GBLearn2 datasources problem(dsource-image)

[Serg Buslovsky]

Hi, All.

I'm trying to use GBLearn2 library for OCR.

So I've got the training working correctly with MNIST-formatted database(using the dsource-mnist class provided in the demo).
But got some problems with the testing stuff:
I'm trying to use dsource-image class for loading the test samples:
;; construct the datasource
(defvar datasource (new dsource-image 0 0.01) )
(defvar files_list (list))
(defvar labels_list (list))
(for (i 6 (length argv))
(progn
(setq files_list (list-merge files_list (list (concat-fname (nth argv i)))))
(setq labels_list (list-merge labels_list (list i)))
))
(==> datasource load-pgms files_list labels_list)
But when I do
(==> thetrainer test-sample datasource testmeter 1)
i'm getting:
*** lisp_c runtime error: Index dimension does not match size and step
** in: C_fprop_C_c_layer
** in: C_fprop_C_net_cscscf
** in: C_fprop_C_idx3_supervised_module

*** ==> : Run-time error in compiled code

It looks like dsource-image is returning the data in some incorrect format.

Any ideas?

[Serg Buslovsky]

Figured out that I'm passing the wrong testingdb datasource class(dsource-image), changed to dsource-idx3l-narrow build around dsource-image.
However it still crashes with the same error.
The code:
(defvar datasource (new dsource-image 0 0.01) )
(defvar files_list (list))
(defvar labels_list (list))
(for (i 6 (length argv))
(progn
(setq files_list (list-merge files_list (list (concat-fname (nth argv i)))))
(setq labels_list (list-merge labels_list (list i)))
))
(==> datasource load-pgms files_list labels_list)

  \(defvar datasource\_idx3l
    \(new dsource-idx3l-narrow datasource \(length files\_list\) 0\)\)

  ;;
  \(==> thetrainer test-sample datasource\_idx3l testmeter 1\)

Eror:
*** lisp_c runtime error: Index dimension does not match size and step
** in: C_fprop_C_c_layer
** in: C_fprop_C_net_cscscf
** in: C_fprop_C_idx3_supervised_module

*** ==> : Run-time error in compiled code