py

docs/source/_toctree.yml

@@ -11,6 +11,8 @@
 - sections:
   - local: quickstart_spaces
     title: Train on Spaces
+  - local: quickstart_py
+    title: Python SDK
   - local: quickstart
     title: Train Locally
   - local: config

docs/source/faq.mdx

@@ -16,6 +16,14 @@ You can safely remove the dataset from the Hub after training is complete.
 If uploaded, the dataset will be stored in your Hugging Face account as a private repository and will only be accessible by you
 and the training process. It is not used once the training is complete.
 
+## My training space paused for no reason mid-training
+
+AutoTrain Training Spaces will pause itself after training is done (or failed). This is done to save resources and costs.
+If your training failed, you can still see the space logs and find out what went wrong. Note: you won't be able to retrive the logs if you restart the space.
+
+Another reason for the space to pause is if the space is space's sleep time kicking in. If you have a long running training job, you must set the sleep time to a much higher value.
+The space will anyways pause itself after the training is done thus saving you costs.
+
 ## I get error `Your installed package nvidia-ml-py is corrupted. Skip patch functions`
 
 This error can be safely ignored. It is a warning from the `nvitop` library and does not affect the functionality of AutoTrain.

docs/source/quickstart.mdx

@@ -1,4 +1,4 @@
-# Quickstart
+# Quickstart Guide for Local Training
 
 This quickstart is for local installation and usage.
 If you want to use AutoTrain on Hugging Face Spaces, please refer to the *AutoTrain on Hugging Face Spaces* section.

docs/source/quickstart_py.mdx

@@ -0,0 +1,111 @@
+# Quickstart with Python
+
+AutoTrain is a library that allows you to train state of the art models on Hugging Face Spaces, or locally. 
+It provides a simple and easy-to-use interface to train models for various tasks like llm finetuning, text classification, 
+image classification, object detection, and more.
+
+In this quickstart guide, we will show you how to train a model using AutoTrain in Python.
+
+## Getting Started
+
+AutoTrain can be installed using pip:
+
+```bash
+$ pip install autotrain-advanced
+```
+
+The example code below shows how to finetune an LLM model using AutoTrain in Python:
+
+```python
+import os
+
+from autotrain.params import LLMTrainingParams
+from autotrain.project import AutoTrainProject
+
+
+params = LLMTrainingParams(
+    model="meta-llama/Llama-3.2-1B-Instruct",
+    data_path="HuggingFaceH4/no_robots",
+    chat_template="tokenizer",
+    text_column="messages",
+    train_split="train",
+    trainer="sft",
+    epochs=3,
+    batch_size=1,
+    lr=1e-5,
+    peft=True,
+    quantization="int4",
+    target_modules="all-linear",
+    padding="right",
+    optimizer="paged_adamw_8bit",
+    scheduler="cosine",
+    gradient_accumulation=8,
+    mixed_precision="bf16",
+    merge_adapter=True,
+    project_name="autotrain-llama32-1b-finetune",
+    log="tensorboard",
+    push_to_hub=True,
+    username=os.environ.get("HF_USERNAME"),
+    token=os.environ.get("HF_TOKEN"),
+)
+
+
+backend = "local"
+project = AutoTrainProject(params=params, backend=backend, process=True)
+project.create()
+```
+
+In this example, we are finetuning the `meta-llama/Llama-3.2-1B-Instruct` model on the `HuggingFaceH4/no_robots` dataset.
+We are training the model for 3 epochs with a batch size of 1 and a learning rate of `1e-5`.
+We are using the `paged_adamw_8bit` optimizer and the `cosine` scheduler.
+We are also using mixed precision training with a gradient accumulation of 8.
+The final model will be pushed to the Hugging Face Hub after training.
+
+To train the model, run the following command:
+
+```bash
+$ export HF_USERNAME=<your-hf-username>
+$ export HF_TOKEN=<your-hf-write-token>
+$ python train.py
+```
+
+This will create a new project directory with the name `autotrain-llama32-1b-finetune` and start the training process.
+Once the training is complete, the model will be pushed to the Hugging Face Hub.
+
+Your HF_TOKEN and HF_USERNAME are only required if you want to push the model or if you are accessing a gated model or dataset.
+
+## AutoTrainProject Class
+
+[[autodoc]] project.AutoTrainProject
+
+## Parameters
+
+### Text Tasks
+
+[[autodoc]] trainers.clm.params.LLMTrainingParams
+
+[[autodoc]] trainers.sent_transformers.params.SentenceTransformersParams
+
+[[autodoc]] trainers.seq2seq.params.Seq2SeqParams
+
+[[autodoc]] trainers.token_classification.params.TokenClassificationParams
+
+[[autodoc]] trainers.extractive_question_answering.params.ExtractiveQuestionAnsweringParams
+
+[[autodoc]] trainers.text_classification.params.TextClassificationParams
+
+[[autodoc]] trainers.text_regression.params.TextRegressionParams
+
+### Image Tasks
+
+[[autodoc]] trainers.image_classification.params.ImageClassificationParams
+
+[[autodoc]] trainers.image_regression.params.ImageRegressionParams
+
+[[autodoc]] trainers.object_detection.params.ObjectDetectionParams
+
+[[autodoc]] trainers.dreambooth.params.DreamBoothTrainingParams
+
+### Tabular Tasks
+
+[[autodoc]] trainers.tabular.params.TabularParams

docs/source/tasks/sentence_transformer.mdx

@@ -68,3 +68,8 @@ For `qa` training, the data should be in the following format:
 | how are you | I am fine |
 | What is your name? | My name is Abhishek |
 | Which is the best programming language? | Python |
+
+
+## Parameters
+
+[[autodoc]] trainers.sent_transformers.params.SentenceTransformersParams

notebooks/python_example.py

@@ -0,0 +1,36 @@
+import os
+
+from autotrain.params import LLMTrainingParams
+from autotrain.project import AutoTrainProject
+
+
+params = LLMTrainingParams(
+    model="meta-llama/Llama-3.2-1B-Instruct",
+    data_path="HuggingFaceH4/no_robots",
+    chat_template="tokenizer",
+    text_column="messages",
+    train_split="train",
+    trainer="sft",
+    epochs=3,
+    batch_size=1,
+    lr=1e-5,
+    peft=True,
+    quantization="int4",
+    target_modules="all-linear",
+    padding="right",
+    optimizer="paged_adamw_8bit",
+    scheduler="cosine",
+    gradient_accumulation=8,
+    mixed_precision="bf16",
+    merge_adapter=True,
+    project_name="autotrain-llama32-1b-finetune",
+    log="tensorboard",
+    push_to_hub=False,
+    username=os.environ.get("HF_USERNAME"),
+    token=os.environ.get("HF_TOKEN"),
+)
+
+
+backend = "local"
+project = AutoTrainProject(params=params, backend=backend, process=True)
+project.create()

setup.cfg

@@ -17,6 +17,7 @@ max-line-length = 119
 per-file-ignores =
     # imported but unused
     __init__.py: F401, E402
+    src/autotrain/params.py: F401
 exclude =
     .git,
     .venv,

src/autotrain/cli/run_dreambooth.py

@@ -4,7 +4,7 @@
 
 from autotrain import logger
 from autotrain.cli import BaseAutoTrainCommand
-from autotrain.cli.utils import common_args, dreambooth_munge_data
+from autotrain.cli.utils import common_args
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.dreambooth.params import DreamBoothTrainingParams
 from autotrain.trainers.dreambooth.utils import VALID_IMAGE_EXTENSIONS, XL_MODELS
@@ -387,7 +387,6 @@ def __init__(self, args):
     def run(self):
         logger.info("Running DreamBooth Training")
         params = DreamBoothTrainingParams(**vars(self.args))
-        params = dreambooth_munge_data(params, local=self.args.backend.startswith("local"))
-        project = AutoTrainProject(params=params, backend=self.args.backend)
+        project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
         job_id = project.create()
         logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_extractive_qa.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import ext_qa_munge_data, get_field_info
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.extractive_question_answering.params import ExtractiveQuestionAnsweringParams
 
@@ -100,7 +100,6 @@ def run(self):
         logger.info("Running Extractive Question Answering")
         if self.args.train:
             params = ExtractiveQuestionAnsweringParams(**vars(self.args))
-            params = ext_qa_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_image_classification.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, img_clf_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.image_classification.params import ImageClassificationParams
 
@@ -108,7 +108,6 @@ def run(self):
         logger.info("Running Image Classification")
         if self.args.train:
             params = ImageClassificationParams(**vars(self.args))
-            params = img_clf_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_image_regression.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, img_reg_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.image_regression.params import ImageRegressionParams
 
@@ -108,7 +108,6 @@ def run(self):
         logger.info("Running Image Regression")
         if self.args.train:
             params = ImageRegressionParams(**vars(self.args))
-            params = img_reg_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_llm.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, llm_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.clm.params import LLMTrainingParams
 
@@ -136,7 +136,6 @@ def run(self):
         logger.info("Running LLM")
         if self.args.train:
             params = LLMTrainingParams(**vars(self.args))
-            params = llm_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_object_detection.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, img_obj_detect_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.object_detection.params import ObjectDetectionParams
 
@@ -108,7 +108,6 @@ def run(self):
         logger.info("Running Object Detection")
         if self.args.train:
             params = ObjectDetectionParams(**vars(self.args))
-            params = img_obj_detect_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_sent_tranformers.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, sent_transformers_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.sent_transformers.params import SentenceTransformersParams
 
@@ -108,7 +108,6 @@ def run(self):
         logger.info("Running Sentence Transformers...")
         if self.args.train:
             params = SentenceTransformersParams(**vars(self.args))
-            params = sent_transformers_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_seq2seq.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, seq2seq_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.seq2seq.params import Seq2SeqParams
 
@@ -92,7 +92,6 @@ def run(self):
         logger.info("Running Seq2Seq Classification")
         if self.args.train:
             params = Seq2SeqParams(**vars(self.args))
-            params = seq2seq_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_tabular.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, tabular_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.tabular.params import TabularParams
 
@@ -101,7 +101,6 @@ def run(self):
         logger.info("Running Tabular Training")
         if self.args.train:
             params = TabularParams(**vars(self.args))
-            params = tabular_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_text_classification.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, text_clf_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.text_classification.params import TextClassificationParams
 
@@ -101,7 +101,6 @@ def run(self):
         logger.info("Running Text Classification")
         if self.args.train:
             params = TextClassificationParams(**vars(self.args))
-            params = text_clf_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")

src/autotrain/cli/run_text_regression.py

@@ -1,7 +1,7 @@
 from argparse import ArgumentParser
 
 from autotrain import logger
-from autotrain.cli.utils import get_field_info, text_reg_munge_data
+from autotrain.cli.utils import get_field_info
 from autotrain.project import AutoTrainProject
 from autotrain.trainers.text_regression.params import TextRegressionParams
 
@@ -101,7 +101,6 @@ def run(self):
         logger.info("Running Text Regression")
         if self.args.train:
             params = TextRegressionParams(**vars(self.args))
-            params = text_reg_munge_data(params, local=self.args.backend.startswith("local"))
-            project = AutoTrainProject(params=params, backend=self.args.backend)
+            project = AutoTrainProject(params=params, backend=self.args.backend, process=True)
             job_id = project.create()
             logger.info(f"Job ID: {job_id}")