Upload 12 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

git_hash.txt

@@ -0,0 +1 @@
+961d20526206465f9ebef9affd131d53a9f516e6

preprocessor_config (1).json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "max_pixels": 12845056,
+    "min_pixels": 3136
+  },
+  "temporal_patch_size": 2
+}

results.json

@@ -0,0 +1 @@
+{"../colpali/data_dir/eval_vidore/tatdqa_test": {"ndcg_at_1": 0.70109, "ndcg_at_3": 0.7923, "ndcg_at_5": 0.81432, "ndcg_at_10": 0.82714, "ndcg_at_20": 0.83121, "ndcg_at_50": 0.83523, "ndcg_at_100": 0.83688, "map_at_1": 0.70109, "map_at_3": 0.77005, "map_at_5": 0.78241, "map_at_10": 0.78785, "map_at_20": 0.78901, "map_at_50": 0.78967, "map_at_100": 0.78981, "recall_at_1": 0.70109, "recall_at_3": 0.85662, "recall_at_5": 0.90948, "recall_at_10": 0.94836, "recall_at_20": 0.96416, "recall_at_50": 0.9842, "recall_at_100": 0.99453, "precision_at_1": 0.70109, "precision_at_3": 0.28554, "precision_at_5": 0.1819, "precision_at_10": 0.09484, "precision_at_20": 0.04821, "precision_at_50": 0.01968, "precision_at_100": 0.00995, "mrr_at_1": 0.7017010935601458, "mrr_at_3": 0.7709599027946537, "mrr_at_5": 0.7835965978128797, "mrr_at_10": 0.7887963123686088, "mrr_at_20": 0.7900069324299046, "mrr_at_50": 0.7906871013277383, "mrr_at_100": 0.7908122337289102, "naucs_at_1_max": 0.31323987398365244, "naucs_at_1_std": -0.06973458951296707, "naucs_at_1_diff1": 0.8257166838058789, "naucs_at_3_max": 0.3926659737473996, "naucs_at_3_std": 0.06571917292636736, "naucs_at_3_diff1": 0.7756394870292411, "naucs_at_5_max": 0.397380358594201, "naucs_at_5_std": 0.13868537387746546, "naucs_at_5_diff1": 0.7396051349982516, "naucs_at_10_max": 0.501090641931301, "naucs_at_10_std": 0.36694999105347903, "naucs_at_10_diff1": 0.7174193926609567, "naucs_at_20_max": 0.4765529924218107, "naucs_at_20_std": 0.3737660736114713, "naucs_at_20_diff1": 0.6900626633277733, "naucs_at_50_max": 0.6429809235278162, "naucs_at_50_std": 0.5965680225588305, "naucs_at_50_diff1": 0.7246754428536203, "naucs_at_100_max": 0.6364067564396122, "naucs_at_100_std": 0.5448683368190432, "naucs_at_100_diff1": 0.6812946217425379}, "../colpali/data_dir/eval_vidore/shiftproject_test": {"ndcg_at_1": 0.81, "ndcg_at_3": 0.8894, "ndcg_at_5": 0.90663, "ndcg_at_10": 0.90952, "ndcg_at_20": 0.90952, "ndcg_at_50": 0.9115, "ndcg_at_100": 0.9115, "map_at_1": 0.81, "map_at_3": 0.87167, "map_at_5": 0.88167, "map_at_10": 0.88267, "map_at_20": 0.88267, "map_at_50": 0.88298, "map_at_100": 0.88298, "recall_at_1": 0.81, "recall_at_3": 0.94, "recall_at_5": 0.98, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.81, "precision_at_3": 0.31333, "precision_at_5": 0.196, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.82, "mrr_at_3": 0.8766666666666667, "mrr_at_5": 0.8866666666666667, "mrr_at_10": 0.8877777777777777, "mrr_at_20": 0.8877777777777777, "mrr_at_50": 0.8880902777777777, "mrr_at_100": 0.8880902777777777, "naucs_at_1_max": 0.0649588768400645, "naucs_at_1_std": -0.2740845513122743, "naucs_at_1_diff1": 0.6852637644716851, "naucs_at_3_max": 0.3546529723000309, "naucs_at_3_std": -0.11009959539371503, "naucs_at_3_diff1": 0.5325241207594116, "naucs_at_5_max": 0.722222222222224, "naucs_at_5_std": 0.3384687208216692, "naucs_at_5_diff1": 0.42250233426704475, "naucs_at_10_max": 0.7222222222222276, "naucs_at_10_std": 0.5541549953314738, "naucs_at_10_diff1": 0.7222222222222276, "naucs_at_20_max": 0.7222222222222276, "naucs_at_20_std": 0.5541549953314738, "naucs_at_20_diff1": 0.7222222222222276, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/syntheticDocQA_artificial_intelligence_test": {"ndcg_at_1": 0.99, "ndcg_at_3": 0.99, "ndcg_at_5": 0.99387, "ndcg_at_10": 0.99387, "ndcg_at_20": 0.99387, "ndcg_at_50": 0.99387, "ndcg_at_100": 0.99387, "map_at_1": 0.99, "map_at_3": 0.99, "map_at_5": 0.992, "map_at_10": 0.992, "map_at_20": 0.992, "map_at_50": 0.992, "map_at_100": 0.992, "recall_at_1": 0.99, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.99, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.99, "mrr_at_3": 0.99, "mrr_at_5": 0.992, "mrr_at_10": 0.992, "mrr_at_20": 0.992, "mrr_at_50": 0.992, "mrr_at_100": 0.992, "naucs_at_1_max": 0.8692810457516276, "naucs_at_1_std": -0.5634920634920657, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 0.8692810457516356, "naucs_at_3_std": -0.5634920634921204, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/syntheticDocQA_government_reports_test": {"ndcg_at_1": 0.91, "ndcg_at_3": 0.95417, "ndcg_at_5": 0.96278, "ndcg_at_10": 0.96278, "ndcg_at_20": 0.96278, "ndcg_at_50": 0.96278, "ndcg_at_100": 0.96278, "map_at_1": 0.91, "map_at_3": 0.945, "map_at_5": 0.95, "map_at_10": 0.95, "map_at_20": 0.95, "map_at_50": 0.95, "map_at_100": 0.95, "recall_at_1": 0.91, "recall_at_3": 0.98, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.91, "precision_at_3": 0.32667, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.92, "mrr_at_3": 0.9566666666666667, "mrr_at_5": 0.9566666666666667, "mrr_at_10": 0.9566666666666667, "mrr_at_20": 0.9566666666666667, "mrr_at_50": 0.9566666666666667, "mrr_at_100": 0.9566666666666667, "naucs_at_1_max": 0.6380329909741665, "naucs_at_1_std": 0.3872289656603365, "naucs_at_1_diff1": 0.9419026870007259, "naucs_at_3_max": 0.8611111111111119, "naucs_at_3_std": 0.5401493930905577, "naucs_at_3_diff1": 0.9346405228758099, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/arxivqa_test_subsampled": {"ndcg_at_1": 0.824, "ndcg_at_3": 0.86576, "ndcg_at_5": 0.88065, "ndcg_at_10": 0.88761, "ndcg_at_20": 0.89161, "ndcg_at_50": 0.89571, "ndcg_at_100": 0.89671, "map_at_1": 0.824, "map_at_3": 0.85533, "map_at_5": 0.86363, "map_at_10": 0.86642, "map_at_20": 0.86749, "map_at_50": 0.8682, "map_at_100": 0.8683, "recall_at_1": 0.824, "recall_at_3": 0.896, "recall_at_5": 0.932, "recall_at_10": 0.954, "recall_at_20": 0.97, "recall_at_50": 0.99, "recall_at_100": 0.996, "precision_at_1": 0.824, "precision_at_3": 0.29867, "precision_at_5": 0.1864, "precision_at_10": 0.0954, "precision_at_20": 0.0485, "precision_at_50": 0.0198, "precision_at_100": 0.00996, "mrr_at_1": 0.826, "mrr_at_3": 0.8566666666666666, "mrr_at_5": 0.8647666666666667, "mrr_at_10": 0.8674404761904762, "mrr_at_20": 0.868713893416525, "mrr_at_50": 0.8694424481678229, "mrr_at_100": 0.8695414304104477, "naucs_at_1_max": 0.7663278157610652, "naucs_at_1_std": -0.09768014573587591, "naucs_at_1_diff1": 0.9187263853904275, "naucs_at_3_max": 0.8138309391467555, "naucs_at_3_std": -0.023388435717895166, "naucs_at_3_diff1": 0.8748014870425184, "naucs_at_5_max": 0.8622919756137744, "naucs_at_5_std": 0.09893172955456826, "naucs_at_5_diff1": 0.8950952930191681, "naucs_at_10_max": 0.8887671010433149, "naucs_at_10_std": 0.18117971826411405, "naucs_at_10_diff1": 0.9102829537612158, "naucs_at_20_max": 0.9081854964207933, "naucs_at_20_std": 0.31092436974789345, "naucs_at_20_diff1": 0.9193899782135069, "naucs_at_50_max": 1.0, "naucs_at_50_std": 0.7428571428571219, "naucs_at_50_diff1": 0.9215686274509768, "naucs_at_100_max": 1.0, "naucs_at_100_std": 0.5613912231558791, "naucs_at_100_diff1": 0.9346405228758466}, "../colpali/data_dir/eval_vidore/docvqa_test_subsampled": {"ndcg_at_1": 0.52993, "ndcg_at_3": 0.58573, "ndcg_at_5": 0.60558, "ndcg_at_10": 0.62266, "ndcg_at_20": 0.63582, "ndcg_at_50": 0.64908, "ndcg_at_100": 0.6581, "map_at_1": 0.52993, "map_at_3": 0.57206, "map_at_5": 0.58293, "map_at_10": 0.58991, "map_at_20": 0.59366, "map_at_50": 0.59581, "map_at_100": 0.59662, "recall_at_1": 0.52993, "recall_at_3": 0.62528, "recall_at_5": 0.67406, "recall_at_10": 0.72727, "recall_at_20": 0.77827, "recall_at_50": 0.84479, "recall_at_100": 0.90022, "precision_at_1": 0.52993, "precision_at_3": 0.20843, "precision_at_5": 0.13481, "precision_at_10": 0.07273, "precision_at_20": 0.03891, "precision_at_50": 0.0169, "precision_at_100": 0.009, "mrr_at_1": 0.5232815964523282, "mrr_at_3": 0.5728011825572801, "mrr_at_5": 0.5826681448632669, "mrr_at_10": 0.5891273360785556, "mrr_at_20": 0.5925723253011941, "mrr_at_50": 0.5943296372892853, "mrr_at_100": 0.59515893309537, "naucs_at_1_max": 0.1489445922678588, "naucs_at_1_std": 0.5944168012150516, "naucs_at_1_diff1": 0.9184631698098695, "naucs_at_3_max": 0.01752251361569101, "naucs_at_3_std": 0.7203256897672715, "naucs_at_3_diff1": 0.8894280781702324, "naucs_at_5_max": -0.04548428009159777, "naucs_at_5_std": 0.7690453232976177, "naucs_at_5_diff1": 0.8644439246273121, "naucs_at_10_max": -0.13892914655317037, "naucs_at_10_std": 0.8207534993760452, "naucs_at_10_diff1": 0.8438993391008719, "naucs_at_20_max": -0.27349363756086625, "naucs_at_20_std": 0.8575543278988474, "naucs_at_20_diff1": 0.8389957258659689, "naucs_at_50_max": -0.43528705588098854, "naucs_at_50_std": 0.9002887869267794, "naucs_at_50_diff1": 0.8407727389585528, "naucs_at_100_max": -0.4873376707138829, "naucs_at_100_std": 0.8779309480426964, "naucs_at_100_diff1": 0.8126388341509826}, "../colpali/data_dir/eval_vidore/syntheticDocQA_healthcare_industry_test": {"ndcg_at_1": 0.96, "ndcg_at_3": 0.97762, "ndcg_at_5": 0.98149, "ndcg_at_10": 0.98149, "ndcg_at_20": 0.98149, "ndcg_at_50": 0.98149, "ndcg_at_100": 0.98149, "map_at_1": 0.96, "map_at_3": 0.97333, "map_at_5": 0.97533, "map_at_10": 0.97533, "map_at_20": 0.97533, "map_at_50": 0.97533, "map_at_100": 0.97533, "recall_at_1": 0.96, "recall_at_3": 0.99, "recall_at_5": 1.0, "recall_at_10": 1.0, "recall_at_20": 1.0, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.96, "precision_at_3": 0.33, "precision_at_5": 0.2, "precision_at_10": 0.1, "precision_at_20": 0.05, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.96, "mrr_at_3": 0.9733333333333333, "mrr_at_5": 0.9753333333333333, "mrr_at_10": 0.9753333333333333, "mrr_at_20": 0.9753333333333333, "mrr_at_50": 0.9753333333333333, "mrr_at_100": 0.9753333333333333, "naucs_at_1_max": 0.6785714285714297, "naucs_at_1_std": -0.529178338001864, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 1.0, "naucs_at_3_std": 0.35807656395889226, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 1.0, "naucs_at_5_std": 1.0, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 1.0, "naucs_at_10_std": 1.0, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 1.0, "naucs_at_20_std": 1.0, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/infovqa_test_subsampled": {"ndcg_at_1": 0.89069, "ndcg_at_3": 0.91896, "ndcg_at_5": 0.92636, "ndcg_at_10": 0.93041, "ndcg_at_20": 0.93306, "ndcg_at_50": 0.93549, "ndcg_at_100": 0.93583, "map_at_1": 0.89069, "map_at_3": 0.91194, "map_at_5": 0.91599, "map_at_10": 0.91774, "map_at_20": 0.91851, "map_at_50": 0.91891, "map_at_100": 0.91894, "recall_at_1": 0.89069, "recall_at_3": 0.93927, "recall_at_5": 0.95749, "recall_at_10": 0.96964, "recall_at_20": 0.97976, "recall_at_50": 0.9919, "recall_at_100": 0.99393, "precision_at_1": 0.89069, "precision_at_3": 0.31309, "precision_at_5": 0.1915, "precision_at_10": 0.09696, "precision_at_20": 0.04899, "precision_at_50": 0.01984, "precision_at_100": 0.00994, "mrr_at_1": 0.888663967611336, "mrr_at_3": 0.9102564102564102, "mrr_at_5": 0.9148110661268556, "mrr_at_10": 0.9165558126084441, "mrr_at_20": 0.9173312497615903, "mrr_at_50": 0.917719246204226, "mrr_at_100": 0.917752984395859, "naucs_at_1_max": 0.6452597879999299, "naucs_at_1_std": -0.1583967206391628, "naucs_at_1_diff1": 0.9382939669972173, "naucs_at_3_max": 0.6792175380169426, "naucs_at_3_std": -0.07352824072374452, "naucs_at_3_diff1": 0.9330460900047705, "naucs_at_5_max": 0.8390629595590892, "naucs_at_5_std": 0.22116664077349757, "naucs_at_5_diff1": 0.9626852988386374, "naucs_at_10_max": 0.8657206615100057, "naucs_at_10_std": 0.2744376241740944, "naucs_at_10_diff1": 0.9564661819784096, "naucs_at_20_max": 0.9183327616354467, "naucs_at_20_std": 0.6804227380555364, "naucs_at_20_diff1": 0.9477594183740975, "naucs_at_50_max": 0.8979159520443043, "naucs_at_50_std": 0.674633185026432, "naucs_at_50_diff1": 0.9346992729676393, "naucs_at_100_max": 0.8638879360590712, "naucs_at_100_std": 0.5661775800352391, "naucs_at_100_diff1": 0.9129323639568517}, "../colpali/data_dir/eval_vidore/syntheticDocQA_energy_test": {"ndcg_at_1": 0.94, "ndcg_at_3": 0.95893, "ndcg_at_5": 0.95893, "ndcg_at_10": 0.96542, "ndcg_at_20": 0.96542, "ndcg_at_50": 0.9676, "ndcg_at_100": 0.9676, "map_at_1": 0.94, "map_at_3": 0.955, "map_at_5": 0.955, "map_at_10": 0.95768, "map_at_20": 0.95768, "map_at_50": 0.95811, "map_at_100": 0.95811, "recall_at_1": 0.94, "recall_at_3": 0.97, "recall_at_5": 0.97, "recall_at_10": 0.99, "recall_at_20": 0.99, "recall_at_50": 1.0, "recall_at_100": 1.0, "precision_at_1": 0.94, "precision_at_3": 0.32333, "precision_at_5": 0.194, "precision_at_10": 0.099, "precision_at_20": 0.0495, "precision_at_50": 0.02, "precision_at_100": 0.01, "mrr_at_1": 0.95, "mrr_at_3": 0.96, "mrr_at_5": 0.96, "mrr_at_10": 0.9625396825396826, "mrr_at_20": 0.9630659983291562, "mrr_at_50": 0.9630659983291562, "mrr_at_100": 0.9630659983291562, "naucs_at_1_max": 0.06847183317771653, "naucs_at_1_std": -0.9000933706816028, "naucs_at_1_diff1": 1.0, "naucs_at_3_max": 0.6150015561780285, "naucs_at_3_std": -1.21708683473389, "naucs_at_3_diff1": 1.0, "naucs_at_5_max": 0.6150015561780299, "naucs_at_5_std": -1.2170868347338937, "naucs_at_5_diff1": 1.0, "naucs_at_10_max": 0.7222222222222276, "naucs_at_10_std": -1.7399626517273863, "naucs_at_10_diff1": 1.0, "naucs_at_20_max": 0.7222222222222276, "naucs_at_20_std": -1.7399626517273863, "naucs_at_20_diff1": 1.0, "naucs_at_50_max": NaN, "naucs_at_50_std": NaN, "naucs_at_50_diff1": NaN, "naucs_at_100_max": NaN, "naucs_at_100_std": NaN, "naucs_at_100_diff1": NaN}, "../colpali/data_dir/eval_vidore/tabfquad_test_subsampled": {"ndcg_at_1": 0.84286, "ndcg_at_3": 0.88737, "ndcg_at_5": 0.89459, "ndcg_at_10": 0.89794, "ndcg_at_20": 0.90691, "ndcg_at_50": 0.90985, "ndcg_at_100": 0.91045, "map_at_1": 0.84286, "map_at_3": 0.87679, "map_at_5": 0.88071, "map_at_10": 0.88203, "map_at_20": 0.88445, "map_at_50": 0.88496, "map_at_100": 0.88502, "recall_at_1": 0.84286, "recall_at_3": 0.91786, "recall_at_5": 0.93571, "recall_at_10": 0.94643, "recall_at_20": 0.98214, "recall_at_50": 0.99643, "recall_at_100": 1.0, "precision_at_1": 0.84286, "precision_at_3": 0.30595, "precision_at_5": 0.18714, "precision_at_10": 0.09464, "precision_at_20": 0.04911, "precision_at_50": 0.01993, "precision_at_100": 0.01, "mrr_at_1": 0.8392857142857143, "mrr_at_3": 0.8744047619047619, "mrr_at_5": 0.8783333333333333, "mrr_at_10": 0.8797108843537415, "mrr_at_20": 0.8821667166283763, "mrr_at_50": 0.8826871907663073, "mrr_at_100": 0.8827487671209872, "naucs_at_1_max": 0.5016709740664825, "naucs_at_1_std": 0.19134881026646522, "naucs_at_1_diff1": 0.9055454059353004, "naucs_at_3_max": 0.6889741403807901, "naucs_at_3_std": 0.34459870904883405, "naucs_at_3_diff1": 0.8472780416514442, "naucs_at_5_max": 0.6857039111941077, "naucs_at_5_std": 0.40758896151053337, "naucs_at_5_diff1": 0.8275495383338527, "naucs_at_10_max": 0.6228446934329287, "naucs_at_10_std": 0.31633986928104846, "naucs_at_10_diff1": 0.8017740429505158, "naucs_at_20_max": 0.9183006535947714, "naucs_at_20_std": 0.7605042016806759, "naucs_at_20_diff1": 0.9183006535947714, "naucs_at_50_max": 1.0, "naucs_at_50_std": 1.0, "naucs_at_50_diff1": 0.8692810457515607, "naucs_at_100_max": 1.0, "naucs_at_100_std": 1.0, "naucs_at_100_diff1": 1.0}}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
+size 11420371

tokenizer_config.json

@@ -0,0 +1,144 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "ColQwen2Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_config.yml

@@ -0,0 +1,73 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2-ba256-ckpt-5e-nopad
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2Processor
+    pretrained_model_name_or_path:  "./models/colqwen2_base" # "./models/paligemma-3b-mix-448"
+    # num_image_tokens: 2048
+    # max_length: 50
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2
+    pretrained_model_name_or_path: "./models/colqwen2_base"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+#    device_map: "auto"
+#    quantization_config:
+#      (): transformers.BitsAndBytesConfig
+#      load_in_4bit: true
+#      bnb_4bit_quant_type: "nf4"
+#      bnb_4bit_compute_dtype:  "bfloat16"
+#      bnb_4bit_use_double_quant: true
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  add_suffix: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 5
+    per_device_train_batch_size: 64
+    gradient_checkpointing: true
+    gradient_checkpointing_kwargs: { "use_reentrant": false }
+    # gradient_checkpointing: true
+    # 6 x 8 gpus = 48 batch size
+    # gradient_accumulation_steps: 4
+    per_device_eval_batch_size: 8
+    eval_strategy: "steps"
+    dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 500
+    logging_steps: 10
+    eval_steps: 100
+    warmup_steps: 100
+    learning_rate: 5e-4
+    save_total_limit: 1
+    # resume_from_checkpoint: true
+    # optim: "paged_adamw_8bit"
+    # wandb logging
+    # wandb_project: "colqwen2"
+    # run_name: "colqwen2-ba32-nolora"
+    report_to: "wandb"
+
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 32
+    lora_alpha: 32
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+