- Welcome to photools.com Community.
Recent posts
#1
General Discussion and Questions / Re: Anybody else having proble...
Last post by jamiesaun85 - Today at 04:36:06 AMQuote from: Mario on April 25, 2025, 08:28:24 AMI have no idea and I don't remember. I guess this is what I saw in the log at the time. I saw this in the log:
AIConnectorOpenAI: 2 HTTP Status Code: 408 'Timeout.'
Maybe you accidentally selected an OpenAI setting in AutoTagger? But then we would not see a timeout error but a "no API key" error.
Anyway, does it work now? As far as I can tell, this problem manifests only on your PC, and this makes this very hard to figure out.
Yes sir, it works well enough. Still doing the same thing, but im not bothred by it. Adding one more file to the Que is not a big deal at all. Thank you
#2
AutoTagger and AI / Re: AutoTagger Prompt Engineer...
Last post by Stenis - Today at 01:57:48 AMWhen tagging some images from Paris with Open AI mini-40 I "think" I have noticed a slightly better quality and detailed info in Autotagger text for JPG-images than RAW. If I deliberately let the AT make different text for a RAW-JPG pair I think I often can see this effect.
The JPEG-example:
Paris France 2017 - Restaurants and shops - "The shop La Cure Gourmande is located on Isle Saint-Louis. It features a wooden exterior with large windows displaying various food items. The interior is filled with colorful products and treats. The shop specializes in pastries and sweets, inviting customers to explore its offerings."
The text I got for the RAW-file was not able to localize the shop to Isle Saint-Louis for example.
Often the RAW-texts seem to have more of fluffy hallucinations than the JPEG-file texts.
Has anybody else seen this pattern?
Can it be because the undeveloped RAW-files iMatch has to handle is harder to interprete than a JPEG?
The problems with the undeveloped RAW-files seems to be a too wide dynamic range that makes them soft and less contrasty and on top of that less sharp.
Since I first had an idea to just tag the RAW and then later export the JPEG that doesn´t really seem to be a good idea.
Another reason for that workflow has been the virtual copies of the RAW I sometimes do in DXO Photolab.
In that case the metadata has to be present in the RAW before I do the virtual copy otherwise in will not get any metadata at all.
I have read recommendations here to first develop the images and then add the metadata but I can´t do that of these reasons.
I also always take RAW+JPG in camera and to do that I just might have found another reason to do so.
In iMatch I have found that the "Run once for all selected files" then works very well is I select a "JPEG"-file last and make that file the master for all that set of files.
The JPEG-example:
Paris France 2017 - Restaurants and shops - "The shop La Cure Gourmande is located on Isle Saint-Louis. It features a wooden exterior with large windows displaying various food items. The interior is filled with colorful products and treats. The shop specializes in pastries and sweets, inviting customers to explore its offerings."
The text I got for the RAW-file was not able to localize the shop to Isle Saint-Louis for example.
Often the RAW-texts seem to have more of fluffy hallucinations than the JPEG-file texts.
Has anybody else seen this pattern?
Can it be because the undeveloped RAW-files iMatch has to handle is harder to interprete than a JPEG?
The problems with the undeveloped RAW-files seems to be a too wide dynamic range that makes them soft and less contrasty and on top of that less sharp.
Since I first had an idea to just tag the RAW and then later export the JPEG that doesn´t really seem to be a good idea.
Another reason for that workflow has been the virtual copies of the RAW I sometimes do in DXO Photolab.
In that case the metadata has to be present in the RAW before I do the virtual copy otherwise in will not get any metadata at all.
I have read recommendations here to first develop the images and then add the metadata but I can´t do that of these reasons.
I also always take RAW+JPG in camera and to do that I just might have found another reason to do so.
In iMatch I have found that the "Run once for all selected files" then works very well is I select a "JPEG"-file last and make that file the master for all that set of files.
#3
General Discussion and Questions / Re: New Open AI Model
Last post by Mario - April 25, 2025, 05:19:29 PMNext release.
#4
General Discussion and Questions / Re: New Open AI Model
Last post by Stenis - April 25, 2025, 04:36:28 PMQuote:
"AutoTagger already has configurations for the Gemini AI and these models, all you need is to get yourself an API key. Pricing is very affordable. Details about Gemini in AutoTagger and how to get an API key is described in the corresponding help topic."
Really? Where? I also have read the documentation too.
By the way, if you need some help testing these new models just tell us.
#5
General Discussion and Questions / Re: AI Tags > IPTC Metadata / ...
Last post by Mario - April 25, 2025, 02:47:07 PMSee Metadata Templates for reference. Very useful feature.
If you often have to add the same combination of tag values (artist info, legal, driver and car info), have a look at AutoFill.
The OP is new to IMatch (we've spent several days of email exchange to get IMatch running smoothly on his PC - which actually required him to buy a new SSD!).
If you often have to add the same combination of tag values (artist info, legal, driver and car info), have a look at AutoFill.
The OP is new to IMatch (we've spent several days of email exchange to get IMatch running smoothly on his PC - which actually required him to buy a new SSD!).
#6
General Discussion and Questions / Re: AI Tags > IPTC Metadata / ...
Last post by Jingo - April 25, 2025, 01:37:33 PMJust to add, you can also store them in IMatch.AI tags and then use a metadata template to auto move them to standard tags as well.. this way, you get to store them in both places.
#7
General Discussion and Questions / Re: Anybody else having proble...
Last post by Mario - April 25, 2025, 08:28:24 AMI have no idea and I don't remember. I guess this is what I saw in the log at the time. I saw this in the log:
AIConnectorOpenAI: 2 HTTP Status Code: 408 'Timeout.'
Maybe you accidentally selected an OpenAI setting in AutoTagger? But then we would not see a timeout error but a "no API key" error.
Anyway, does it work now? As far as I can tell, this problem manifests only on your PC, and this makes this very hard to figure out.
AIConnectorOpenAI: 2 HTTP Status Code: 408 'Timeout.'
Maybe you accidentally selected an OpenAI setting in AutoTagger? But then we would not see a timeout error but a "no API key" error.
Anyway, does it work now? As far as I can tell, this problem manifests only on your PC, and this makes this very hard to figure out.
#8
General Discussion and Questions / Re: Use geonames premium servi...
Last post by Mario - April 25, 2025, 08:24:38 AMI suggest you open a feature request so I can see if this is of interest for a substantial number of users. Implementing this will require some work (new dialog elements, making the URL dynamic across all code points, documentation updates etc.).
#9
General Discussion and Questions / Re: AI Tags > IPTC Metadata / ...
Last post by Mario - April 25, 2025, 08:23:11 AMYou control where you store the content of Traits. You can either store them in user-defined AI Tags or you store them in regular XMP tags. User-defined AI tags are maintained in the IMatch database but not written to the image or XMP sidecar file.
This allows you to separate between tag data you consider "internal" and "public". Or to separate AI-made from human-made content.
You specify the target tag for traits in the trait editor. Here is a "Headline" trait I usually use, and I store it in the official XMP headline tag:
The same is true for the three standard prompts, you can either store them in normal XMP tags (description / hierarchical keywords) to separate AI-generated content from human-made content by storing the data into AI tags.
This allows you to separate between tag data you consider "internal" and "public". Or to separate AI-made from human-made content.
You specify the target tag for traits in the trait editor. Here is a "Headline" trait I usually use, and I store it in the official XMP headline tag:
The same is true for the three standard prompts, you can either store them in normal XMP tags (description / hierarchical keywords) to separate AI-generated content from human-made content by storing the data into AI tags.
#10
General Discussion and Questions / Re: Anybody else having proble...
Last post by jamiesaun85 - April 25, 2025, 02:19:55 AMQuote from: Mario on April 15, 2025, 07:18:27 PMThen you are switching to OpenAI (?) instead of LM Studio and OpenAI returns
Quote from: Mario on April 15, 2025, 07:18:27 PMSo, LM Studio produces responds with an invalid / empty HTTP response. Then you switch to OpenAI, which responds with a timeout.
Umm mm, no. Definitely not. I have never used any AI other than Ollama and now, LM studio. I have no idea why it would try to switch to open AI. I even have all the other options hidden in the settings for AutoTagger. I literally cannot select OpenAI unless I un-hide it first.
What's that all about? Why would it even try to use OpenAI?