Training in progress, step 159

README.md

@@ -1,18 +1,18 @@
 ---
-base_model: 314e/dexit-VLM-Qwen3_5-Entity-test5-integration
+base_model: Qwen/Qwen3.6-27B
 library_name: transformers
 model_name: result
 tags:
 - generated_from_trainer
 - unsloth
-- sft
 - trl
+- sft
 licence: license
 ---
 
 # Model Card for result
 
-This model is a fine-tuned version of [314e/dexit-VLM-Qwen3_5-Entity-test5-integration](https://huggingface.co/314e/dexit-VLM-Qwen3_5-Entity-test5-integration).
+This model is a fine-tuned version of [Qwen/Qwen3.6-27B](https://huggingface.co/Qwen/Qwen3.6-27B).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -21,7 +21,7 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 from transformers import pipeline
 
 question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="developer-314e/result", device="cuda")
+generator = pipeline("text-generation", model="developer-314e/result", device_map="auto")
 output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
 print(output["generated_text"])
 ```
@@ -31,12 +31,13 @@ print(output["generated_text"])
  
 
 
+
 This model was trained with SFT.
 
 ### Framework versions
 
-- TRL: 0.22.2
-- Transformers: 5.3.0
+- TRL: 1.14.0
+- Transformers: 5.14.1
 - Pytorch: 2.7.1
 - Datasets: 4.3.0
 - Tokenizers: 0.22.2
@@ -48,12 +49,11 @@ This model was trained with SFT.
 Cite TRL as:
     
 ```bibtex
-@misc{vonwerra2022trl,
-	title        = {{TRL: Transformer Reinforcement Learning}},
-	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
-	year         = 2020,
-	journal      = {GitHub repository},
-	publisher    = {GitHub},
-	howpublished = {\url{https://github.com/huggingface/trl}}
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
 }
 ```

adapter_config.json

@@ -7,7 +7,7 @@
     "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5",
     "unsloth_fixed": true
   },
-  "base_model_name_or_path": "314e/dexit-VLM-Qwen3_5-Entity-test5-integration",
+  "base_model_name_or_path": "Qwen/Qwen3.6-27B",
   "bias": "none",
   "corda_config": null,
   "ensure_weight_tying": false,
@@ -16,29 +16,32 @@
   "fan_in_fan_out": false,
   "inference_mode": true,
   "init_lora_weights": true,
+  "kasa_config": null,
   "layer_replication": null,
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 128,
+  "lora_alpha": 512,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "lora_ga_config": null,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
+  "monteclora_config": null,
   "peft_type": "LORA",
-  "peft_version": "0.19.1",
+  "peft_version": "0.21.0",
   "qalora_group_size": 16,
-  "r": 64,
+  "r": 256,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense).*?(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mlp|feed_forward|ffn|dense)\\.(?:(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
+  "target_modules": "(?:.*?(?:language|text).*?(?:self_attn|attention|attn|mixer|mlp|feed_forward|ffn|dense|mixer).*?(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj))|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn|mixer|mlp|feed_forward|ffn|dense|mixer)\\.(?:(?:qkv|proj|linear_fc1|linear_fc2|out_proj|in_proj_qkv|in_proj_z|in_proj_b|in_proj_a|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
   "target_parameters": null,
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,
   "use_bdlora": null,
   "use_dora": false,
   "use_qalora": false,
-  "use_rslora": false
+  "use_rslora": false,
+  "velora_config": null
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7c6e60946115db086c62b78b6fb5e2751bca36c3fc390d2b12ee344b129cd16
-size 815758512
+oid sha256:43553fb098d706aaa9a447e81f2fe1f47a77862963ab1b87fd6b702767899d79
+size 7470734200

chat_template.jinja

@@ -97,7 +97,7 @@
             {%- endif %}
         {%- endif %}
         {%- set reasoning_content = reasoning_content|trim %}
-        {%- if loop.index0 > ns.last_query_index %}
+        {%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}
             {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
         {%- else %}
             {{- '<|im_start|>' + message.role + '\n' + content }}
@@ -116,11 +116,10 @@
                 {%- else %}
                     {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
                 {%- endif %}
-                {%- if tool_call.arguments is mapping %}
-                    {%- for args_name in tool_call.arguments %}
-                        {%- set args_value = tool_call.arguments[args_name] %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
                         {{- '<parameter=' + args_name + '>\n' }}
-                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
                         {{- args_value }}
                         {{- '\n</parameter>\n' }}
                     {%- endfor %}
@@ -147,9 +146,9 @@
 {%- endfor %}
 {%- if add_generation_prompt %}
     {{- '<|im_start|>assistant\n' }}
-    {%- if enable_thinking is defined and enable_thinking is true %}
-        {{- '<think>\n' }}
-    {%- else %}
+    {%- if enable_thinking is defined and enable_thinking is false %}
         {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
     {%- endif %}
 {%- endif %}

distribution.json

@@ -0,0 +1,3 @@
+{
+    "facesheet": 6058
+}

doctype_entity_distribution.json

@@ -0,0 +1,30 @@
+{
+    "facesheet": {
+        "accession_number": 1513,
+        "collection_date": 2801,
+        "facility_address": 5652,
+        "facility_city": 5678,
+        "facility_name": 5647,
+        "facility_state": 5694,
+        "facility_zip": 5721,
+        "other_insurance_name_1": 1962,
+        "other_insurance_number_1": 1967,
+        "patient_dob": 5673,
+        "patient_first_name": 6051,
+        "patient_last_name": 6056,
+        "patient_sex": 5636,
+        "provider_npi": 5017,
+        "sample_code": 5012,
+        "medicaid_insurance_number": 2052,
+        "medicare_insurance_number": 1597,
+        "patient_address": 2303,
+        "patient_city": 2321,
+        "patient_state": 2429,
+        "patient_zip": 2363,
+        "phone_number": 1269,
+        "social_security_number": 180,
+        "patient_middle_name": 974,
+        "other_insurance_name_2": 17,
+        "other_insurance_number_2": 18
+    }
+}

entity_distribution.json

@@ -0,0 +1,28 @@
+{
+    "accession_number": 1513,
+    "collection_date": 2801,
+    "facility_address": 5652,
+    "facility_city": 5678,
+    "facility_name": 5647,
+    "facility_state": 5694,
+    "facility_zip": 5721,
+    "other_insurance_name_1": 1962,
+    "other_insurance_number_1": 1967,
+    "patient_dob": 5673,
+    "patient_first_name": 6051,
+    "patient_last_name": 6056,
+    "patient_sex": 5636,
+    "provider_npi": 5017,
+    "sample_code": 5012,
+    "medicaid_insurance_number": 2052,
+    "medicare_insurance_number": 1597,
+    "patient_address": 2303,
+    "patient_city": 2321,
+    "patient_state": 2429,
+    "patient_zip": 2363,
+    "phone_number": 1269,
+    "social_security_number": 180,
+    "patient_middle_name": 974,
+    "other_insurance_name_2": 17,
+    "other_insurance_number_2": 18
+}

file.log

@@ -0,0 +1,43 @@
+2026-09-26 08:47:48.779 | INFO     | app.tasks.entity.vlm.dataset:create:716 - Using pre-split dataset mode
+2026-09-26 08:47:48.811 | INFO     | app.tasks.entity.vlm.dataset:_download_current_splits:1641 - Downloading train dataset from s3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/train.json
+2026-09-26 08:47:49.224 | INFO     | app.tasks.entity.vlm.dataset:_download_split_documents:1498 - Processing %s %s/%s with %d documents
+2026-09-26 08:48:24.172 | INFO     | app.tasks.entity.vlm.dataset:_download_current_splits:1641 - Downloading val dataset from s3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/eval.json
+2026-09-26 08:48:24.400 | INFO     | app.tasks.entity.vlm.dataset:_download_split_documents:1498 - Processing %s %s/%s with %d documents
+2026-09-26 08:48:29.540 | INFO     | app.tasks.entity.vlm.dataset:_download_current_splits:1641 - Downloading test dataset from s3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/test.json
+2026-09-26 08:48:29.744 | INFO     | app.tasks.entity.vlm.dataset:_download_split_documents:1498 - Processing %s %s/%s with %d documents
+2026-09-26 08:48:40.851 | INFO     | app.tasks.entity.vlm.dataset:_download_miscellaneous:1688 - Downloading miscellaneous data from s3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/miscellaneous.json
+2026-09-26 08:48:41.032 | INFO     | app.tasks.entity.vlm.dataset:_download_previous_datasets:1597 - Found 0 previous dataset(s) to consolidate
+2026-09-26 08:48:41.033 | INFO     | app.tasks.entity.vlm.dataset:download_presplit_datasets:1739 - Successfully downloaded all datasets to /home/ubuntu/dataset
+2026-09-26 08:49:02.820 | INFO     | app.tasks.entity.vlm.dataset:remove_null_files:640 - Removed 2 files with null values
+2026-09-26 08:49:05.099 | INFO     | app.tasks.entity.vlm.reports:get_distribution:1520 - Original number of ground truth files 6058
+2026-09-26 08:49:10.146 | INFO     | app.tasks.entity.vlm.reports:get_distribution:1535 - After removing empty files 6058
+2026-09-26 08:49:10.379 | INFO     | app.tasks.entity.vlm.reports:get_entity_distribution:1635 - Creating entity distribution for 6058 files
+2026-09-26 08:49:14.975 | DEBUG    | app.commonmlops:get_oauth_token:834 - OAuth token request status: 200
+2026-09-26 08:49:15.263 | DEBUG    | app.commonmlops:get_training_job_queue:891 - Get training job queue status code: 200, api_url: https://aitest.api.dexit.tech/api/v1/mlops/trainingjobqueue/06a665f8-ebd0-7481-8000-83304d751ace
+2026-09-26 08:49:15.499 | DEBUG    | app.commonmlops:get_oauth_token:834 - OAuth token request status: 200
+2026-09-26 08:49:15.500 | DEBUG    | app.commonmlops:update_training_job_queue:937 - Updating training job queue with merged payload: {'id': '06a665f8-ebd0-7481-8000-83304d751ace', 'modeltype': 'entity', 'modelname': 'vlm_qwen3_6', 'jobtype': 'feedback', 'basemodelversionid': '019f0d56-3844-7c0a-8fab-4aa6a6d1f12e', 'datasetid': '06a665f5-06bf-7f15-8000-acbbe709e051', 'status': 'triggered', 'createddate': '2026-07-26T19:27:10.738422Z', 'createdby': '00000000-0000-0000-0000-000000000000', 'startdate': '2026-09-26T08:47:16.517659Z', 'enddate': None, 'accuracyreport': {'overall': {'recall': 0.9659282581805915, 'f1_score': 0.966580398334646, 'precision': 0.9672334196599482}, 'patient': {'recall': 0.9570468593654881, 'accuracy': 0.965850180851059, 'f1_score': 0.9578867442511313, 'precision': 0.9588553005522813}, 'facility': {'recall': 0.9681215596546779, 'accuracy': 0.9468811825459762, 'f1_score': 0.9706030656585861, 'precision': 0.9731011770220469}, 'patient-dob': {'recall': 0.9914821124361158, 'accuracy': 0.9848484848484849, 'f1_score': 0.9919045590115041, 'precision': 0.9923273657289002}, 'patient-sex': {'recall': 0.9991423670668954, 'accuracy': 0.9983935742971888, 'f1_score': 0.9991423670668954, 'precision': 0.9991423670668954}, 'patient-zip': {'recall': 0.9683168316831683, 'accuracy': 0.9847634322373697, 'f1_score': 0.9809428284854563, 'precision': 0.9939024390243902}, 'patient.dob': {'recall': 0.9821124361158433, 'accuracy': 0.9706582077716098, 'f1_score': 0.9842082799829279, 'precision': 0.9863130881094953}, 'patient.sex': {'recall': 0.9991423670668954, 'accuracy': 0.9983935742971888, 'f1_score': 0.9991423670668954, 'precision': 0.9991423670668954}, 'patient.zip': {'recall': 0.9663366336633663, 'accuracy': 0.9847512038523274, 'f1_score': 0.9809045226130653, 'precision': 0.9959183673469387}, 'sample-code': {'recall': 0.9757045675413022, 'accuracy': 0.9577814569536424, 'f1_score': 0.9752306945118989, 'precision': 0.974757281553398}, 'facility-zip': {'recall': 0.9839119390347163, 'accuracy': 0.9776714513556619, 'f1_score': 0.9880952380952381, 'precision': 0.9923142613151152}, 'facility.zip': {'recall': 0.9847586790855207, 'accuracy': 0.9808, 'f1_score': 0.9897872340425531, 'precision': 0.9948674080410608}, 'patient-city': {'recall': 0.9817444219066938, 'accuracy': 0.988791032826261, 'f1_score': 0.9857433808553971, 'precision': 0.9897750511247444}, 'patient.city': {'recall': 0.9817444219066938, 'accuracy': 0.988, 'f1_score': 0.9847405900305187, 'precision': 0.9877551020408163}, 'phone-number': {'recall': 0.9784172661870504, 'accuracy': 0.9855769230769231, 'f1_score': 0.9679715302491103, 'precision': 0.9577464788732394}, 'provider-npi': {'recall': 1.0, 'accuracy': 0.9991546914623838, 'f1_score': 0.9995147986414362, 'precision': 0.9990300678952473}, 'facility-city': {'recall': 0.9982920580700256, 'accuracy': 0.9975922953451043, 'f1_score': 0.9987184963690731, 'precision': 0.9991452991452991}, 'facility-name': {'recall': 0.9718189581554227, 'accuracy': 0.9505882352941176, 'f1_score': 0.9730654125694741, 'precision': 0.9743150684931506}, 'facility.city': {'recall': 0.9965841161400513, 'accuracy': 0.9967871485943776, 'f1_score': 0.998289136013687, 'precision': 1.0}, 'facility.name': {'recall': 0.9624252775405636, 'accuracy': 0.9339035769828927, 'f1_score': 0.9636596836254809, 'precision': 0.9648972602739726}, 'patient-state': {'recall': 0.8416030534351145, 'accuracy': 0.875, 'f1_score': 0.8424068767908309, 'precision': 0.8432122370936902}, 'patient.state': {'recall': 0.8530534351145038, 'accuracy': 0.8843226788432268, 'f1_score': 0.8546845124282982, 'precision': 0.8563218390804598}, 'facility-state': {'recall': 0.9361158432708688, 'accuracy': 0.8876233864844343, 'f1_score': 0.9369138959931799, 'precision': 0.9377133105802048}, 'facility.state': {'recall': 0.9335604770017035, 'accuracy': 0.8867781155015197, 'f1_score': 0.936351986330628, 'precision': 0.9391602399314481}, 'collection-date': {'recall': 0.9469026548672567, 'accuracy': 0.9458784346378019, 'f1_score': 0.9427312775330398, 'precision': 0.9385964912280702}, 'patient-address': {'recall': 0.9020408163265307, 'accuracy': 0.9278510473235065, 'f1_score': 0.9048106448311156, 'precision': 0.9075975359342916}, 'patient.address': {'recall': 0.9, 'accuracy': 0.9255813953488372, 'f1_score': 0.901840490797546, 'precision': 0.9036885245901639}, 'accession-number': {'recall': 0.9967948717948718, 'accuracy': 0.9991546914623838, 'f1_score': 0.9983948635634029, 'precision': 1.0}, 'facility-address': {'recall': 0.9718189581554227, 'accuracy': 0.9490994518402506, 'f1_score': 0.9722340879965827, 'precision': 0.9726495726495726}, 'facility.address': {'recall': 0.9632792485055508, 'accuracy': 0.9361370716510904, 'f1_score': 0.9649272882805817, 'precision': 0.9665809768637532}, 'patient-last-name': {'recall': 0.9887550200803212, 'accuracy': 0.9785373608903021, 'f1_score': 0.9891522699879469, 'precision': 0.9895498392282959}, 'patient.last-name': {'recall': 0.9879518072289156, 'accuracy': 0.9785202863961814, 'f1_score': 0.9891435464414957, 'precision': 0.9903381642512077}, 'patient-first-name': {'recall': 0.9927652733118971, 'accuracy': 0.985634477254589, 'f1_score': 0.9927652733118971, 'precision': 0.9927652733118971}, 'patient.first-name': {'recall': 0.9911575562700965, 'accuracy': 0.9848242811501597, 'f1_score': 0.9923541247484909, 'precision': 0.9935535858178888}, 'patient-middle-name': {'recall': 0.9471153846153846, 'accuracy': 0.980722891566265, 'f1_score': 0.9425837320574162, 'precision': 0.9380952380952381}, 'patient.middle-name': {'recall': 0.9519230769230769, 'accuracy': 0.9776, 'f1_score': 0.9339622641509433, 'precision': 0.9166666666666666}, 'other-insurance-name-1': {'recall': 0.8142857142857143, 'accuracy': 0.8877630553390491, 'f1_score': 0.8260869565217392, 'precision': 0.8382352941176471}, 'other-insurance-name-2': {'recall': 1.0, 'accuracy': 0.9991967871485944, 'f1_score': 0.6666666666666666, 'precision': 0.5}, 'social-security-number': {'recall': 0.868421052631579, 'accuracy': 0.9863453815261044, 'f1_score': 0.7951807228915663, 'precision': 0.7333333333333333}, 'other-insurance-number-1': {'recall': 0.8928571428571429, 'accuracy': 0.9353035143769968, 'f1_score': 0.9025270758122744, 'precision': 0.9124087591240876}, 'other-insurance-number-2': {'recall': 1.0, 'accuracy': 0.9991967871485944, 'f1_score': 0.6666666666666666, 'precision': 0.5}, 'medicaid-insurance-number': {'recall': 0.9063231850117096, 'accuracy': 0.9304897314375987, 'f1_score': 0.8979118329466357, 'precision': 0.8896551724137931}, 'medicare-insurance-number': {'recall': 0.9626436781609196, 'accuracy': 0.9744204636290967, 'f1_score': 0.9544159544159544, 'precision': 0.9463276836158192}}, 'metadata': {'clearml_details': {'hf_model_id': '314e/dexit-aitest-vlm_qwen3_6-entity-integration', 'clearml_task_id': '337cc64a6155492fb9dcd31f8cb93617', 'trained_entities': ['accession_number', 'collection_date', 'facility_address', 'facility_city', 'facility_name', 'facility_state', 'facility_zip', 'medicaid_insurance_number', 'medicare_insurance_number', 'other_insurance_name_1', 'other_insurance_name_2', 'other_insurance_number_1', 'other_insurance_number_2', 'patient_address', 'patient_city', 'patient_dob', 'patient_first_name', 'patient_last_name', 'patient_middle_name', 'patient_sex', 'patient_state', 'patient_zip', 'phone_number', 'provider_npi', 'sample_code', 'social_security_number'], 'new_modelversion_tag': 'v7.0.0', 'base_modelversion_tag': 'v1.0.0'}, 'hyperparameters': {'DATASET_PARAMS': {'DATASET__EVAL_S3_PATH': 's3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/eval.json', 'DATASET__TEST_S3_PATH': 's3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/test.json', 'DATASET__TRAIN_S3_PATH': 's3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/train.json', 'DATASET__MISCELLANEOUS_S3_PATH': 's3://aitest-dexit-tech/data/__mlops__/dataset/entity/06a665f5-06bf-7f15-8000-acbbe709e051/miscellaneous.json'}, 'MODEL_CONFIGURATION': {'SEED': 42, 'MODELNAME': 'vlm_qwen3_6', 'MODELTYPE': 'entity', 'ENVIRONMENT': 'integration', 'HF_MODEL_ID': '314e/dexit-aitest-vlm_qwen3_6-entity-integration', 'MODEL_FAMILY': 'qwen', 'ARTIFACTS_DIR': '/home/ubuntu/result', 'MODELVERSIONTAG': 'v1.0.0', 'HF_BASE_MODEL_ID': 'Qwen/Qwen3.6-27B', 'IMAGE_PATCH_SIZE': 16, 'THRESHOLD_OFFSET': 0.05, 'TRAIN_OOTB_MODEL': True, 'NEW_MODELVERSIONTAG': 'v7.0.0'}, 'MODEL_HYPERPARAMETERS': {'RANK': 256, 'ALPHA': 512, 'TRAIN_EPOCHS': 2, 'EVAL_BATCH_SIZE': 1, 'GRAD_ACCUM_STEPS': 4, 'TRAIN_BATCH_SIZE': 4, 'INFERENCE_BATCH_SIZE': 6}}, 'workflow_details': {'temporal_workflow_id': 'local-model-retraining-entity-8bd7891e', 'skypilot_cluster_name': 'dexit-aitest-skypilot-entity-1790411922-425638', 'temporal_workflow_run_id': '01a0dcdd-a70c-7d0d-9906-de3892df566a'}, 'modelconfig_snapshot': {'modelname': 'vlm_qwen3_6', 'modeltype': 'entity', 'deployment': {'auto_model_deployment_manual': False, 'auto_model_deployment_feedback': False}, 'feedback_config': {'bucket_low': 0.5, 'bucket_high': 0.7, 'auto_trigger_retraining': True, 'non_corrected_sample_ratio': 0.8, 'min_feedback_count_for_retraining': 1500, 'low_bucket_threshold_offset_multiplier': 0.5, 'auto_trigger_retraining_on_new_entities': False, 'high_bucket_threshold_offset_multiplier': 1.5, 'low_precision_bucket_corrected_sample_rate': 0.7, 'mid_precision_bucket_corrected_sample_rate': 0.7, 'high_precision_bucket_corrected_sample_rate': 0.8, 'low_precision_bucket_threshold_sample_count': 100}, 'training_config': {'seed': 42, 'dataset': {'eval_size': 0.1, 'test_size': 0.2, 'train_size': 0.7}, 'cloud_providers': [{'accelerators': ['A100-80GB:1', 'H100:1'], 'provider_name': 'runpod'}], 'hyperparameters': {'rank': 256, 'alpha': 512, 'train_epochs': 2, 'grad_accum_steps': 4, 'train_batch_size': 4, 'inference_batch_size': 6}, 'hf_base_model_id': 'Qwen/Qwen3.6-27B', 'train_ootb_model': True, 'previous_datasets': {'eval': {'enabled': True, 'distribution': {'weights': {'n-1': 1.0}, 'strategy': 'equal', 'decay_factor': 0.8}, 'max_versions_to_include': 'all'}, 'test': {'enabled': True, 'distribution': {'weights': {'n-1': 1.0}, 'strategy': 'equal', 'decay_factor': 0.8}, 'max_versions_to_include': 'all'}, 'train': {'enabled': True, 'distribution': {'weights': {'n-1': 1.0}, 'strategy': 'equal', 'decay_factor': 0.8}, 'max_versions_to_include': 'all'}}}, 'threshold_offset': 0.05}}, 'deploymentstatus': None, 'deployeddate': None}
+2026-09-26 08:49:15.793 | DEBUG    | app.commonmlops:update_training_job_queue:940 - Update training job queue status code: 200, api_url: https://aitest.api.dexit.tech/api/v1/mlops/trainingjobqueue/06a665f8-ebd0-7481-8000-83304d751ace
+2026-09-26 08:49:15.795 | SUCCESS  | app.commonmlops:update_trained_entities:979 - Updated trained_entities on training job queue
+2026-09-26 08:49:16.232 | INFO     | app.tasks.entity.vlm.reports:upload_initial_artifacts:1703 - Uploaded intial artifacts to ClearML
+2026-09-26 08:49:16.232 | INFO     | __main__:_run_training_phase:635 - Moving into standard training loop
+2026-09-26 08:49:16.233 | INFO     | __main__:setup_directories:210 - Loading pre-split dataset statistics
+2026-09-26 08:49:18.070 | INFO     | __main__:load_presplit_statistics:187 - Found 4250 training samples
+2026-09-26 08:49:18.587 | INFO     | __main__:load_presplit_statistics:189 - Found 1245 test samples
+2026-09-26 08:49:18.825 | INFO     | __main__:load_presplit_statistics:191 - Found 563 validation samples
+2026-09-26 08:49:18.825 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1052 - 
+Dataset Split Statistics:
+2026-09-26 08:49:18.826 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1053 - Total samples: 6058
+2026-09-26 08:49:18.826 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1055 - Train set: 4250 (70.2%)
+2026-09-26 08:49:18.826 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1056 - Test set: 1245 (20.6%)
+2026-09-26 08:49:18.827 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1057 - Validation set: 563 (9.3%)
+2026-09-26 08:49:18.828 | INFO     | app.tasks.entity.vlm.dataset:print_split_statistics:1066 - Split statistics saved to /home/ubuntu/result/split_statistics.json
+2026-09-26 08:49:18.899 | INFO     | __main__:standard_training_loop:332 - Initializing model
+2026-09-26 08:49:18.899 | INFO     | __main__:initialize_model:242 - Setting up model for retraining
+2026-09-26 08:49:18.900 | INFO     | __main__:initialize_model:245 - Training from base model: Qwen/Qwen3.6-27B
+2026-09-26 08:49:18.900 | INFO     | app.tasks.entity.vlm.model:__post_init__:321 - loading model Qwen/Qwen3.6-27B for training with unsloth
+2026-09-26 08:55:06.793 | INFO     | app.tasks.entity.vlm.model:calculate_max_steps:417 - num_samples: 4250, batch_size: 4, gradient_accumulation_steps: 4, epochs: 2
+2026-09-26 08:55:06.794 | INFO     | app.tasks.entity.vlm.model:calculate_max_steps:421 - calculating max steps
+2026-09-26 08:55:06.836 | INFO     | __main__:standard_training_loop:335 - Setting up trainer
+2026-09-26 08:55:06.836 | INFO     | __main__:create_trainer:284 - Datasets: 4250 train, 563 eval samples
+2026-09-26 08:55:07.496 | INFO     | __main__:standard_training_loop:338 - Starting training
+2026-09-26 08:55:07.680 | INFO     | app.tasks.entity.vlm.model:train_model:624 - everything is set up, launching training

null_files.json

@@ -0,0 +1,3 @@
+{
+    "null_files": []
+}

processor_config.json

@@ -1,6 +1,5 @@
 {
   "image_processor": {
-    "data_format": "channels_first",
     "do_convert_rgb": true,
     "do_normalize": true,
     "do_rescale": true,
@@ -10,7 +9,7 @@
       0.5,
       0.5
     ],
-    "image_processor_type": "Qwen2VLImageProcessorFast",
+    "image_processor_type": "Qwen2VLImageProcessor",
     "image_std": [
       0.5,
       0.5,
@@ -28,8 +27,6 @@
   },
   "processor_class": "Qwen3VLProcessor",
   "video_processor": {
-    "data_format": "channels_first",
-    "default_to_square": true,
     "do_convert_rgb": true,
     "do_normalize": true,
     "do_rescale": true,

removed_files.json

@@ -0,0 +1,4 @@
+[
+    "/home/ubuntu/dataset/train/facesheet/06973491-2c33-78ad-8000-28b22f42c5ca.json",
+    "/home/ubuntu/dataset/train/facesheet/06973491-2c33-7d17-8000-50d328faf93c.json"
+]

runs/Sep26_08-55-09_bc0f704212de/events.out.tfevents.1790412909.bc0f704212de.10077.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49561cfcc6c212ce161eadf70a5a8e5f61c40d38314960b1c355178f94fe89d1
+size 15806

sorted_distribution.json

@@ -0,0 +1,3 @@
+{
+    "facesheet": 6058
+}

split_statistics.json

@@ -0,0 +1 @@
+{"total": 6058, "train": 4250, "test": 1245, "val": 563}

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
-size 19989343
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer_config.json

@@ -10,6 +10,7 @@
   "errors": "replace",
   "image_token": "<|image_pad|>",
   "is_local": false,
+  "local_files_only": false,
   "model_max_length": 262144,
   "model_specific_special_tokens": {
     "audio_bos_token": "<|audio_start|>",
@@ -20,12 +21,12 @@
     "vision_bos_token": "<|vision_start|>",
     "vision_eos_token": "<|vision_end|>"
   },
-  "pad_token": "<|vision_pad|>",
+  "pad_token": "<|endoftext|>",
   "padding_side": "right",
   "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
   "processor_class": "Qwen3VLProcessor",
   "split_special_tokens": false,
-  "tokenizer_class": "TokenizersBackend",
+  "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null,
   "video_token": "<|video_pad|>",
   "vision_bos_token": "<|vision_start|>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26d676196297336f9f717002a8c4632780d3f2627bb420a44f75d61e6afbe983
-size 5585
+oid sha256:5e964d7edeca076b3878da9163aee9f069f9a0e5ff17a3e05ce7c2176e0d86b1
+size 5905