diff --git a/databricks_notebooks/bulk/Export_All.py b/databricks_notebooks/bulk/Export_All.py
index 8970771c..2ef136df 100644
--- a/databricks_notebooks/bulk/Export_All.py
+++ b/databricks_notebooks/bulk/Export_All.py
@@ -62,15 +62,29 @@
 
 # COMMAND ----------
 
-from mlflow_export_import.bulk.export_all import export_all
-
-export_all(
-    output_dir = output_dir, 
-    stages = stages,
-    export_latest_versions = export_latest_versions,
-    run_start_time = run_start_date,
-    export_permissions = export_permissions,
-    export_deleted_runs = export_deleted_runs,
-    notebook_formats = notebook_formats, 
-    use_threads = use_threads
-)
+# MAGIC %%capture captured
+# MAGIC 
+# MAGIC from mlflow_export_import.bulk.export_all import export_all
+# MAGIC 
+# MAGIC export_all(
+# MAGIC     output_dir = output_dir, 
+# MAGIC     stages = stages,
+# MAGIC     export_latest_versions = export_latest_versions,
+# MAGIC     run_start_time = run_start_date,
+# MAGIC     export_permissions = export_permissions,
+# MAGIC     export_deleted_runs = export_deleted_runs,
+# MAGIC     notebook_formats = notebook_formats, 
+# MAGIC     use_threads = use_threads
+# MAGIC )
+
+# COMMAND ----------
+
+# DBTITLE 1,write log file
+filepath = "/mnt/public-blobs/dcoles/mlflow_export_log.txt"
+
+dbutils.fs.rm(filepath)
+dbutils.fs.put(filepath, captured.stdout)
+
+# COMMAND ----------
+
+
diff --git a/databricks_notebooks/bulk/Export_Experiments.py b/databricks_notebooks/bulk/Export_Experiments.py
index 739d6acc..fc916130 100644
--- a/databricks_notebooks/bulk/Export_Experiments.py
+++ b/databricks_notebooks/bulk/Export_Experiments.py
@@ -1,8 +1,8 @@
 # Databricks notebook source
 # MAGIC %md ## Export Experiments
-# MAGIC 
+# MAGIC
 # MAGIC Export multiple experiments and all their runs.
-# MAGIC 
+# MAGIC
 # MAGIC Widgets
 # MAGIC * `1. Experiments` - comma delimited list of either experiment IDs or experiment names. `all` will export all experiments.
 # MAGIC * `2. Output directory` - shared directory between source and destination workspaces.
@@ -53,6 +53,22 @@
 
 # COMMAND ----------
 
+# DBTITLE 1,set up log file
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y-%m-%d-%H:%M:%S")
+ 
+logfile = f"export_experiments.{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+print("Logging to", logfile)
+
+# COMMAND ----------
+
 assert_widget(experiments, "1. Experiments")
 assert_widget(output_dir, "2. Output directory")
 
diff --git a/databricks_notebooks/bulk/Export_Models.py b/databricks_notebooks/bulk/Export_Models.py
index 7e7c3a9e..106d166a 100644
--- a/databricks_notebooks/bulk/Export_Models.py
+++ b/databricks_notebooks/bulk/Export_Models.py
@@ -1,8 +1,8 @@
 # Databricks notebook source
 # MAGIC %md ## Export Models
-# MAGIC 
+# MAGIC
 # MAGIC Export specified models, their version runs and the experiments that the runs belong to.
-# MAGIC 
+# MAGIC
 # MAGIC Widgets
 # MAGIC * `1. Models` - comma seperated registered model names to be exported. `all` will export all models.
 # MAGIC * `2. Output directory` - shared directory between source and destination workspaces.
@@ -13,7 +13,7 @@
 # MAGIC * `7. Export deleted runs`
 # MAGIC * `8. Notebook formats`
 # MAGIC * `9. Use threads`
-# MAGIC 
+# MAGIC
 # MAGIC See: https://github.com/mlflow/mlflow-export-import/blob/master/README_bulk.md#registered-models.
 
 # COMMAND ----------
@@ -69,6 +69,22 @@
 
 # COMMAND ----------
 
+# DBTITLE 1,set up log file
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y-%m-%d-%H:%M:%S")
+ 
+logfile = f"export_models.{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+print("Logging to", logfile)
+
+# COMMAND ----------
+
 assert_widget(models, "1. Models")
 assert_widget(output_dir, "2. Output directory")
 
@@ -91,33 +107,3 @@
     notebook_formats = notebook_formats,
     use_threads = use_threads
 )
-
-# COMMAND ----------
-
-# MAGIC %md ### Display exported files
-
-# COMMAND ----------
-
-# MAGIC %sh 
-# MAGIC echo $OUTPUT_DIR
-# MAGIC ls -l $OUTPUT_DIR
-
-# COMMAND ----------
-
-# MAGIC %sh cat $OUTPUT_DIR/manifest.json
-
-# COMMAND ----------
-
-# MAGIC %sh ls -l $OUTPUT_DIR/models
-
-# COMMAND ----------
-
-# MAGIC %sh cat $OUTPUT_DIR/models/models.json
-
-# COMMAND ----------
-
-# MAGIC %sh ls -l $OUTPUT_DIR/experiments
-
-# COMMAND ----------
-
-# MAGIC %sh cat $OUTPUT_DIR/experiments/experiments.json
diff --git a/databricks_notebooks/bulk/Import_Experiments.py b/databricks_notebooks/bulk/Import_Experiments.py
index 035d5858..70fd2b19 100644
--- a/databricks_notebooks/bulk/Import_Experiments.py
+++ b/databricks_notebooks/bulk/Import_Experiments.py
@@ -1,12 +1,12 @@
 # Databricks notebook source
 # MAGIC %md ## Import Experiments
-# MAGIC 
+# MAGIC
 # MAGIC Widgets
 # MAGIC * `1. Input directory` - directory of exported experiments.
 # MAGIC * `2. Import source tags`
 # MAGIC * `3. Experiment rename file` - Experiment rename file.
 # MAGIC * `4. Use threads` - use multi-threaded import.
-# MAGIC 
+# MAGIC
 # MAGIC See https://github.com/mlflow/mlflow-export-import/blob/master/README_bulk.md#Import-experiments.
 
 # COMMAND ----------
@@ -15,6 +15,22 @@
 
 # COMMAND ----------
 
+# DBTITLE 1,set up log file
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y-%m-%d-%H:%M:%S")
+ 
+logfile = f"import_experiments.{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+print("Logging to", logfile)
+
+# COMMAND ----------
+
 dbutils.widgets.text("1. Input directory", "") 
 input_dir = dbutils.widgets.get("1. Input directory")
 input_dir = input_dir.replace("dbfs:","/dbfs")
@@ -40,6 +56,8 @@
 
 # COMMAND ----------
 
+#%%capture captured
+
 from mlflow_export_import.bulk.import_experiments import import_experiments
 
 import_experiments(
@@ -48,3 +66,7 @@
     experiment_renames = experiment_rename_file,
     use_threads = use_threads
 )
+
+# COMMAND ----------
+
+
diff --git a/databricks_notebooks/bulk/Import_Models.py b/databricks_notebooks/bulk/Import_Models.py
index 4f475944..024a4a4e 100644
--- a/databricks_notebooks/bulk/Import_Models.py
+++ b/databricks_notebooks/bulk/Import_Models.py
@@ -1,6 +1,6 @@
 # Databricks notebook source
 # MAGIC %md ## Import Models
-# MAGIC 
+# MAGIC
 # MAGIC Widgets 
 # MAGIC * `1. Input directory` - directory of exported models. 
 # MAGIC * `2. Delete model` - delete the current contents of model
@@ -8,7 +8,7 @@
 # MAGIC * `4. Experiment rename file` - Experiment rename file.
 # MAGIC * `5. Import source tags`
 # MAGIC * `6. Use threads` - use multi-threaded import
-# MAGIC 
+# MAGIC
 # MAGIC See https://github.com/mlflow/mlflow-export-import/blob/master/README_bulk.md#Import-registered-models
 
 # COMMAND ----------
@@ -47,17 +47,35 @@
 
 # COMMAND ----------
 
+# DBTITLE 1,set up log file
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y-%m-%d-%H:%M:%S")
+
+logfile = f"import_models.{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+print("Logging to", logfile)
+
+# COMMAND ----------
+
 assert_widget(input_dir, "1. Input directory")
 
 # COMMAND ----------
 
-from mlflow_export_import.bulk.import_models import import_all
+from mlflow_export_import.bulk.import_models import import_models
 
-import_all(
-    input_dir = input_dir,
-    delete_model = delete_model,
-    import_source_tags = import_source_tags,
-    experiment_renames = experiment_rename_file,
-    model_renames = model_rename_file,
-    use_threads = use_threads
+import_models(
+  input_dir = input_dir,
+  delete_model = delete_model,
+  use_src_user_id = True,
+  verbose=True,
+  import_source_tags = import_source_tags,
+  experiment_renames = experiment_rename_file,
+  model_renames = model_rename_file,
+  use_threads = use_threads
 )
diff --git a/databricks_notebooks/bulk/export_common.py b/databricks_notebooks/bulk/export_common.py
new file mode 100644
index 00000000..a0a004bb
--- /dev/null
+++ b/databricks_notebooks/bulk/export_common.py
@@ -0,0 +1,2 @@
+# Databricks notebook source
+# MAGIC %run ./Common
diff --git a/databricks_notebooks/bulk/import_common.py b/databricks_notebooks/bulk/import_common.py
new file mode 100644
index 00000000..dfb4fc59
--- /dev/null
+++ b/databricks_notebooks/bulk/import_common.py
@@ -0,0 +1,14 @@
+# Databricks notebook source
+# DBTITLE 1,install mlflow-export-import from local
+# MAGIC %pip install ../../../mlflow-export-import --use-feature=in-tree-build
+
+# COMMAND ----------
+
+def assert_widget(value, name):
+    if len(value.rstrip())==0: 
+        raise Exception(f"ERROR: '{name}' widget is required")
+
+# COMMAND ----------
+
+import mlflow
+mlflow_client = mlflow.client.MlflowClient()
diff --git a/mlflow_export_import/bulk/export_experiments.py b/mlflow_export_import/bulk/export_experiments.py
index 671d0c38..e53567ed 100644
--- a/mlflow_export_import/bulk/export_experiments.py
+++ b/mlflow_export_import/bulk/export_experiments.py
@@ -99,7 +99,7 @@ def export_experiments(
     export_results = []
     futures = []
     notebook_formats = utils.string_to_list(notebook_formats)
-    with ThreadPoolExecutor(max_workers=max_workers) as executor:
+    with ThreadPoolExecutor() as executor:
         for exp_id_or_name in experiments:
             run_ids = experiments_dct.get(exp_id_or_name, None)
             future = executor.submit(_export_experiment,
diff --git a/mlflow_export_import/bulk/export_models.py b/mlflow_export_import/bulk/export_models.py
index b74dc1da..2aa123ab 100644
--- a/mlflow_export_import/bulk/export_models.py
+++ b/mlflow_export_import/bulk/export_models.py
@@ -104,7 +104,7 @@ def _export_models(
 
     notebook_formats = utils.string_to_list(notebook_formats),
     futures = []
-    with ThreadPoolExecutor(max_workers=max_workers) as executor:
+    with ThreadPoolExecutor() as executor:
         for model_name in model_names:
             dir = os.path.join(output_dir, model_name)
             future = executor.submit(export_model,
diff --git a/mycode/.gitignore b/mycode/.gitignore
new file mode 100644
index 00000000..4ead4ed0
--- /dev/null
+++ b/mycode/.gitignore
@@ -0,0 +1,5 @@
+export_*
+exported_*
+import_*
+imported_*
+scratch*
\ No newline at end of file
diff --git a/mycode/archive/cli-import-models-v2.py b/mycode/archive/cli-import-models-v2.py
new file mode 100644
index 00000000..8f904941
--- /dev/null
+++ b/mycode/archive/cli-import-models-v2.py
@@ -0,0 +1,102 @@
+# Databricks notebook source
+# MAGIC %md ### Manually loop over models instead of using `import-models`, which is unreliable
+# MAGIC
+# MAGIC TODO: needs work with model and experiment names
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# DBTITLE 1,we can see the s3 mount from %sh :)
+# MAGIC %sh ls /dbfs/mnt/ccidsdatascidatalake/
+
+# COMMAND ----------
+
+# DBTITLE 1,set env vars
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"import_models_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open("/dbfs/FileStore/shared_uploads/darrell.coles@crowncastle.com/aws_databricks_credentials") as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC #echo $DATABRICKS_HOST
+# MAGIC #echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# DBTITLE 0,`export-models` options
+# MAGIC %sh 
+# MAGIC import-model --help
+
+# COMMAND ----------
+
+# DBTITLE 1,make model-experiment file
+import json
+
+input_dir = "/dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models" 
+
+def get_most_recent_experiment_name(model):
+  try:
+    newest = max(model["mlflow"]["registered_model"]["versions"], key=lambda v:v["last_updated_timestamp"])
+    return newest["_experiment_name"]
+  except:
+    return None
+
+with open(f"{input_dir}/models.json") as f:
+  models = json.load(f)
+  model_names = models["info"]["model_names"]
+
+with open("model_experiment","w") as model_experiment:
+  for m in model_names:
+    with open(f"{input_dir}/{m}/model.json") as f:
+      model = json.load(f)
+      experiment_name = get_most_recent_experiment_name(model)
+      if experiment_name:
+        model_experiment.write(f"{m},{experiment_name}\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution
+# MAGIC %sh 
+# MAGIC
+# MAGIC input_dir=/dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models
+# MAGIC
+# MAGIC while IFS=, read model experiment; 
+# MAGIC do 
+# MAGIC   import-model \
+# MAGIC     --input-dir '$input_dir/$model' \
+# MAGIC     --model '$model' \
+# MAGIC     --experiment-name '$experiment' \
+# MAGIC     --delete-model True \
+# MAGIC     --import-source-tags True \
+# MAGIC     --verbose True;
+# MAGIC done < model_experiment
+
+# COMMAND ----------
+
+
diff --git a/mycode/aws-model-hashes b/mycode/aws-model-hashes
new file mode 100644
index 00000000..7630ae31
--- /dev/null
+++ b/mycode/aws-model-hashes
@@ -0,0 +1,69 @@
+alv_regressor: bc08c046ef468213921d1ceaea551d35
+churn-arima: Model file not found
+churn-model: cfae6c5c021a01238fab224158046823
+FDM: cba24da13caf82c694dcc443d2abb6dc
+five-year-alv: 3d1e4337f348a53d54701fff0e1f8b9f
+nearby_location_classifier: Model file not found
+nearby_transmitter_classificiation_1km: cd36bab1df90665e6957ec5dfa6d973e
+pycaret_small_cell_clf: ce16bf9bbcfe0226860c1d77c03bf737
+random-forest-small-cell-classifier: 02c70a5af3406809b34f93a72b2c14f2
+rm_sos_model: b79915de67585a407af1bc188cf59350
+SC_TAM_xgb_ATT_14: Model file not found
+SC_TAM_xgb_CC_ATT_13: c9b1bff151699b358199e8c3f29dd0fd
+SC_TAM_xgb_CC_ATT_14: 2fdfd9d04f790e59e39568ee4bb720b1
+SC_TAM_xgb_CC_ATT_15: 9fb0a58f88c76f442a2ae95d1c5f5572
+SC_TAM_xgb_CC_ATT_16: b126e73954f0147d5db0f2f9f4e5d41e
+SC_TAM_xgb_CC_TMO_13: 86867c9f74a829b49e2615fcb025c4f2
+SC_TAM_xgb_CC_TMO_14: fa57802a4515434b3f477446341476ef
+SC_TAM_xgb_CC_TMO_15: 455b2ff6a2937013bd4b3f7ea57866f9
+SC_TAM_xgb_CC_TMO_16: 898a81cff0b8faa63d7cde8ed306f977
+SC_TAM_xgb_CC_total_13: 0568cd270cb84ac7a5e3ba192b430bfb
+SC_TAM_xgb_CC_total_14: 46be3c00931fcd55320befb4a8a7f69b
+SC_TAM_xgb_CC_total_15: 3073035766c0fb6db6bbd0cfc3928c9d
+SC_TAM_xgb_CC_total_16: 33c815846d6c4d77ee3095b2348707be
+SC_TAM_xgb_CC_VZW_13: 10099377eb81084545b7035707bc09c7
+SC_TAM_xgb_CC_VZW_14: 09dd59a9721327c9e8d7a8c1b2929b9d
+SC_TAM_xgb_CC_VZW_15: e38cfe16baeef765596bb0130d3d6061
+SC_TAM_xgb_CC_VZW_16: 1cb7e9c8ae4535979b2d47a47999cad8
+SC_TAM_xgb_TMO_14: Model file not found
+SC_TAM_xgb_total_14: Model file not found
+SC_TAM_xgb_VZW_14: Model file not found
+small_cell_classifier: e88cbd08d01e0bbe8e5ff6608541455a
+small_cell_classifier_fs: Model file not found
+ta_step_size_classifier: afaf962b88719e786d59597e0553921c
+TAM v1 MVP: Model file not found
+tam_RF_CC_12: 4eaa75365903f7a30946194a9b3cb94c
+tam_RF_CC_13: c8d6a83d36cb4d28eb9f83dd829c99b3
+tam_RF_CC_14: 67508cac03d341de200abaa958b77a31
+tam_RF_CC_15: a0ac24c68a95df7687ee3d28eeabea36
+tam_RF_CC_16: a842437542bb380e02f6b9fa1922e2f4
+tam_RF_CC_17: 524709a93e2063b113343085f05e1091
+tamByMNO_RF_CC_ATT_12: 4f8bc9bd6f0dbda8bbf0d6838d8ac5ad
+tamByMNO_RF_CC_ATT_13: e11fe2bbd705870bca8b8d793c83327f
+tamByMNO_RF_CC_ATT_14: 25fbcb74d2d3bfd55339874713dd7541
+tamByMNO_RF_CC_ATT_15: d9ead39efbea06ac569c8477b6840fdf
+tamByMNO_RF_CC_ATT_16: 408133539c2cbc02bbdfce3eae38923c
+tamByMNO_RF_CC_ATT_17: c99b6dfa4e66de57faa9a32d128ee6ee
+tamByMNO_RF_CC_TMO_13: b5c04f4f050f1a3640ca9963cb469db6
+tamByMNO_RF_CC_TMO_14: cb5fe7361fc17881ea60bfa0fd23060b
+tamByMNO_RF_CC_TMO_15: 8bed2612506e7be6c7da54fde3dd5017
+tamByMNO_RF_CC_TMO_16: 18e3849545c621bb83203cfe00f6cf92
+tamByMNO_RF_CC_TMO_17: e27df7ef79937a2b779e194adbd1413e
+tamByMNO_RF_CC_VZW_13: 6c913d63f1420c10ebd15662a7f68b1d
+tamByMNO_RF_CC_VZW_14: da88b8c761d0c5b33d20b535c3d3ae0b
+tamByMNO_RF_CC_VZW_15: 30ffdb42cb9f0e50f9a176fa56649a65
+tamByMNO_RF_CC_VZW_16: 508aeed31329c6bea3f8e6c76334987c
+tamByMNO_RF_CC_VZW_17: 433e554237d605ae97f39bfce6e66f66
+timing-advance-step-size-ternary-classifier: b1b0b19abd834a86b582253577acd646
+tsa-gt-reg: aeb0e6623cadfd936b781b9bb4da39e3
+tsa-mono-reg: 1503821b4c9689fde4726471b77989aa
+tsa-sst-reg: b8aecdddfd906087b5335b5b9f96a9a3
+wireless_throughput_model_AT&T: b443b41808ce9eaa6a4eb566f0093302
+wireless_throughput_model_AT&T_log: 6db791d5c8b1b5b7b8e8ee396d37cd4e
+wireless_throughput_model_T-Mobile: 8650f744e0440ae9474978948505ac50
+wireless_throughput_model_T-Mobile_log: 09f16d64726d3d721240eda91271ef07
+wireless_throughput_model_Verizon: e6311060a594a4250e9bfaba97688695
+wireless_throughput_model_Verizon_log: 3023cc6114e2f16cdf89a2c1b22d99d1
+wireless_throughput_Verizon: Model file not found
+ws_activity_classifier: c159a33ed4788b4e5925befe8da66efe
+xgboost_dev: Model file not found
\ No newline at end of file
diff --git a/mycode/aws-model-hashes-uc b/mycode/aws-model-hashes-uc
new file mode 100644
index 00000000..b7257b6c
--- /dev/null
+++ b/mycode/aws-model-hashes-uc
@@ -0,0 +1,68 @@
+ds_nonprod.migrated_models.wireless_throughput_model_verizon_log: 3023cc6114e2f16cdf89a2c1b22d99d1
+ds_nonprod.migrated_models.tam_rf_cc_15: a0ac24c68a95df7687ee3d28eeabea36
+ds_nonprod.migrated_models.wireless_throughput_verizon: Model not found
+ds_nonprod.migrated_models.churn-arima: Model not found
+ds_nonprod.migrated_models.tam_rf_cc_12: 4eaa75365903f7a30946194a9b3cb94c
+ds_nonprod.migrated_models.tambymno_rf_cc_vzw_16: 508aeed31329c6bea3f8e6c76334987c
+ds_nonprod.migrated_models.sc_tam_xgb_cc_tmo_16: 898a81cff0b8faa63d7cde8ed306f977
+ds_nonprod.migrated_models.ta_step_size_classifier: afaf962b88719e786d59597e0553921c
+ds_nonprod.migrated_models.ws_activity_classifier: c159a33ed4788b4e5925befe8da66efe
+ds_nonprod.migrated_models.random-forest-small-cell-classifier: 02c70a5af3406809b34f93a72b2c14f2
+ds_nonprod.migrated_models.tambymno_rf_cc_vzw_15: 30ffdb42cb9f0e50f9a176fa56649a65
+ds_nonprod.migrated_models.tam_rf_cc_17: 524709a93e2063b113343085f05e1091
+ds_nonprod.migrated_models.tambymno_rf_cc_tmo_15: 8bed2612506e7be6c7da54fde3dd5017
+ds_nonprod.migrated_models.sc_tam_xgb_total_14: Model not found
+ds_nonprod.migrated_models.tambymno_rf_cc_att_15: d9ead39efbea06ac569c8477b6840fdf
+ds_nonprod.migrated_models.tam_rf_cc_13: c8d6a83d36cb4d28eb9f83dd829c99b3
+ds_nonprod.migrated_models.wireless_throughput_model_at&t: b443b41808ce9eaa6a4eb566f0093302
+ds_nonprod.migrated_models.five-year-alv: 3d1e4337f348a53d54701fff0e1f8b9f
+ds_nonprod.migrated_models.sc_tam_xgb_cc_vzw_14: 09dd59a9721327c9e8d7a8c1b2929b9d
+ds_nonprod.migrated_models.sc_tam_xgb_cc_tmo_15: 455b2ff6a2937013bd4b3f7ea57866f9
+ds_nonprod.migrated_models.tam_rf_cc_14: 67508cac03d341de200abaa958b77a31
+ds_nonprod.migrated_models.tambymno_rf_cc_tmo_16: 18e3849545c621bb83203cfe00f6cf92
+ds_nonprod.migrated_models.sc_tam_xgb_cc_tmo_14: fa57802a4515434b3f477446341476ef
+ds_nonprod.migrated_models.sc_tam_xgb_cc_vzw_15: e38cfe16baeef765596bb0130d3d6061
+ds_nonprod.migrated_models.sc_tam_xgb_cc_total_14: 46be3c00931fcd55320befb4a8a7f69b
+ds_nonprod.migrated_models.sc_tam_xgb_vzw_14: Model not found
+ds_nonprod.migrated_models.tambymno_rf_cc_att_12: 4f8bc9bd6f0dbda8bbf0d6838d8ac5ad
+ds_nonprod.migrated_models.sc_tam_xgb_cc_att_15: 9fb0a58f88c76f442a2ae95d1c5f5572
+ds_nonprod.migrated_models.tambymno_rf_cc_tmo_17: e27df7ef79937a2b779e194adbd1413e
+ds_nonprod.migrated_models.tambymno_rf_cc_att_14: 25fbcb74d2d3bfd55339874713dd7541
+ds_nonprod.migrated_models.tsa-mono-reg: Model not found
+ds_nonprod.migrated_models.wireless_throughput_model_at&t_log: 6db791d5c8b1b5b7b8e8ee396d37cd4e
+ds_nonprod.migrated_models.tambymno_rf_cc_att_17: c99b6dfa4e66de57faa9a32d128ee6ee
+ds_nonprod.migrated_models.xgboost_dev: Model not found
+ds_nonprod.migrated_models.sc_tam_xgb_cc_total_16: 33c815846d6c4d77ee3095b2348707be
+ds_nonprod.migrated_models.rm_sos_model: b79915de67585a407af1bc188cf59350
+ds_nonprod.migrated_models.tambymno_rf_cc_vzw_14: da88b8c761d0c5b33d20b535c3d3ae0b
+ds_nonprod.migrated_models.sc_tam_xgb_cc_total_15: 3073035766c0fb6db6bbd0cfc3928c9d
+ds_nonprod.migrated_models.nearby_location_classifier: Model not found
+ds_nonprod.migrated_models.small_cell_classifier_fs: Model not found
+ds_nonprod.migrated_models.sc_tam_xgb_cc_att_16: b126e73954f0147d5db0f2f9f4e5d41e
+ds_nonprod.migrated_models.sc_tam_xgb_cc_total_13: 0568cd270cb84ac7a5e3ba192b430bfb
+ds_nonprod.migrated_models.sc_tam_xgb_cc_att_14: 2fdfd9d04f790e59e39568ee4bb720b1
+ds_nonprod.migrated_models.tambymno_rf_cc_tmo_14: cb5fe7361fc17881ea60bfa0fd23060b
+ds_nonprod.migrated_models.timing-advance-step-size-ternary-classifier: b1b0b19abd834a86b582253577acd646
+ds_nonprod.migrated_models.tsa-sst-reg: Model not found
+ds_nonprod.migrated_models.tambymno_rf_cc_att_16: 408133539c2cbc02bbdfce3eae38923c
+ds_nonprod.migrated_models.sc_tam_xgb_cc_vzw_16: 1cb7e9c8ae4535979b2d47a47999cad8
+ds_nonprod.migrated_models.tambymno_rf_cc_att_13: e11fe2bbd705870bca8b8d793c83327f
+ds_nonprod.migrated_models.sc_tam_xgb_tmo_14: Model not found
+ds_nonprod.migrated_models.tambymno_rf_cc_vzw_13: 6c913d63f1420c10ebd15662a7f68b1d
+ds_nonprod.migrated_models.tsa-gt-reg: Model not found
+ds_nonprod.migrated_models.fdm: cba24da13caf82c694dcc443d2abb6dc
+ds_nonprod.migrated_models.nearby_transmitter_classificiation_1km: cd36bab1df90665e6957ec5dfa6d973e
+ds_nonprod.migrated_models.tambymno_rf_cc_tmo_13: b5c04f4f050f1a3640ca9963cb469db6
+ds_nonprod.migrated_models.wireless_throughput_model_t-mobile_log: 09f16d64726d3d721240eda91271ef07
+ds_nonprod.migrated_models.sc_tam_xgb_cc_tmo_13: 86867c9f74a829b49e2615fcb025c4f2
+ds_nonprod.migrated_models.wireless_throughput_model_verizon: Model not found
+ds_nonprod.migrated_models.tambymno_rf_cc_vzw_17: 433e554237d605ae97f39bfce6e66f66
+ds_nonprod.migrated_models.alv_regressor: bc08c046ef468213921d1ceaea551d35
+ds_nonprod.migrated_models.sc_tam_xgb_cc_att_13: c9b1bff151699b358199e8c3f29dd0fd
+ds_nonprod.migrated_models.churn-model: Model not found
+ds_nonprod.migrated_models.sc_tam_xgb_att_14: Model not found
+ds_nonprod.migrated_models.wireless_throughput_model_t-mobile: 8650f744e0440ae9474978948505ac50
+ds_nonprod.migrated_models.sc_tam_xgb_cc_vzw_13: 10099377eb81084545b7035707bc09c7
+ds_nonprod.migrated_models.tam_rf_cc_16: a842437542bb380e02f6b9fa1922e2f4
+ds_nonprod.migrated_models.pycaret_small_cell_clf: Model not found
+ds_nonprod.migrated_models.small_cell_classifier: 375a1ac8c163d8e70980abc6ab01292a
\ No newline at end of file
diff --git a/mycode/azure-model-hashes b/mycode/azure-model-hashes
new file mode 100644
index 00000000..4408ba4d
--- /dev/null
+++ b/mycode/azure-model-hashes
@@ -0,0 +1,70 @@
+alv_regressor: bc08c046ef468213921d1ceaea551d35
+churn-arima: Model file not found
+churn-model: cfae6c5c021a01238fab224158046823
+FDM: cba24da13caf82c694dcc443d2abb6dc
+five-year-alv: 3d1e4337f348a53d54701fff0e1f8b9f
+nearby_location_classifier: Model file not found
+nearby_transmitter_classificiation_1km: 4f7014a660f5dd7b0bdf519a1cc8733e
+pycaret_small_cell_clf: ce16bf9bbcfe0226860c1d77c03bf737
+random-forest-small-cell-classifier: 02c70a5af3406809b34f93a72b2c14f2
+rfds_vgg: Model file not found
+rm_sos_model: b79915de67585a407af1bc188cf59350
+SC_TAM_xgb_ATT_14: Model file not found
+SC_TAM_xgb_CC_ATT_13: c9b1bff151699b358199e8c3f29dd0fd
+SC_TAM_xgb_CC_ATT_14: e780e65b67f3d5f8ee10d33c26bca286
+SC_TAM_xgb_CC_ATT_15: 9fb0a58f88c76f442a2ae95d1c5f5572
+SC_TAM_xgb_CC_ATT_16: 582919c7b5a1ec7006ca740e41eb1c8a
+SC_TAM_xgb_CC_TMO_13: 86867c9f74a829b49e2615fcb025c4f2
+SC_TAM_xgb_CC_TMO_14: a91791267a14d95e94572c37ba6c4925
+SC_TAM_xgb_CC_TMO_15: 455b2ff6a2937013bd4b3f7ea57866f9
+SC_TAM_xgb_CC_TMO_16: d2337a992ab0bf48f38f5d6b608c0a49
+SC_TAM_xgb_CC_total_13: 0568cd270cb84ac7a5e3ba192b430bfb
+SC_TAM_xgb_CC_total_14: 46be3c00931fcd55320befb4a8a7f69b
+SC_TAM_xgb_CC_total_15: 3073035766c0fb6db6bbd0cfc3928c9d
+SC_TAM_xgb_CC_total_16: 33c815846d6c4d77ee3095b2348707be
+SC_TAM_xgb_CC_VZW_13: 5bcd17f601967212a59132710d116b72
+SC_TAM_xgb_CC_VZW_14: 3bc0d66861a84ab7a66b20588cacfcf4
+SC_TAM_xgb_CC_VZW_15: e38cfe16baeef765596bb0130d3d6061
+SC_TAM_xgb_CC_VZW_16: a4bdc13587d437ed12dd3021e506ad16
+SC_TAM_xgb_TMO_14: Model file not found
+SC_TAM_xgb_total_14: Model file not found
+SC_TAM_xgb_VZW_14: Model file not found
+small_cell_classifier: e88cbd08d01e0bbe8e5ff6608541455a
+small_cell_classifier_fs: Model file not found
+ta_step_size_classifier: afaf962b88719e786d59597e0553921c
+TAM v1 MVP: Model file not found
+tam_RF_CC_12: 4eaa75365903f7a30946194a9b3cb94c
+tam_RF_CC_13: c8d6a83d36cb4d28eb9f83dd829c99b3
+tam_RF_CC_14: 67508cac03d341de200abaa958b77a31
+tam_RF_CC_15: a0ac24c68a95df7687ee3d28eeabea36
+tam_RF_CC_16: a842437542bb380e02f6b9fa1922e2f4
+tam_RF_CC_17: 524709a93e2063b113343085f05e1091
+tamByMNO_RF_CC_ATT_12: 4f8bc9bd6f0dbda8bbf0d6838d8ac5ad
+tamByMNO_RF_CC_ATT_13: d90498883b237ef6b71932ebf3dfee75
+tamByMNO_RF_CC_ATT_14: 25fbcb74d2d3bfd55339874713dd7541
+tamByMNO_RF_CC_ATT_15: d9ead39efbea06ac569c8477b6840fdf
+tamByMNO_RF_CC_ATT_16: 38d12e47d3a9774b73749c3f31460521
+tamByMNO_RF_CC_ATT_17: c99b6dfa4e66de57faa9a32d128ee6ee
+tamByMNO_RF_CC_TMO_13: c7c0fa3532104107db03b4cc46da53a3
+tamByMNO_RF_CC_TMO_14: cb5fe7361fc17881ea60bfa0fd23060b
+tamByMNO_RF_CC_TMO_15: 8bed2612506e7be6c7da54fde3dd5017
+tamByMNO_RF_CC_TMO_16: 18e3849545c621bb83203cfe00f6cf92
+tamByMNO_RF_CC_TMO_17: e27df7ef79937a2b779e194adbd1413e
+tamByMNO_RF_CC_VZW_13: 33f8b87ffcfc246912cff23b9c46929e
+tamByMNO_RF_CC_VZW_14: da88b8c761d0c5b33d20b535c3d3ae0b
+tamByMNO_RF_CC_VZW_15: 30ffdb42cb9f0e50f9a176fa56649a65
+tamByMNO_RF_CC_VZW_16: 508aeed31329c6bea3f8e6c76334987c
+tamByMNO_RF_CC_VZW_17: 433e554237d605ae97f39bfce6e66f66
+timing-advance-step-size-ternary-classifier: b1b0b19abd834a86b582253577acd646
+tsa-gt-reg: aeb0e6623cadfd936b781b9bb4da39e3
+tsa-mono-reg: 1503821b4c9689fde4726471b77989aa
+tsa-sst-reg: b8aecdddfd906087b5335b5b9f96a9a3
+wireless_throughput_model_AT&T: 99c32a9286879b6057b75e60384c30ca
+wireless_throughput_model_AT&T_log: 6db791d5c8b1b5b7b8e8ee396d37cd4e
+wireless_throughput_model_T-Mobile: e6e83b20667bd5726ba39c183ece3527
+wireless_throughput_model_T-Mobile_log: 09f16d64726d3d721240eda91271ef07
+wireless_throughput_model_Verizon: e6311060a594a4250e9bfaba97688695
+wireless_throughput_model_Verizon_log: 3023cc6114e2f16cdf89a2c1b22d99d1
+wireless_throughput_Verizon: Model file not found
+ws_activity_classifier: c159a33ed4788b4e5925befe8da66efe
+xgboost_dev: Model file not found
\ No newline at end of file
diff --git a/mycode/cli-export-experiments.py b/mycode/cli-export-experiments.py
new file mode 100644
index 00000000..cd03bb7f
--- /dev/null
+++ b/mycode/cli-export-experiments.py
@@ -0,0 +1,78 @@
+# Databricks notebook source
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# DBTITLE 1,we can see the s3 mount from %sh :)
+# MAGIC %sh ls /dbfs/mnt/aws-ds-non-prod
+
+# COMMAND ----------
+
+# MAGIC %run ./credentials
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.dropdown("platform","",["", "azure", "aws"])
+platform = dbutils.widgets.get("platform")
+
+credentials_path = get_credentials_path(platform)
+
+# COMMAND ----------
+
+# DBTITLE 1,set env vars
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"export_experiments_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open(credentials_path) as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC # echo $DATABRICKS_HOST
+# MAGIC # echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# MAGIC %md ## Execute
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution | all
+# MAGIC %sh 
+# MAGIC export-experiments \
+# MAGIC   --output-dir /dbfs/mnt/aws-ds-non-prod/mlflow-migration-experiments/experiments \
+# MAGIC   --experiments all \
+# MAGIC   --run-start-time 2024-11-09 \
+# MAGIC   --export-permissions True \
+# MAGIC   --notebook-formats SOURCE \
+# MAGIC   --use-threads True
diff --git a/mycode/cli-export-models.py b/mycode/cli-export-models.py
new file mode 100644
index 00000000..a863d799
--- /dev/null
+++ b/mycode/cli-export-models.py
@@ -0,0 +1,90 @@
+# Databricks notebook source
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# MAGIC %run ./credentials
+
+# COMMAND ----------
+
+# DBTITLE 1,we can see the s3 mount from %sh :)
+# MAGIC %sh ls /dbfs/mnt/aws-ds-non-prod
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+# MAGIC Make sure the mount to the target s3 bucket is set up. See the **mount** notebook in this directory
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.dropdown("platform","",["", "azure", "aws"])
+platform = dbutils.widgets.get("platform")
+
+credentials_path = get_credentials_path(platform)
+
+# COMMAND ----------
+
+# DBTITLE 1,set env vars
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"export_models_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open(credentials_path) as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC # echo $DATABRICKS_HOST
+# MAGIC # echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# DBTITLE 0,`export-models` options
+# MAGIC %sh 
+# MAGIC export-models --help
+
+# COMMAND ----------
+
+# MAGIC %md ## Execute
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution
+# MAGIC %sh 
+# MAGIC export-models \
+# MAGIC   --output-dir /dbfs/mnt/aws-ds-non-prod/mlflow-migration-models-01 \
+# MAGIC   --models all \
+# MAGIC   --stages 'Production' \
+# MAGIC   --export-permissions True \
+# MAGIC   --notebook-formats SOURCE \
+# MAGIC   --export-version-model True \
+# MAGIC   --use-threads True
+
+# COMMAND ----------
+
+
diff --git a/mycode/cli-import-experiments.py b/mycode/cli-import-experiments.py
new file mode 100644
index 00000000..b549676f
--- /dev/null
+++ b/mycode/cli-import-experiments.py
@@ -0,0 +1,90 @@
+# Databricks notebook source
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# MAGIC %run ./credentials
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+
+# COMMAND ----------
+
+# DBTITLE 1,we can see the s3 mount from %sh :)
+# MAGIC %fs ls /mnt/datalake
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.dropdown("platform","",["", "azure", "aws"])
+platform = dbutils.widgets.get("platform")
+
+credentials_path = get_credentials_path(platform)
+
+# COMMAND ----------
+
+# DBTITLE 1,set env vars
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"import_experiments_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open(credentials_path) as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC # echo $DATABRICKS_HOST
+# MAGIC # echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# MAGIC %md ## Execution
+
+# COMMAND ----------
+
+# DBTITLE 0,`export-models` options
+# MAGIC %sh 
+# MAGIC import-experiments --help
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution
+# MAGIC %sh 
+# MAGIC import-experiments \
+# MAGIC   --input-dir /dbfs/mnt/datalake/mlflow-migration-02/experiments \
+# MAGIC   --import-source-tags True \
+# MAGIC   --import-permissions True \
+# MAGIC   --use-threads True
+
+# COMMAND ----------
+
+# MAGIC %sh ls /dbfs/mnt/datalake
+
+# COMMAND ----------
+
+
diff --git a/mycode/cli-import-models.py b/mycode/cli-import-models.py
new file mode 100644
index 00000000..69bc8450
--- /dev/null
+++ b/mycode/cli-import-models.py
@@ -0,0 +1,184 @@
+# Databricks notebook source
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# DBTITLE 1,we can see the s3 mount from %sh :)
+# MAGIC %sh ls /dbfs/mnt/aws-ds-non-prod
+
+# COMMAND ----------
+
+# MAGIC %run ./credentials
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+
+# COMMAND ----------
+
+dbutils.widgets.dropdown("platform","",["", "azure", "aws"])
+platform = dbutils.widgets.get("platform")
+
+credentials_path = get_credentials_path(platform)
+
+# COMMAND ----------
+
+# DBTITLE 1,set env vars
+import os 
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"import_models_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open(credentials_path) as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC echo $DATABRICKS_HOST
+# MAGIC echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# DBTITLE 0,`export-models` options
+# MAGIC %sh 
+# MAGIC import-models --help
+
+# COMMAND ----------
+
+# MAGIC %md ## Execution
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution
+# MAGIC %sh 
+# MAGIC import-models \
+# MAGIC   --input-dir /dbfs/mnt/aws-ds-non-prod/mlflow-migration-models-01 \
+# MAGIC   --delete-model True \
+# MAGIC   --import-permissions True \
+# MAGIC   --import-source-tags True \
+# MAGIC   --verbose True
+# MAGIC   --use-threads False
+
+# COMMAND ----------
+
+# MAGIC %md ## Appendices
+
+# COMMAND ----------
+
+# DBTITLE 1,count models in workspace model registry
+import mlflow
+len(mlflow.search_registered_models())
+
+# COMMAND ----------
+
+# MAGIC %sh
+# MAGIC import-model --help
+
+# COMMAND ----------
+
+# MAGIC %sh 
+# MAGIC import-model \
+# MAGIC   --input-dir /dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models/pycaret_small_cell_clf \
+# MAGIC   --model pycaret_small_cell_clf \
+# MAGIC   --experiment-name /Users/darrell.coles@crowncastle.com/databricks_automl/pycaret-sc-clf-test-1 \
+# MAGIC   --delete-model True \
+# MAGIC   # --import-permissions True \
+# MAGIC   --import-source-tags True \
+# MAGIC   --verbose True
+
+# COMMAND ----------
+
+# MAGIC %sh cat /dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models/models.json
+
+# COMMAND ----------
+
+def get_most_recent_experiment_name(model):
+  try:
+    newest = max(model["mlflow"]["registered_model"]["versions"], key=lambda v:v["last_updated_timestamp"])
+    return newest["_experiment_name"]
+  except:
+    return None
+
+# COMMAND ----------
+
+import json
+
+input_dir = "/dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models" 
+
+with open(f"{input_dir}/models.json") as f:
+  models = json.load(f)
+  model_names = models["info"]["model_names"]
+
+with open("model_experiment","w") as model_experiment:
+  for m in model_names:
+    with open(f"{input_dir}/{m}/model.json") as f:
+      model = json.load(f)
+      experiment_name = get_most_recent_experiment_name(model)
+      if experiment_name:
+        model_experiment.write(f"{m},{experiment_name}\n")
+
+# COMMAND ----------
+
+# MAGIC %sh cat model_experiment
+
+# COMMAND ----------
+
+# MAGIC %sh 
+# MAGIC
+# MAGIC input_dir=/dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/models
+# MAGIC
+# MAGIC while IFS=, read model experiment; 
+# MAGIC do 
+# MAGIC   import-model \
+# MAGIC     --input-dir $input_dir/$model \
+# MAGIC     --model $model \
+# MAGIC     --experiment-name $experiment_name \
+# MAGIC     --delete-model True \
+# MAGIC     --import-source-tags True \
+# MAGIC     --verbose True;
+# MAGIC done < model_experiment
+
+# COMMAND ----------
+
+# MAGIC %sh
+# MAGIC while IFS=, read field1 field2;
+# MAGIC   do 
+# MAGIC     echo model: $field1
+# MAGIC     echo experiment: $field2
+# MAGIC done < model_experiment
+
+# COMMAND ----------
+
+import os
+os.environ
+
+# COMMAND ----------
+
+# MAGIC %sh $elas
+
+# COMMAND ----------
+
+
diff --git a/mycode/compare-model-hashes-uc.py b/mycode/compare-model-hashes-uc.py
new file mode 100644
index 00000000..a6e2107b
--- /dev/null
+++ b/mycode/compare-model-hashes-uc.py
@@ -0,0 +1,60 @@
+# Databricks notebook source
+# DBTITLE 1,azure model hashes
+with open("azure-model-hashes", "r") as f:
+  lines = f.readlines()
+
+az_model_hashes = dict()
+for line in lines:
+  k, v = line.lower().strip("\n").split(": ")
+  if v=="model file not found":
+    v=None
+  az_model_hashes[k] = v
+
+# COMMAND ----------
+
+# DBTITLE 1,aws model hashes
+with open("aws-model-hashes-uc", "r") as f:
+  lines = f.readlines()
+
+aws_model_hashes = dict()
+for line in lines:
+  k, v = line.lower().replace("ds_nonprod.migrated_models.","").strip("\n").split(": ")
+  if v=='model not found': 
+    v=None
+  aws_model_hashes[k] = v
+
+# COMMAND ----------
+
+aws_model_hashes
+
+# COMMAND ----------
+
+# DBTITLE 1,num registered models on each platform
+len(az_model_hashes), len(aws_model_hashes)
+
+# COMMAND ----------
+
+# DBTITLE 1,missing models
+set(az_model_hashes.keys()) - set(aws_model_hashes.keys()), set(aws_model_hashes.keys()) - set(az_model_hashes.keys()) 
+
+# COMMAND ----------
+
+import pandas as pd
+
+azure_df = pd.DataFrame(az_model_hashes, index=["azure"])
+aws_df = pd.DataFrame(aws_model_hashes, index=["aws"])
+model_hashes_pdf = pd.concat((azure_df, aws_df)).T
+
+model_hashes_pdf["match"] = model_hashes_pdf.azure == model_hashes_pdf.aws
+
+model_hashes = spark.createDataFrame(model_hashes_pdf.reset_index())
+
+display(model_hashes)
+
+# COMMAND ----------
+
+display(model_hashes.where("not match and azure is not null and aws is not null"))
+
+# COMMAND ----------
+
+
diff --git a/mycode/compare-model-hashes.py b/mycode/compare-model-hashes.py
new file mode 100644
index 00000000..bb4c21cc
--- /dev/null
+++ b/mycode/compare-model-hashes.py
@@ -0,0 +1,52 @@
+# Databricks notebook source
+# DBTITLE 1,azure model hashes
+with open("azure-model-hashes", "r") as f:
+  lines = f.readlines()
+
+az_model_hashes = dict()
+for line in lines:
+  k, v = line.strip("\n").split(": ")
+  az_model_hashes[k] = v
+
+# COMMAND ----------
+
+# DBTITLE 1,aws model hashes
+with open("aws-model-hashes", "r") as f:
+  lines = f.readlines()
+
+aws_model_hashes = dict()
+for line in lines:
+  k, v = line.strip("\n").split(": ")
+  aws_model_hashes[k] = v
+
+# COMMAND ----------
+
+# DBTITLE 1,num registered models on each platform
+len(az_model_hashes), len(aws_model_hashes)
+
+# COMMAND ----------
+
+# DBTITLE 1,missing models
+set(az_model_hashes.keys()) - set(aws_model_hashes.keys()), set(aws_model_hashes.keys()) - set(az_model_hashes.keys()) 
+
+# COMMAND ----------
+
+import pandas as pd
+
+azure_df = pd.DataFrame(az_model_hashes, index=["azure"])
+aws_df = pd.DataFrame(aws_model_hashes, index=["aws"])
+model_hashes_pdf = pd.concat((azure_df, aws_df)).T
+
+model_hashes_pdf["match"] = model_hashes_pdf.azure == model_hashes_pdf.aws
+
+model_hashes = spark.createDataFrame(model_hashes_pdf.reset_index())
+
+display(model_hashes)
+
+# COMMAND ----------
+
+display(model_hashes.where("not match"))
+
+# COMMAND ----------
+
+
diff --git a/mycode/count-exported-mlflow-objects.py b/mycode/count-exported-mlflow-objects.py
new file mode 100644
index 00000000..a12a6ac3
--- /dev/null
+++ b/mycode/count-exported-mlflow-objects.py
@@ -0,0 +1,98 @@
+# Databricks notebook source
+import re
+import pandas as pd
+import numpy as np
+
+# COMMAND ----------
+
+# MAGIC %md ## Experiments
+
+# COMMAND ----------
+
+# MAGIC %sh 
+# MAGIC find /dbfs/mnt/ccidsdatascidatalake/mlflow-migration-models/experiments -type d -maxdepth 2 > exported_experiments
+
+# COMMAND ----------
+
+exp_pat = "^.*\/dbfs\/mnt\/datalake\/mlflow-migration-models\/experiments\/(.*)$"
+
+filepath = "exported_experiments"
+capture = [re.findall(exp_pat, line) for line in open(filepath)]
+
+# COMMAND ----------
+
+def parse_exp_run(s):
+  res = s.split("/")
+  return res if len(res)==2 else [*res,None]
+
+parsed = np.array([parse_exp_run(c[0]) for c in capture if c])
+
+df_parsed = pd.DataFrame(dict(experiment_id=parsed[:,0], run_id=parsed[:,1]))
+
+exp_df = df_parsed.groupby("experiment_id").run_id.count().to_frame()
+exp_df.columns = ["n_runs"]
+
+# COMMAND ----------
+
+# MAGIC %md ## Models
+
+# COMMAND ----------
+
+# MAGIC %sh 
+# MAGIC ls /dbfs/mnt/datalake/mlflow-migration-models/models/ > exported_models
+
+# COMMAND ----------
+
+with open("exported_models") as f:
+  n_models = len(f.readlines())
+
+# COMMAND ----------
+
+# MAGIC %md ## Report
+
+# COMMAND ----------
+
+print("There are", exp_df.shape[0], "total experiments")
+print("There are", exp_df.query("n_runs > 0").shape[0], "experiments with at least one run")
+print("There are", exp_df.query("n_runs == 0").shape[0], "experiments with no runs")
+print("There are", exp_df.n_runs.sum(), "runs")
+print("There are", n_models, "registered models")
+
+# COMMAND ----------
+
+# import os 
+
+# def exp_id(exp):
+#   try:
+#     return exp.experiment_id
+#   except:
+#     return "none"
+  
+# def exp_id_after_import(exp):
+#   try:
+#     return exp.experiment_id
+#   except:
+#     return "none"
+
+# def email(exp):
+#   try:
+#     return exp.tags['mlflow.ownerEmail']
+#   except:
+#     return "none"
+  
+# def n_runs(exp):
+#   try:
+#     return len(mlflow.search_runs(exp_id(exp)))
+#   except:
+#     0
+    
+# result = []
+# for exp in mlflow.search_experiments():
+#   result.append([os.path.basename(exp.name), email(exp), n_runs(exp)])
+  
+# df = spark.createDataFrame(result,["experiment_name","owner","n_runs"])
+# display(df)
+
+# COMMAND ----------
+
+
diff --git a/mycode/count-legacy-production-models.py b/mycode/count-legacy-production-models.py
new file mode 100644
index 00000000..72d82334
--- /dev/null
+++ b/mycode/count-legacy-production-models.py
@@ -0,0 +1,28 @@
+# Databricks notebook source
+import mlflow 
+from mlflow import MlflowClient
+client = MlflowClient()
+
+# COMMAND ----------
+
+# DBTITLE 1,count all models
+len(client.search_registered_models())
+
+# COMMAND ----------
+
+# DBTITLE 1,count models with prod stage
+count = 0
+
+for registered_model in mlflow.search_registered_models():
+  has_prod_stg = False
+  if "Production" in [version.current_stage for version in registered_model.latest_versions]:
+    has_prod_stg = True
+    count +=1
+  if not has_prod_stg:
+    print(registered_model.name, "DOES NOT HAVE a Production stage")
+
+print("\n", count, "models have Production stages all together")
+
+# COMMAND ----------
+
+
diff --git a/mycode/count-objects-in-mlflow.py b/mycode/count-objects-in-mlflow.py
new file mode 100644
index 00000000..356bd7d2
--- /dev/null
+++ b/mycode/count-objects-in-mlflow.py
@@ -0,0 +1,90 @@
+# Databricks notebook source
+import mlflow
+import pandas as pd
+import numpy as np
+from datetime import datetime
+
+# COMMAND ----------
+
+wij = dbutils.widgets
+
+wij.text("start_time","")
+start_time = wij.get("start_time")
+start_time = start_time if start_time else None
+
+print("start_time:", start_time)
+
+# COMMAND ----------
+
+# DBTITLE 1,inspection functions
+def exp_id(exp):
+  try:
+    return exp.experiment_id
+  except:
+    return "none"
+  
+def exp_id_after_import(exp):
+  try:
+    return exp.experiment_id
+  except:
+    return "none"
+
+def email(exp):
+  try:
+    return exp.tags['mlflow.ownerEmail']
+  except:
+    return "none"
+    
+def run_start_time(run):
+  """
+  Parameters
+  ==========
+    run (pandas.Series): run parameters
+  
+  Returns
+  =======
+    datetime
+  """
+  return datetime(
+    year=run.start_time.year,
+    month=run.start_time.month,
+    day=run.start_time.day)
+  
+def n_runs(exp, start_time=None):
+  """
+  Parameters
+  ==========
+  start_time (str): date; format 'YYYY-MM-DD'  
+  """
+  runs = mlflow.search_runs(exp.experiment_id)
+  try:
+    if start_time:
+      year, month, day = start_time.split("-")
+      dt_thresh = datetime(year=int(year), month=int(month), day=int(day))
+      return len([1 for _, run in runs.iterrows() if run_start_time(run) >= dt_thresh])
+    else:
+      return len(runs)
+  except:
+    0
+
+# COMMAND ----------
+
+# DBTITLE 1,experiment stats
+data = np.array([(exp_id(exp), n_runs(exp, start_time), email(exp)) for exp in mlflow.search_experiments()])
+
+df = pd.DataFrame(dict(experiment_id=data[:,0], n_runs=data[:,1], owner_email=data[:,2]))
+df["n_runs"] = df.n_runs.astype(int)
+
+print("There are", df.shape[0], "total experiments")
+print("There are", df.query("n_runs > 0").shape[0], "experiments with at least one run")
+print("There are", df.query("n_runs == 0").shape[0], "experiments with no runs")
+print("There are", df.n_runs.sum(), "runs")
+
+# COMMAND ----------
+
+# DBTITLE 1,model stats
+print("There are", len(mlflow.search_registered_models()), "registered models")
+
+# COMMAND ----------
+
+
diff --git a/mycode/credentials.py b/mycode/credentials.py
new file mode 100644
index 00000000..35cb7836
--- /dev/null
+++ b/mycode/credentials.py
@@ -0,0 +1,10 @@
+# Databricks notebook source
+def get_credentials_path(platform):
+  
+  az_creds_path = "/dbfs/FileStore/shared_uploads/darrell.coles@crowncastle.com/azure_databricks_credentials"
+  aws_creds_path = "/Workspace/Users/darrell.coles@crowncastle.com/aws_databricks_credentials"
+
+  match platform:
+    case "": raise ValueError("platform must be specified")
+    case "azure": return az_creds_path
+    case "aws": return aws_creds_path
diff --git a/mycode/delete-all-mlflow-experiments.py b/mycode/delete-all-mlflow-experiments.py
new file mode 100644
index 00000000..3bc4717c
--- /dev/null
+++ b/mycode/delete-all-mlflow-experiments.py
@@ -0,0 +1,20 @@
+# Databricks notebook source
+import mlflow
+mlflow.__version__
+
+# COMMAND ----------
+
+for exp in mlflow.search_experiments():
+  eid = exp.experiment_id
+  try:
+    mlflow.delete_experiment(experiment_id=eid)
+  except:
+    print(f"Experiment {eid} not deleted")
+
+# COMMAND ----------
+
+mlflow.search_experiments()
+
+# COMMAND ----------
+
+
diff --git a/mycode/delete-all-mlflow-registered-models.py b/mycode/delete-all-mlflow-registered-models.py
new file mode 100644
index 00000000..0cac72e2
--- /dev/null
+++ b/mycode/delete-all-mlflow-registered-models.py
@@ -0,0 +1,27 @@
+# Databricks notebook source
+from mlflow import MlflowClient
+
+# COMMAND ----------
+
+dbutils.widgets.dropdown("mlflow_model_registry","",["", "unity_catalog","workspace"])
+model_registry = dbutils.widgets.get("mlflow_model_registry")
+
+client = MlflowClient(registry_uri="databricks-uc") if model_registry == "unity_catalog" else MlflowClient()
+
+# COMMAND ----------
+
+for model in client.search_registered_models():
+  model_name = model.name
+  if model_registry == "unity_catalog":
+    if not model_name.startswith("ds_nonprod.migrated_models"):
+      continue
+  else:
+    # transition all stages to 'Archived' in workspace registry
+    dbutils.notebook.run("./set-all-registered-model-stages-to-archived", -1, {"model_name": model_name})
+
+  print(f"Deleting model: {model_name}")
+  client.delete_registered_model(model_name)
+
+# COMMAND ----------
+
+
diff --git a/mycode/get-model-hash.py b/mycode/get-model-hash.py
new file mode 100644
index 00000000..0aaabaf7
--- /dev/null
+++ b/mycode/get-model-hash.py
@@ -0,0 +1,86 @@
+# Databricks notebook source
+# MAGIC %md
+# MAGIC This notebook takes the name of a registered mlflow model and hashes its `champion` or `Production` version, depending on whether it's in Azure (Databricks Workspace) or AWS (Unity Catalog).
+# MAGIC
+# MAGIC 1. The directory of the `champion`/`Production` model is exported to the local machine with MLflow artifact utilities 
+# MAGIC 1. From this directory, hash the first file that ends with `pkl`
+
+# COMMAND ----------
+
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+import mlflow
+from mlflow import MlflowClient
+from hashlib import md5 
+import os
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+
+# COMMAND ----------
+
+# DBTITLE 1,utils
+def hash(path: str):
+  with open(path,"rb") as f:
+    return md5(f.read()).hexdigest()
+  
+def hash_model_directory(model_dir):
+  try:
+    fname = [file for file in os.listdir(model_dir) if file.endswith("pkl")][0]
+    model_hash = hash(model_dir+fname)
+    result = f"{model_name}: {model_hash}"
+  except:
+    result = f"{model_name}: Model file not found"
+  return result
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.text("registered-model-name","")
+model_name = dbutils.widgets.get("registered-model-name")
+
+dbutils.widgets.dropdown("mlflow_model_registry", "workspace",["workspace", "unity_catalog"])
+model_registry = dbutils.widgets.get("mlflow_model_registry")
+
+# COMMAND ----------
+
+# DBTITLE 1,dependent variables
+if model_registry == "workspace":
+  mlflow.set_registry_uri("databricks")
+elif model_registry == "unity_catalog":
+  mlflow.set_registry_uri("databricks-uc")
+else:
+  raise Exception("Invalid model registry")
+
+if model_registry == "workspace":
+  client=mlflow.tracking.MlflowClient()
+  try:
+    version = client.get_latest_versions(model_name, stages=["Production"])[0].version
+    model_uri = f"models:/{model_name}/{version}"
+  except:
+    model_uri = ""
+else:
+  model_uri = f"models:/{model_name}@champion"
+
+# COMMAND ----------
+
+# MAGIC %md ## Execute
+
+# COMMAND ----------
+
+try:
+  model_dir = mlflow.artifacts.download_artifacts(model_uri)
+  result = hash_model_directory(model_dir)
+except:
+  result = f"{model_name}: Model not found"
+
+# COMMAND ----------
+
+dbutils.notebook.exit(result)
+
+# COMMAND ----------
+
+
diff --git a/mycode/get-model-hashes.py b/mycode/get-model-hashes.py
new file mode 100644
index 00000000..39fa7463
--- /dev/null
+++ b/mycode/get-model-hashes.py
@@ -0,0 +1,29 @@
+# Databricks notebook source
+# DBTITLE 1,Libs
+import mlflow
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.dropdown("mlflow_model_registry", "workspace",["workspace", "unity_catalog"])
+model_registry = dbutils.widgets.get("mlflow_model_registry")
+
+# set up mlflow model registry uri
+if model_registry == "workspace":
+  mlflow.set_registry_uri("databricks")
+elif model_registry == "unity_catalog":
+  mlflow.set_registry_uri("databricks-uc")
+else:
+  raise Exception("Invalid model registry")
+
+# COMMAND ----------
+
+# DBTITLE 1,execute
+for model in mlflow.search_registered_models():
+  if (model_registry == "unity_catalog" and model.name.startswith("ds_nonprod.migrated_models")) or model_registry == "workspace": 
+    result = dbutils.notebook.run("get-model-hash", -1, {"registered-model-name": model.name, "mlflow_model_registry": model_registry})
+    print(result)
+
+# COMMAND ----------
+
+
diff --git a/mycode/manual-import-model.py b/mycode/manual-import-model.py
new file mode 100644
index 00000000..ad5a5c13
--- /dev/null
+++ b/mycode/manual-import-model.py
@@ -0,0 +1,150 @@
+# Databricks notebook source
+# MAGIC %md ### Manually import models that failed with the `import-models` utility
+# MAGIC
+# MAGIC You can find which models failed by searching the import_models log for "the source run_id was probably deleted"
+
+# COMMAND ----------
+
+# MAGIC %md ## Libs
+
+# COMMAND ----------
+
+# DBTITLE 1,install latest pkg from github
+# MAGIC %sh 
+# MAGIC #pip install mlflow-export-import
+# MAGIC pip install git+https:///github.com/mlflow/mlflow-export-import/#egg=mlflow-export-import
+
+# COMMAND ----------
+
+# MAGIC %run ./credentials
+
+# COMMAND ----------
+
+import os
+
+# COMMAND ----------
+
+# MAGIC %md ## Setup
+
+# COMMAND ----------
+
+# DBTITLE 1,variables
+dbutils.widgets.dropdown("platform","",["", "azure", "aws"])
+platform = dbutils.widgets.get("platform")
+
+credentials_path = get_credentials_path(platform)
+
+# COMMAND ----------
+
+# DBTITLE 1,set model env vars
+wij = dbutils.widgets
+
+wij.text("input_dir","/dbfs/mnt/datalake/mlflow-migration-models/models")
+input_dir = wij.get("input_dir")
+
+wij.text("model_name", "churn-model")
+model_name = wij.get("model_name")
+
+# set env variables for shell command
+os.environ["INPUT_DIR"]=input_dir
+os.environ["MODEL_NAME"]=model_name
+
+# COMMAND ----------
+
+# DBTITLE 1,set MLflow env vars
+from datetime import datetime
+import pytz
+
+cst = pytz.timezone('US/Central')
+now = datetime.now(tz=cst)
+date = now.strftime("%Y%m%d_%H%M")
+ 
+logfile = f"import_model_{model_name}_{date}.log"
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE"] = logfile 
+
+os.environ["MLFLOW_EXPORT_IMPORT_LOG_FORMAT"]="%(threadName)s-%(levelname)s-%(message)s"
+
+os.environ["MLFLOW_TRACKING_URI"]="databricks"
+
+with open(credentials_path) as f:
+  os.environ["DATABRICKS_HOST"]  = f.readline().strip("\n")
+  os.environ["DATABRICKS_TOKEN"] = f.readline().strip("\n")
+
+# COMMAND ----------
+
+# DBTITLE 1,set experiment env vars
+import json
+
+def get_most_recent_experiment_name(model: dict):
+  try:
+    newest = max(model["mlflow"]["registered_model"]["versions"], key=lambda v:v["last_updated_timestamp"])
+    return newest["_experiment_name"]
+  except:
+    return None
+
+with open(f"{input_dir}/{model_name}/model.json") as f:
+  model = json.load(f)
+  os.environ["EXPERIMENT_NAME"] = get_most_recent_experiment_name(model).replace(" ","_")
+
+# COMMAND ----------
+
+# DBTITLE 1,remove failed model registrations before import
+import json
+import os
+
+fname = f"{input_dir}/{model_name}/model.json"
+
+# read model.json
+with open(fname, "r") as f:
+  d = json.load(f)
+
+# remove failed versions
+for version in d["mlflow"]["registered_model"]["versions"]:
+  if version["status"] == 'FAILED_REGISTRATION':
+    d["mlflow"]["registered_model"]["versions"].remove(version)
+
+# rename old json
+os.rename(fname, f"{fname}_old")
+
+# write clean json
+with open(fname, "w") as f:
+  d = json.dump(d, f)
+
+# COMMAND ----------
+
+# DBTITLE 1,verify env vars
+# MAGIC %sh
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_OUTPUT_FILE
+# MAGIC echo $MLFLOW_EXPORT_IMPORT_LOG_FORMAT
+# MAGIC echo $MLFLOW_TRACKING_URI
+# MAGIC echo $INPUT_DIR
+# MAGIC echo $MODEL_NAME
+# MAGIC echo $EXPERIMENT_NAME
+# MAGIC #echo $DATABRICKS_HOST
+# MAGIC #echo $DATABRICKS_TOKEN
+
+# COMMAND ----------
+
+# MAGIC %md ## Execute
+
+# COMMAND ----------
+
+# DBTITLE 0,`export-models` options
+# MAGIC %sh 
+# MAGIC import-model --help
+
+# COMMAND ----------
+
+# DBTITLE 1,cli execution
+# MAGIC %sh 
+# MAGIC import-model \
+# MAGIC   --input-dir $INPUT_DIR/$MODEL_NAME \
+# MAGIC   --model $MODEL_NAME \
+# MAGIC   --experiment-name $EXPERIMENT_NAME \
+# MAGIC   --delete-model True \
+# MAGIC   # --import-source-tags True \
+# MAGIC   --verbose True
+
+# COMMAND ----------
+
+
diff --git a/mycode/models-by-owner.py b/mycode/models-by-owner.py
new file mode 100644
index 00000000..5c69bb5f
--- /dev/null
+++ b/mycode/models-by-owner.py
@@ -0,0 +1,27 @@
+# Databricks notebook source
+import mlflow
+import pyspark.sql.functions as F
+import pandas as pd
+
+# COMMAND ----------
+
+registered_models = mlflow.search_registered_models()
+
+model_name, user_id = [], []
+for model in registered_models:
+  model_name.append(model.name) 
+  user_id.append(model.latest_versions[0].user_id)
+
+user_models = pd.DataFrame({'model': model_name, 'user': user_id})
+
+# COMMAND ----------
+
+display(
+  spark.createDataFrame(user_models)
+  .groupBy("user")
+  .agg(F.collect_set("model").alias("models"))
+  )
+
+# COMMAND ----------
+
+
diff --git a/mycode/mount.py b/mycode/mount.py
new file mode 100644
index 00000000..b55005a7
--- /dev/null
+++ b/mycode/mount.py
@@ -0,0 +1,25 @@
+# Databricks notebook source
+# MAGIC %md 
+# MAGIC I added the AWS keys to Databricks secrets using the databricks cli. For Azure DEV, the profile name is **azure-dev**. 
+# MAGIC
+# MAGIC Raj had to give me the AWS keys because I don't have permission to create  AWS IAM users with S3 access. 
+# MAGIC
+# MAGIC ### Reference
+# MAGIC - https://docs.databricks.com/aws/en/dbfs/mounts#mount-a-bucket-using-aws-keys
+# MAGIC - https://docs.databricks.com/aws/en/security/secrets#create-a-secret-scope
+# MAGIC - https://docs.databricks.com/aws/en/security/secrets#create-a-secret
+
+# COMMAND ----------
+
+# DBTITLE 1,s3 bucket in ds-non-prod
+access_key = dbutils.secrets.get(scope = "aws", key = "aws-access-key")
+secret_key = dbutils.secrets.get(scope = "aws", key = "aws-secret-key")
+encoded_secret_key = secret_key.replace("/", "%2F")
+aws_bucket_name = "s3-us-east-2-dev-ds-non-prod-s3-app-ovisg"
+mount_name = "aws-ds-non-prod"
+
+dbutils.fs.mount(f"s3a://{access_key}:{encoded_secret_key}@{aws_bucket_name}", f"/mnt/{mount_name}")
+display(dbutils.fs.ls(f"/mnt/{mount_name}"))
+
+
+
diff --git a/mycode/set-all-registered-model-stages-to-archived.py b/mycode/set-all-registered-model-stages-to-archived.py
new file mode 100644
index 00000000..160a6780
--- /dev/null
+++ b/mycode/set-all-registered-model-stages-to-archived.py
@@ -0,0 +1,23 @@
+# Databricks notebook source
+from mlflow import MlflowClient
+
+# COMMAND ----------
+
+dbutils.widgets.text("model_name","")
+model_name = dbutils.widgets.get("model_name")
+
+# COMMAND ----------
+
+client = MlflowClient()
+
+for mv in client.search_model_versions(f"name='{model_name}'"):
+  version = int(mv.version)
+  if mv.current_stage != "Archived":
+    try:
+      client.transition_model_version_stage(name=model_name, version=version, stage="Archived")
+    except:
+      print(f"'{model_name}' version {version} state transition failed")
+
+# COMMAND ----------
+
+