cehrgpt 0.1.0__py3-none-any.whl → 0.1.2__py3-none-any.whl

cehrgpt/runners/sample_packing_trainer.py

@@ -35,6 +35,13 @@ class SamplePackingTrainer(Trainer):
                 self.max_tokens_per_batch,
             )
 
+        self.negative_sampling_probability = kwargs.pop(
+            "negative_sampling_probability", None
+        )
+        if self.negative_sampling_probability:
+            LOG.info(
+                "negative_sampling_probability: %s", self.negative_sampling_probability
+            )
         self.train_lengths = kwargs.pop("train_lengths", None)
         self.validation_lengths = kwargs.pop("validation_lengths", None)
         super().__init__(*args, **kwargs)
@@ -70,6 +77,14 @@ class SamplePackingTrainer(Trainer):
             data_collator = self._get_collator_with_removed_columns(
                 data_collator, description="training"
             )
+
+        labels = None
+        if (
+            self.negative_sampling_probability is not None
+            and "classifier_label" in train_dataset.column_names
+        ):
+            labels = train_dataset["classifier_label"]
+
         # Create our custom batch sampler
         batch_sampler = SamplePackingBatchSampler(
             lengths=lengths,
@@ -77,6 +92,8 @@ class SamplePackingTrainer(Trainer):
             max_position_embeddings=self.max_position_embeddings,
             drop_last=self.args.dataloader_drop_last,
             seed=self.args.seed,
+            negative_sampling_probability=self.negative_sampling_probability,
+            labels=labels,
         )
         dataloader_params = {
             "collate_fn": data_collator,

cehrgpt/time_to_event/config/1_year_cabg.yaml

@@ -0,0 +1,23 @@
+task_name: "cabg_prediction"
+outcome_events: [
+    "43528001",
+    "43528003",
+    "43528004",
+    "43528002",
+    "4305852",
+    "4168831",
+    "2107250",
+    "2107216",
+    "2107222",
+    "2107231",
+    "4336464",
+    "4231998",
+    "4284104",
+    "2100873",
+]
+future_visit_start: 0
+future_visit_end: -1
+prediction_window_start: 0
+prediction_window_end: 365
+max_new_tokens: 1024
+include_descendants: true

cehrgpt/time_to_event/time_to_event_model.py

@@ -80,20 +80,9 @@ class TimeToEventModel:
         return token in self.outcome_events
 
     def simulate(
-        self, partial_history: Union[np.ndarray, List[str]]
+        self,
+        partial_history: Union[np.ndarray, List[str]],
     ) -> List[List[str]]:
-
-        sequence_is_demographics = len(partial_history) == 4 and partial_history[
-            0
-        ].startswith("year")
-        sequence_ends_ve = is_visit_end(partial_history[-1])
-
-        if not (sequence_is_demographics | sequence_ends_ve):
-            raise ValueError(
-                "There are only two types of sequences allowed. 1) the sequence only contains "
-                "demographics; 2) the sequence ends on VE;"
-            )
-
         token_ids = self.tokenizer.encode(partial_history)
         prompt = torch.tensor(token_ids).unsqueeze(0).to(self.device)
 

cehrgpt/time_to_event/time_to_event_prediction.py

@@ -118,9 +118,9 @@ def main(args):
     LOG.info(f"Top P {args.top_p}")
     LOG.info(f"Top K {args.top_k}")
 
-    cehrgpt_model.resize_position_embeddings(
-        cehrgpt_model.config.max_position_embeddings + task_config.max_new_tokens
-    )
+    # cehrgpt_model.resize_position_embeddings(
+    #     cehrgpt_model.config.max_position_embeddings + task_config.max_new_tokens
+    # )
 
     generation_config = TimeToEventModel.get_generation_config(
         tokenizer=cehrgpt_tokenizer,
@@ -190,14 +190,22 @@ def main(args):
             args.max_n_trial,
         )
         visit_counter = sum([int(is_visit_end(_)) for _ in partial_history])
+        predicted_boolean_probability = (
+            sum([event != "0" for event in concept_time_to_event.outcome_events])
+            / len(concept_time_to_event.outcome_events)
+            if concept_time_to_event
+            else 0.0
+        )
         tte_outputs.append(
             {
-                "person_id": record["person_id"],
-                "index_date": record["index_date"],
+                "subject_id": record["person_id"],
+                "prediction_time": record["index_date"],
                 "visit_counter": visit_counter,
-                "label": label,
+                "boolean_value": label,
+                "predicted_boolean_probability": predicted_boolean_probability,
+                "predicted_boolean_value": None,
                 "time_to_event": time_to_event,
-                "prediction": (
+                "trials": (
                     asdict(concept_time_to_event) if concept_time_to_event else None
                 ),
             }
@@ -263,9 +271,13 @@ def filter_out_existing_results(
     parquet_files = glob.glob(os.path.join(prediction_output_folder_name, "*parquet"))
     if parquet_files:
         cohort_members = set()
-        results_dataframe = pd.read_parquet(parquet_files)[["person_id", "index_date"]]
+        results_dataframe = pd.read_parquet(parquet_files)[
+            ["subject_id", "prediction_time"]
+        ]
         for row in results_dataframe.itertuples():
-            cohort_members.add((row.person_id, row.index_date.strftime("%Y-%m-%d")))
+            cohort_members.add(
+                (row.subject_id, row.prediction_time.strftime("%Y-%m-%d"))
+            )
 
         def filter_func(batched):
             return [
@@ -292,12 +304,14 @@ def flush_to_disk_if_full(
         pd.DataFrame(
             tte_outputs,
             columns=[
-                "person_id",
-                "index_date",
+                "subject_id",
+                "prediction_time",
                 "visit_counter",
-                "label",
+                "boolean_value",
+                "predicted_boolean_probability",
+                "predicted_boolean_value",
                 "time_to_event",
-                "prediction",
+                "trials",
             ],
         ).to_parquet(output_parquet_file)
         tte_outputs.clear()

cehrgpt/tools/linear_prob/compute_cehrgpt_features.py

@@ -1,8 +1,8 @@
+import datetime
 import glob
 import os
 import shutil
 import uuid
-from datetime import datetime
 from functools import partial
 from pathlib import Path
 from typing import Optional, Union
@@ -29,8 +29,12 @@ from cehrgpt.models.hf_cehrgpt import (
     CEHRGPT2Model,
     extract_features_from_packed_sequence,
 )
+from cehrgpt.models.special_tokens import LINEAR_PROB_TOKEN
 from cehrgpt.models.tokenization_hf_cehrgpt import CehrGptTokenizer
-from cehrgpt.runners.data_utils import prepare_finetune_dataset
+from cehrgpt.runners.data_utils import (
+    extract_cohort_sequences,
+    prepare_finetune_dataset,
+)
 from cehrgpt.runners.gpt_runner_util import parse_runner_args
 from cehrgpt.runners.hf_cehrgpt_pretrain_runner import tokenizer_exists
 
@@ -112,6 +116,11 @@ def main():
         .eval()
         .to(device)
     )
+
+    if LINEAR_PROB_TOKEN not in cehrgpt_tokenizer.get_vocab():
+        cehrgpt_tokenizer.add_tokens(LINEAR_PROB_TOKEN)
+        cehrgpt_model.resize_token_embeddings(cehrgpt_tokenizer.vocab_size)
+
     prepared_ds_path = generate_prepared_ds_path(
         data_args, model_args, data_folder=data_args.cohort_folder
     )
@@ -137,39 +146,31 @@ def main():
 
     if processed_dataset is None:
         if is_main_process(training_args.local_rank):
-            # Organize them into a single DatasetDict
-            final_splits = prepare_finetune_dataset(
-                data_args, training_args, cehrgpt_args, cache_file_collector
-            )
-            if cehrgpt_args.expand_tokenizer:
-                new_tokenizer_path = os.path.expanduser(training_args.output_dir)
-                if tokenizer_exists(new_tokenizer_path):
-                    cehrgpt_tokenizer = CehrGptTokenizer.from_pretrained(
-                        new_tokenizer_path
-                    )
-                else:
-                    cehrgpt_tokenizer = CehrGptTokenizer.expand_trained_tokenizer(
-                        cehrgpt_tokenizer=cehrgpt_tokenizer,
-                        dataset=final_splits["train"],
-                        data_args=data_args,
-                        concept_name_mapping={},
-                    )
-                    cehrgpt_tokenizer.save_pretrained(
-                        os.path.expanduser(training_args.output_dir)
-                    )
-
+            # If the full dataset has been tokenized, we don't want to tokenize the cohort containing
+            # the subset of the data. We should slice out the portion of the tokenized sequences for each sample
+            if cehrgpt_args.tokenized_full_dataset_path is not None:
+                processed_dataset = extract_cohort_sequences(
+                    data_args, cehrgpt_args, cache_file_collector
+                )
+            else:
+                # Organize them into a single DatasetDict
+                final_splits = prepare_finetune_dataset(
+                    data_args, training_args, cehrgpt_args, cache_file_collector
+                )
                 # TODO: temp solution, this column is mixed typed and causes an issue when transforming the data
-            if not data_args.streaming:
-                all_columns = final_splits["train"].column_names
-                if "visit_concept_ids" in all_columns:
-                    final_splits = final_splits.remove_columns(["visit_concept_ids"])
-
-            processed_dataset = create_cehrgpt_finetuning_dataset(
-                dataset=final_splits,
-                cehrgpt_tokenizer=cehrgpt_tokenizer,
-                data_args=data_args,
-                cache_file_collector=cache_file_collector,
-            )
+                if not data_args.streaming:
+                    all_columns = final_splits["train"].column_names
+                    if "visit_concept_ids" in all_columns:
+                        final_splits = final_splits.remove_columns(
+                            ["visit_concept_ids"]
+                        )
+
+                processed_dataset = create_cehrgpt_finetuning_dataset(
+                    dataset=final_splits,
+                    cehrgpt_tokenizer=cehrgpt_tokenizer,
+                    data_args=data_args,
+                    cache_file_collector=cache_file_collector,
+                )
             if not data_args.streaming:
                 processed_dataset.save_to_disk(prepared_ds_path)
                 processed_dataset.cleanup_cache_files()
@@ -218,10 +219,6 @@ def main():
             len(processed_dataset["test"]),
         )
 
-    LOG.info(f"cehrgpt_model.config.vocab_size: {cehrgpt_model.config.vocab_size}")
-    LOG.info(f"cehrgpt_tokenizer.vocab_size: {cehrgpt_tokenizer.vocab_size}")
-    if cehrgpt_model.config.vocab_size < cehrgpt_tokenizer.vocab_size:
-        cehrgpt_model.resize_token_embeddings(cehrgpt_tokenizer.vocab_size)
     if (
         cehrgpt_model.config.max_position_embeddings
         < model_args.max_position_embeddings
@@ -244,6 +241,7 @@ def main():
             SamplePackingCehrGptDataCollator,
             cehrgpt_args.max_tokens_per_batch,
             cehrgpt_model.config.max_position_embeddings,
+            add_end_token_in_sample_packing=cehrgpt_args.add_end_token_in_sample_packing,
         )
         train_batch_sampler = SamplePackingBatchSampler(
             lengths=train_set["num_of_concepts"],
@@ -278,6 +276,7 @@ def main():
         include_ttv_prediction=False,
         use_sub_time_tokenization=False,
         include_demographics=cehrgpt_args.include_demographics,
+        add_linear_prob_token=True,
     )
 
     train_loader = DataLoader(
@@ -298,30 +297,38 @@ def main():
         batch_sampler=test_batch_sampler,
     )
 
-    # Loading demographics
-    print("Loading demographics as a dictionary")
-    demographics_df = pd.concat(
-        [
-            pd.read_parquet(
-                data_dir,
-                columns=[
-                    "person_id",
-                    "index_date",
-                    "gender_concept_id",
-                    "race_concept_id",
-                ],
-            )
-            for data_dir in [data_args.data_folder, data_args.test_data_folder]
-        ]
-    )
-    demographics_df["index_date"] = demographics_df.index_date.dt.date
-    demographics_dict = {
-        (row["person_id"], row["index_date"]): {
-            "gender_concept_id": row["gender_concept_id"],
-            "race_concept_id": row["race_concept_id"],
+    if data_args.is_data_in_meds:
+        demographics_dict = dict()
+    else:
+        # Loading demographics
+        print("Loading demographics as a dictionary")
+        demographics_df = pd.concat(
+            [
+                pd.read_parquet(
+                    data_dir,
+                    columns=[
+                        "person_id",
+                        "index_date",
+                        "gender_concept_id",
+                        "race_concept_id",
+                    ],
+                )
+                for data_dir in [data_args.data_folder, data_args.test_data_folder]
+            ]
+        )
+
+        demographics_df["index_date"] = (
+            demographics_df["index_date"].dt.tz_localize("UTC")
+            - datetime.datetime(1970, 1, 1, tzinfo=datetime.timezone.utc)
+        ).dt.total_seconds()
+
+        demographics_dict = {
+            (row["person_id"], row["index_date"]): {
+                "gender_concept_id": row["gender_concept_id"],
+                "race_concept_id": row["race_concept_id"],
+            }
+            for _, row in demographics_df.iterrows()
         }
-        for _, row in demographics_df.iterrows()
-    }
 
     data_loaders = [("train", train_loader), ("test", test_dataloader)]
 
@@ -351,9 +358,16 @@ def main():
                 prediction_time_posix = batch.pop("index_date").numpy().squeeze()
                 if prediction_time_posix.ndim == 0:
                     prediction_time_posix = np.asarray([prediction_time_posix])
+
                 prediction_time = list(
-                    map(datetime.fromtimestamp, prediction_time_posix)
+                    map(
+                        lambda posix_time: datetime.datetime.utcfromtimestamp(
+                            posix_time
+                        ).replace(tzinfo=None),
+                        prediction_time_posix,
+                    )
                 )
+
                 labels = (
                     batch.pop("classifier_label")
                     .float()
@@ -365,6 +379,10 @@ def main():
                 if labels.ndim == 0:
                     labels = np.asarray([labels])
 
+                # Right now the model does not support this column, we need to pop it
+                if "epoch_times" in batch:
+                    batch.pop("epoch_times")
+
                 batch = {k: v.to(device) for k, v in batch.items()}
                 # Forward pass
                 cehrgpt_output = cehrgpt_model(

{cehrgpt-0.1.0.dist-info → cehrgpt-0.1.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cehrgpt
-Version: 0.1.0
+Version: 0.1.2
 Summary: CEHR-GPT: Generating Electronic Health Records with Chronological Patient Timelines
 Author-email: Chao Pang <chaopang229@gmail.com>, Xinzhuo Jiang <xj2193@cumc.columbia.edu>, Krishna Kalluri <kk3326@cumc.columbia.edu>, Elise Minto <em3697@cumc.columbia.edu>, Jason Patterson <jp3477@cumc.columbia.edu>, Nishanth Parameshwar Pavinkurve <np2689@cumc.columbia.edu>, Karthik Natarajan <kn2174@cumc.columbia.edu>
 License: MIT License
@@ -12,14 +12,15 @@ Classifier: Programming Language :: Python :: 3
 Requires-Python: >=3.10.0
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: cehrbert==1.4.1
-Requires-Dist: cehrbert_data==0.0.7
+Requires-Dist: cehrbert==1.4.5
+Requires-Dist: cehrbert_data==0.0.11
 Requires-Dist: openai==1.54.3
 Requires-Dist: optuna==4.0.0
-Requires-Dist: transformers==4.44.0
+Requires-Dist: transformers==4.44.1
 Requires-Dist: tokenizers==0.19.0
 Requires-Dist: peft==0.10.0
 Requires-Dist: lightgbm
+Requires-Dist: polars
 Provides-Extra: dev
 Requires-Dist: pre-commit; extra == "dev"
 Requires-Dist: pytest; extra == "dev"
@@ -36,9 +37,9 @@ Dynamic: license-file
 
 [![PyPI - Version](https://img.shields.io/pypi/v/cehrgpt)](https://pypi.org/project/cehrgpt/)
 ![Python](https://img.shields.io/badge/-Python_3.11-blue?logo=python&logoColor=white)
-[![tests](https://github.com/knatarajan-lab/cehrgpt-public/actions/workflows/tests.yaml/badge.svg)](https://github.com/knatarajan-lab/cehrgpt-public/actions/workflows/tests.yml)
-[![license](https://img.shields.io/badge/License-MIT-green.svg?labelColor=gray)](https://github.com/knatarajan-lab/cehrgpt-public/blob/main/LICENSE)
-[![contributors](https://img.shields.io/github/contributors/knatarajan-lab/cehrgpt-public.svg)](https://github.com/knatarajan-lab/cehrgpt-public/graphs/contributors)
+[![tests](https://github.com/knatarajan-lab/cehrgpt/actions/workflows/tests.yaml/badge.svg)](https://github.com/knatarajan-lab/cehrgpt/actions/workflows/tests.yaml)
+[![license](https://img.shields.io/badge/License-MIT-green.svg?labelColor=gray)](https://github.com/knatarajan-lab/cehrgpt/blob/main/LICENSE)
+[![contributors](https://img.shields.io/github/contributors/knatarajan-lab/cehrgpt.svg)](https://github.com/knatarajan-lab/cehrgpt/graphs/contributors)
 
 ## Description
 CEHRGPT is a synthetic data generation model developed to handle structured electronic health records (EHR) with enhanced privacy and reliability. It leverages state-of-the-art natural language processing techniques to create realistic, anonymized patient data that can be used for research and development without compromising patient privacy.
@@ -104,6 +105,100 @@ sh scripts/omop_pipeline.sh \
   $OMOP_VOCAB_DIR
 ```
 
+# MEDS Support
+
+This section demonstrates how to pretrain CEHR-GPT using MIMIC-IV data in the MEDS (Medical Event Data Standard) format.
+
+## Prerequisites
+
+Set up the required environment variables before beginning:
+
+```bash
+export CEHR_GPT_MODEL_DIR=""    # Path to CEHR-GPT model directory
+export MEDS_DIR=""              # Path to MEDS data directory
+export MEDS_READER_DIR=""       # Path to MEDS reader output directory
+```
+
+## Step 1: Create MIMIC MEDS Data
+
+Transform your MIMIC files into MEDS format by following the instructions in the [MEDS_transforms](https://github.com/mmcdermott/MEDS_transforms/) repository.
+
+## Step 2: Create the MEDS Reader
+
+Convert the MEDS data for use with CEHR-GPT:
+
+```bash
+meds_reader_convert $MEDS_DIR $MEDS_READER_DIR --num_threads 10
+```
+
+## Step 3: Pretrain CEHR-GPT
+
+Run the pretraining process using the prepared MEDS data:
+
+```bash
+python -u -m cehrgpt.runners.hf_cehrgpt_pretrain_runner \
+  --model_name_or_path $CEHR_GPT_MODEL_DIR \
+  --tokenizer_name_or_path $CEHR_GPT_MODEL_DIR \
+  --output_dir $CEHR_GPT_MODEL_DIR \
+  --data_folder $MEDS_READER_DIR \
+  --dataset_prepared_path "$CEHR_GPT_MODEL_DIR/dataset_prepared" \
+  --do_train true --seed 42 \
+  --dataloader_num_workers 16 --dataloader_prefetch_factor 8 \
+  --hidden_size 768 --num_hidden_layers 14 --max_position_embeddings 8192 \
+  --evaluation_strategy epoch --save_strategy epoch \
+  --sample_packing --max_tokens_per_batch 16384 \
+  --warmup_steps 500 --weight_decay 0.01 \
+  --num_train_epochs 50 --learning_rate 0.0002 \
+  --use_early_stopping --early_stopping_threshold 0.001 \
+  --is_data_in_meds --inpatient_att_function_type day \
+  --att_function_type day --include_inpatient_hour_token \
+  --include_auxiliary_token --include_demographic_prompt \
+  --meds_to_cehrbert_conversion_type "MedsToBertMimic4"
+```
+
+## Step 4: Generate MEDS Trajectories
+
+### Environment Setup for Trajectory Generation
+
+Configure additional environment variables for trajectory generation with task labels (`subject_id`, `prediction_time`, `boolean_value` [optional]):
+
+```bash
+# MEDS_LABEL_COHORT_DIR must contain a set of parquet files
+export MEDS_LABEL_COHORT_DIR=""     # Path to cohort labels directory
+export MEDS_TRAJECTORY_DIR=""       # Path for trajectory output
+```
+
+### Generate Trajectories
+
+Create synthetic patient trajectories using the trained model:
+
+> **Important:** The total sequence length (`generation_input_length` + `generation_max_new_tokens`) cannot exceed the `max_position_embeddings` value (8192) defined during pretraining.
+
+```bash
+python -u -m cehrgpt.generation.cehrgpt_conditional_generation \
+  --cohort_folder $MEDS_LABEL_COHORT_DIR \
+  --data_folder $MEDS_READER_DIR \
+  --dataset_prepared_path "$CEHR_GPT_MODEL_DIR/dataset_prepared" \
+  --model_name_or_path $CEHR_GPT_MODEL_DIR \
+  --tokenizer_name_or_path $CEHR_GPT_MODEL_DIR \
+  --output_dir $MEDS_TRAJECTORY_DIR \
+  --per_device_eval_batch_size 16 \
+  --num_of_trajectories_per_sample 2 \
+  --generation_input_length 4096 \
+  --generation_max_new_tokens 4096 \
+  --is_data_in_meds \
+  --att_function_type day --inpatient_att_function_type day \
+  --meds_to_cehrbert_conversion_type MedsToBertMimic4 \
+  --include_auxiliary_token --include_demographic_prompt \
+  --include_inpatient_hour_token
+```
+
+### Parameters Explanation
+
+- `generation_input_length`: Controls the length of input context for generation
+- `generation_max_new_tokens`: Maximum number of new tokens to generate
+- `num_of_trajectories_per_sample`: Number of trajectories to generate per patient sample
+
 ## Citation
 ```
 @article{cehrgpt2024,

{cehrgpt-0.1.0.dist-info → cehrgpt-0.1.2.dist-info}/RECORD

@@ -1,8 +1,9 @@
 __init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cehrgpt/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cehrgpt/cehrgpt_args.py,sha256=zPLp9Qjlq5PapWx3R15BNnyaX8zV3dxr4PuWj71r0Lg,3516
-cehrgpt/gpt_utils.py,sha256=bksHCXMX4j859VSv1Q284rVr4gn1Y8dCx4a_V-g4mug,10939
+cehrgpt/gpt_utils.py,sha256=IA5qw-hxcKkGO07AB47lDNRU6mlb9jblpKO7KeLLN78,11342
 cehrgpt/analysis/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cehrgpt/analysis/irregularity.py,sha256=Rfl_daMvSh9cZ68vUwfmuH-JYCFXdAph2ITHHffYC0Y,1047
 cehrgpt/analysis/privacy/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cehrgpt/analysis/privacy/attribute_inference.py,sha256=0ANVW0I5uvOl6IxQ15-vMVQd0mugOgSGReBUQQESImg,9368
 cehrgpt/analysis/privacy/attribute_inference_config.yml,sha256=hfLfpBlDqqsNOynpRHK414vV24edKA6ta-inmEhM2ao,103272
@@ -11,22 +12,23 @@ cehrgpt/analysis/privacy/nearest_neighbor_inference.py,sha256=qoJgWW7VsUMzjMGpTa
 cehrgpt/analysis/privacy/reid_inference.py,sha256=Pypd3QJXQNY8VljpnIEa5zeAbTZHMjQOazaL-9VsBGw,13955
 cehrgpt/analysis/privacy/utils.py,sha256=CRA4H9mPLBjMQGKzZ_x_3ro3tMap-NjsMDVqSOjHSVQ,8226
 cehrgpt/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cehrgpt/data/hf_cehrgpt_dataset.py,sha256=t9vpN05e--CiKgIlxLP0aLacISnvWWDPXtuFuJi3ksE,3736
-cehrgpt/data/hf_cehrgpt_dataset_collator.py,sha256=DOvIF4Wzkd8-IO3zpIRZkX1j0IdvefaiSnrDn1YivCk,27912
-cehrgpt/data/hf_cehrgpt_dataset_mapping.py,sha256=eI8CTk6yJ4DlNJWrNAkEmhWh353NeLqg5rwPpKqKT-U,17308
-cehrgpt/data/sample_packing_sampler.py,sha256=0uKTbvtXpfS81esy_3epJ88eohyJPK46bfmxhle1fws,5419
+cehrgpt/data/hf_cehrgpt_dataset.py,sha256=hwJlGW7XiJIr6cXtmwvReQf9yLZJPD-dvJGvRg5ERqU,3755
+cehrgpt/data/hf_cehrgpt_dataset_collator.py,sha256=juM5HeZScgj8w15Bl1qC83Swld4gY6avh0QkSWLqITA,45465
+cehrgpt/data/hf_cehrgpt_dataset_mapping.py,sha256=_QDX9NXfmQ_S3kOf3yndb3AhoEeFiSzAOv836uYW0AY,26230
+cehrgpt/data/sample_packing_sampler.py,sha256=vovGMtmhG70DRkSCeiaDEJ_rjKZ38y-YLaI1kkhFEkI,6747
 cehrgpt/generation/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cehrgpt/generation/cehrgpt_conditional_generation.py,sha256=AM76yaPyw1B-bcdei24HO0uspGZWHGKWpYpHywotTIQ,11972
 cehrgpt/generation/chatgpt_generation.py,sha256=SrnLwHLdNtnAOEg36gNjqfoT9yd12iyPgpZffL2AFJo,4428
-cehrgpt/generation/generate_batch_hf_gpt_sequence.py,sha256=uSEh8aMmPD61nGewIaPSkIqm-2AxDjCBiu4cBfxHxU4,11503
-cehrgpt/generation/omop_converter_batch.py,sha256=-c0AlDVy5pJ5Afhr8ERiCHhoRrEk8ozJi3g0yFdWaMI,25348
+cehrgpt/generation/generate_batch_hf_gpt_sequence.py,sha256=P8al4-zqymqEkCHCCu2sqz_45akcKF2o_AtQIjJdVmQ,11919
+cehrgpt/generation/omop_converter_batch.py,sha256=LUmCD-t_6ZP1YfNDZCqYewl-XIIaIgRZ_dAxuR_VdCQ,26275
 cehrgpt/generation/omop_entity.py,sha256=Q5Sr0AlyuPAm1FRPfnJO13q-u1fqRgYVHXruZ9g4xNE,19400
 cehrgpt/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cehrgpt/models/config.py,sha256=Y3CiXZWniLP9_RlpU80Oe9gjn5leLmTYnNe_fWqfJLQ,10158
-cehrgpt/models/hf_cehrgpt.py,sha256=3EQIOfa--oz4f8bM8KzbDi98G3XrUEQkox1vmBN001M,83321
-cehrgpt/models/hf_modeling_outputs.py,sha256=LaWa1jI6BRIKMEjWOy1QUeOfTur5y_p2c-JyuGVTdtw,10301
+cehrgpt/models/config.py,sha256=nOAKgH5420HLCcy7n1hE7MbqR861Iq4DTutKoAd25tg,11090
+cehrgpt/models/hf_cehrgpt.py,sha256=3P7bOLDr7NMSedGszhmlJJN4Mhpd_65-x6uzwvSjigE,92837
+cehrgpt/models/hf_modeling_outputs.py,sha256=5X4WEYKqT37phv_e5ZAv3A_N0wqdAUJLJRm6TxS6dDQ,10356
 cehrgpt/models/pretrained_embeddings.py,sha256=vLLVs17TLpXRqCVEWQxGGwPHkUJUO7laNTeBuyBK_yk,3238
-cehrgpt/models/special_tokens.py,sha256=-a7HPJBbdIH0qQ6B3CcRKqvpG6FZlm4nbVPTswGSJ4U,485
-cehrgpt/models/tokenization_hf_cehrgpt.py,sha256=jjCRqS29IzMnKp40jNOs80UKh2z9lK5S6M02GSB-4mk,42351
+cehrgpt/models/special_tokens.py,sha256=lrw45B4tea4Dsajn09Cz6w5D2TfHmYXikZkgwnstu_o,521
+cehrgpt/models/tokenization_hf_cehrgpt.py,sha256=cAxHTctpVBxfWfC3XcwDQavN1zwWN9Nid_Fajd5zQWQ,53159
 cehrgpt/omop/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cehrgpt/omop/condition_era.py,sha256=hPZALz2XaWnro_1bwIYNkI48foOJjueyg3CZ1BliCno,626
 cehrgpt/omop/observation_period.py,sha256=TRMgv5Ya2RaS2im7oQ6BLC_5JL9EJYNYR62ApxIuHvg,1211
@@ -37,22 +39,23 @@ cehrgpt/omop/queries/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hS
 cehrgpt/omop/queries/condition_era.py,sha256=LFB6vBAvshHJxtYIRkl7cfrF0kf7ay0piBKpmHBwrpE,2578
 cehrgpt/omop/queries/observation_period.py,sha256=fpzr5DMNw-QLoSwp2Iatfch88E3hyhZ75usiIdG3A0U,6410
 cehrgpt/runners/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cehrgpt/runners/data_utils.py,sha256=ScZZnfXwgXKaMvKgFzdb4vtQ7F_lw97O5uNsFbfsyP4,10620
+cehrgpt/runners/data_utils.py,sha256=i-krtBx_6rvPYtdLdDoWwOTtJcaovd0wH8gBYmgN2l4,16013
 cehrgpt/runners/gpt_runner_util.py,sha256=YJQSRW9Mo4TjXSOUOTf6BUFcs1MGFiXU5T4ztKZcYhU,3485
-cehrgpt/runners/hf_cehrgpt_finetune_runner.py,sha256=bkPl30Y9CSXBlmMkH-3cA3-aW8XJK36Q-adx___WjkE,26921
-cehrgpt/runners/hf_cehrgpt_pretrain_runner.py,sha256=ViVa_flEGdk_SO0psMR7ho-o79igsz_l1x80u81WJ3A,23875
-cehrgpt/runners/hf_gpt_runner_argument_dataclass.py,sha256=VrqgDSiAMfGyHEIodoOg_8LU5O0ndWf9EE0YOKDFKKA,7019
-cehrgpt/runners/hyperparameter_search_util.py,sha256=pWFmGo9Ezju4YmuZ-ohbAbYB0GGMfIDVUCyvcTxS1iU,9153
-cehrgpt/runners/sample_packing_trainer.py,sha256=aezX30vxpP1DDcH5hO-yn395NqBKi2Xhb0mFNHi9OBs,7340
+cehrgpt/runners/hf_cehrgpt_finetune_runner.py,sha256=1OgxLm4T7iHv5pKi2QaSdaz9ogWo2n3sSUGp6cHDF9s,28309
+cehrgpt/runners/hf_cehrgpt_pretrain_runner.py,sha256=ERSnvB38fPYVghtKQeNTZ8VfeXnoRcCHB0cWISWaZ84,26523
+cehrgpt/runners/hf_gpt_runner_argument_dataclass.py,sha256=fJR4RHPqal1YI6_KUH-WlkoQLSZuBT5bKUGfPHDFrWI,9350
+cehrgpt/runners/hyperparameter_search_util.py,sha256=YWdFQ1igQs-G_wqWUrUzYraGiz8OSpSYyvid-I5nhWA,9262
+cehrgpt/runners/sample_packing_trainer.py,sha256=Zb7Aqwnk8-VqrjEKUVeg5XzZWmHxXOU2sDn1YURS-FU,7960
 cehrgpt/simulations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cehrgpt/simulations/generate_plots.py,sha256=BTZ71r8Kah0PMorkiO3vw55_p_9U1Z8KiD3GsPfaV0s,2520
 cehrgpt/simulations/run_simulation.sh,sha256=DcJ6B19jIteUO0pZ0Tc21876lB9XxQHFAxlre7MtAzk,795
 cehrgpt/simulations/time_embedding_simulation.py,sha256=HZ-imXH-bN-QYZN1PAfcERmNtaWIwKjbf0UrZduwCiA,8687
 cehrgpt/simulations/time_token_simulation.py,sha256=sLg8vVXydvR_zk3BbqyrlA7sDIdhFnS-s5pSKcCilSc,6057
 cehrgpt/time_to_event/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cehrgpt/time_to_event/time_to_event_model.py,sha256=tfXa24l_0q1TBZ68BPRrHRC_3KRWYxrWGIv4myJlIb8,8497
-cehrgpt/time_to_event/time_to_event_prediction.py,sha256=Ajesq2gSsILghWHCTLiiBhcyOCa7m6JPPMdi_xvBlR4,12624
+cehrgpt/time_to_event/time_to_event_model.py,sha256=Plm0bZxvlAbnMl82DTBXWvaXLvrqcdkzcP_celX8WC4,8055
+cehrgpt/time_to_event/time_to_event_prediction.py,sha256=W2e7UqIV7ELdfTy997HS66vggjnhdncCKt840knI0Dw,13183
 cehrgpt/time_to_event/time_to_event_utils.py,sha256=KN4hwGgxy2nJtO7osbYQBF3-HpmGUWefNfexzPYiEwc,1937
+cehrgpt/time_to_event/config/1_year_cabg.yaml,sha256=SFF2-F5D02pDSMRddDrEUoERBCd0t2Hzln_xC-Mo2hA,407
 cehrgpt/time_to_event/config/30_day_readmission.yaml,sha256=Hn5KnEXMtSV_CtCpmAU4wjkc0-gTXvniaH991TSbUXA,234
 cehrgpt/time_to_event/config/next_visit_type_prediction.yaml,sha256=WMj2ZutEvHKIMyGG51xtXaL6MyRANKvpg9xT8ouctLc,319
 cehrgpt/time_to_event/config/t2dm_hf.yaml,sha256=_oMQzh2eJTYzEaMOpmhAzbX-qmdsKlkORELL6HxOxHo,202
@@ -63,10 +66,10 @@ cehrgpt/tools/generate_pretrained_embeddings.py,sha256=lhFSacGv8bMld6qigKZN8Op8e
 cehrgpt/tools/merge_synthetic_real_dataasets.py,sha256=O1dbQ32Le0t15fwymwAh9mfNVLEWuFwW53DNvESrWbY,7589
 cehrgpt/tools/upload_omop_tables.py,sha256=vdBAbkeAsGPA4NsyhNjelPVj3gS8yzmS1sKNM1Qk96g,3791
 cehrgpt/tools/linear_prob/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cehrgpt/tools/linear_prob/compute_cehrgpt_features.py,sha256=jVgAmBrZKp7ABfqKkzwV5Vl_G9jDCjPl98NSVmSwHpE,19291
+cehrgpt/tools/linear_prob/compute_cehrgpt_features.py,sha256=Hpx7WvAWm2WwPHFfimCADXh019I7bwdzJ4_5_YCxQzU,19817
 cehrgpt/tools/linear_prob/train_with_cehrgpt_features.py,sha256=w0UvzMKYGenN_KDVnbzutmy8IPLUxW5hPvpKKxDSL5U,5820
-cehrgpt-0.1.0.dist-info/licenses/LICENSE,sha256=LOfC32zkfUIdGm8e_098jPbt8OHKtNWymDzxn2pA9Zk,1093
-cehrgpt-0.1.0.dist-info/METADATA,sha256=V02vsptjJRD_bybXVRFXPrJa-By9CX4j-oAA3EfXFq4,4933
-cehrgpt-0.1.0.dist-info/WHEEL,sha256=Nw36Djuh_5VDukK0H78QzOX-_FQEo6V37m3nkm96gtU,91
-cehrgpt-0.1.0.dist-info/top_level.txt,sha256=akNCJBbMSLV8nkOzdVzdy13hMJ5CIQURnAS_YYEDVwA,17
-cehrgpt-0.1.0.dist-info/RECORD,,
+cehrgpt-0.1.2.dist-info/licenses/LICENSE,sha256=LOfC32zkfUIdGm8e_098jPbt8OHKtNWymDzxn2pA9Zk,1093
+cehrgpt-0.1.2.dist-info/METADATA,sha256=D7gGKrQThiLivViFeNm711NCP8J-wXfkueMGb6RKqV0,8481
+cehrgpt-0.1.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cehrgpt-0.1.2.dist-info/top_level.txt,sha256=akNCJBbMSLV8nkOzdVzdy13hMJ5CIQURnAS_YYEDVwA,17
+cehrgpt-0.1.2.dist-info/RECORD,,

{cehrgpt-0.1.0.dist-info → cehrgpt-0.1.2.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.7.1)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any