
-
<li>New startup screen!</li>
27
-
<li>Major UI and performance improvements for the <ahref="{{site.baseurl}}/documentation/multicut/multicut.html">Multicut Workflow</a></li>
28
-
<li>Stability/Usability improvements in the <ahref="{{site.baseurl}}/documentation/carving/carving.html">Carving Workflow</a></li>
29
-
<li>By popular demand: redo/undo for brush-strokes</li>
30
-
<li>Many bugfixes throughout</li>
20
+
<li>Introduces support for the OME-Zarr file format</li>
<li>Native build for Apple Silicon (M1, M2, M3, M4) and Intel Macs</li>
23
+
<li>Tiff reading improvements</li>
24
+
<li>Added spherical texture object features, contributed by Oane Gros (<ahref="https://dx.plos.org/10.1371/journal.pcbi.1012349">publication</a>)</li>
25
+
<li>Better user interface in dark mode</li>
31
26
</ul>
32
27
33
28
<br>
34
-
ilastik 1.4.0 is distributed in two versions for Windows and Linux.
35
-
In addition to our regular build, you can download GPU-enabled builds.
36
-
With these you can <ahref="{{site.baseurl}}/documentation/nn/nn.html">run pre-trained neural network models</a> from the <ahref="https://bioimage.io/#/?partner=ilastik">bioimage.io Model Zoo</a> faster, if you have a compatible nvidia GPU on your system.
37
-
All other workflows, besides the <i>Neural Network Workflow (local)</i> will work on the CPU, like before.
38
-
<br>
39
-
If you are interested in running the ilastik Neural Network Workflow on a <i>remote</i> GPU, e.g. if you have a powerful server in your facility, you have to check for additional installation instructions for the server part in the <ahref="https://github.com/ilastik/tiktorch#tiktorch">tiktorch repository</a>.
40
-
After you install the server part, you can use your regular ilastik installation as a client with the <i>Neural Network Workflow (remote)</i>, ilastik will take care of the communication with the server.
41
-
<br>
42
-
Commercial solver support with gurobi 951 for tracking with learning.
43
29
44
-
Note that on OSX there is no 3D preview in Carving, all other functionality should work as expected.
45
-
We recommend OSX users to download the current <ahref="{{site.baseurl}}/download.html#beta">beta version</a>.
46
-
If you have an Apple Silicon (M1/M2/M3), then please select the "Apple Silicon (arm64)" build.
30
+
<p><strong>Notes:</strong></p>
31
+
<ul>
32
+
<li>Commercial solver support with gurobi <code>12.0.*</code> for tracking with learning.</li>
33
+
<li>Native builds for Apple Silicon (M1, M2, M3, M4) and Intel Macs are available, which means better performance on these machines overall.
34
+
Notes for the silicon version:
35
+
<ul>
36
+
<li>Currently only 2D networks are supported in the Neural Network and Trainable Domain Adaptation Workflows.
37
+
These, however, run accelerated leveraging Metal Performance Shaders (mps).</li>
<li>Native build for Apple Silicon (M1, M2, M3)</li>
102
-
<li>Tiff reading improvements</li>
103
-
<li>More responsive user interface in Multicut</li>
104
-
<li>Added spherical texture features, contributed by Oane Gros (<ahref="https://www.biorxiv.org/content/10.1101/2024.07.25.605050v1">preprint</a>)</li>
105
-
<li>Better user interface in dark mode</li>
106
-
<li>Introduces support for the OME-Zarr file format</li>
114
+
<li><ahref="{{site.baseurl}}/documentation/nn/nn.html">Neural Network Workflow</a>: run pre-trained neural networks from <ahref="https://bioimage.io">Bioimage.io Model Zoo</a> in ilastik
115
+
<ul>
116
+
<li>Networks can run on CPU, GPU (choose the GPU build for Windows, or Linux), or on a remote server</li>
117
+
<li>No Neural Network training, yet - we’re working on it.</li>
118
+
</ul>
119
+
</li>
120
+
<li>New startup screen!</li>
121
+
<li>Major UI and performance improvements for the <ahref="{{site.baseurl}}/documentation/multicut/multicut.html">Multicut Workflow</a></li>
122
+
<li>Stability/Usability improvements in the <ahref="{{site.baseurl}}/documentation/carving/carving.html">Carving Workflow</a></li>
123
+
<li>By popular demand: redo/undo for brush-strokes</li>
124
+
<li>Many bugfixes throughout</li>
107
125
</ul>
108
126
109
127
<br>
128
+
ilastik 1.4.0 is distributed in two versions for Windows and Linux.
129
+
In addition to our regular build, you can download GPU-enabled builds.
130
+
With these you can <ahref="{{site.baseurl}}/documentation/nn/nn.html">run pre-trained neural network models</a> from the <ahref="https://bioimage.io/#/?partner=ilastik">bioimage.io Model Zoo</a> faster, if you have a compatible nvidia GPU on your system.
131
+
All other workflows, besides the <i>Neural Network Workflow (local)</i> will work on the CPU, like before.
132
+
<br>
133
+
If you are interested in running the ilastik Neural Network Workflow on a <i>remote</i> GPU, e.g. if you have a powerful server in your facility, you have to check for additional installation instructions for the server part in the <ahref="https://github.com/ilastik/tiktorch#tiktorch">tiktorch repository</a>.
134
+
After you install the server part, you can use your regular ilastik installation as a client with the <i>Neural Network Workflow (remote)</i>, ilastik will take care of the communication with the server.
135
+
<br>
136
+
Commercial solver support with gurobi 951 for tracking with learning.
110
137
111
-
<p><strong>Notes:</strong></p>
112
-
<ul>
113
-
<li>Commercial solver support with gurobi 951 for tracking with learning.</li>
114
-
<li>This beta version is currently distributed without tensorflow</li>
115
-
<li>Experimental native build for Apple Silicon (M1, M2, M3) is available, which means better performance on these machines overall.
116
-
Notes for the silicon version:
117
-
<ul>
118
-
<li>Currently only 2D networks are supported in the Neural Network and Trainable Domain Adaptation Workflows.
119
-
These, however, run accelerated leveraging Metal Performance Shaders (mps).</li>
120
-
</ul>
121
-
</li>
122
-
</ul>
138
+
Note that on OSX there is no 3D preview in Carving, all other functionality should work as expected.
139
+
We recommend OSX users to download the current <ahref="{{site.baseurl}}/download.html#beta">beta version</a>.
140
+
If you have an Apple Silicon (M1/M2/M3), then please select the "Apple Silicon (arm64)" build.
0 commit comments