diff --git a/modules/training.py b/modules/training.py index bc5b3878..f63f2990 100644 --- a/modules/training.py +++ b/modules/training.py @@ -25,35 +25,40 @@ def get_json_dataset(path: str): def create_train_interface(): with gr.Tab('Train LoRA', elem_id='lora-train-tab'): - loraName = gr.Textbox(label="Name", info="The name of your new LoRA file") + lora_name = gr.Textbox(label="Name", info="The name of your new LoRA file") with gr.Row(): # TODO: Implement multi-device support. - microBatchSize = gr.Slider(label='Micro Batch Size', value=4, minimum=1, maximum=128, step=1, info='Per-device batch size (NOTE: multiple devices not yet implemented). Increasing this will increase VRAM usage.') - batchSize = gr.Slider(label='Batch Size', value=128, minimum=1, maximum=1024, step=4, info='Global batch size. The two batch sizes together determine gradient accumulation (gradientAccum = batch / microBatch). Higher gradient accum values lead to better quality training.') + micro_batch_size = gr.Slider(label='Micro Batch Size', value=4, minimum=1, maximum=128, step=1, info='Per-device batch size (NOTE: multiple devices not yet implemented). Increasing this will increase VRAM usage.') + batch_size = gr.Slider(label='Batch Size', value=128, minimum=1, maximum=1024, step=4, info='Global batch size. The two batch sizes together determine gradient accumulation (gradientAccum = batch / microBatch). Higher gradient accum values lead to better quality training.') + with gr.Row(): epochs = gr.Number(label='Epochs', value=1, info='Number of times every entry in the dataset should be fed into training. So 1 means feed each item in once, 5 means feed it in five times, etc.') - learningRate = gr.Textbox(label='Learning Rate', value='3e-4', info='Learning rate, in scientific notation. 3e-4 is a good starting base point. 1e-2 is extremely high, 1e-6 is extremely low.') + learning_rate = gr.Textbox(label='Learning Rate', value='3e-4', info='Learning rate, in scientific notation. 3e-4 is a good starting base point. 1e-2 is extremely high, 1e-6 is extremely low.') + # TODO: What is the actual maximum rank? Likely distinct per model. This might be better to somehow be on a log scale. - loraRank = gr.Slider(label='LoRA Rank', value=8, minimum=1, maximum=1024, step=4, info='LoRA Rank, or dimension count. Higher values produce a larger file with better control over the model\'s content. Smaller values produce a smaller file with less overall control. Small values like 4 or 8 are great for stylistic guidance, high values like 128 or 256 are good for teaching content upgrades. Higher ranks also require higher VRAM.') - loraAlpha = gr.Slider(label='LoRA Alpha', value=16, minimum=1, maximum=2048, step=4, info='LoRA Alpha. This divided by the rank becomes the scaling of the LoRA. Higher means stronger. A good standard value is twice your Rank.') + lora_rank = gr.Slider(label='LoRA Rank', value=8, minimum=1, maximum=1024, step=4, info='LoRA Rank, or dimension count. Higher values produce a larger file with better control over the model\'s content. Smaller values produce a smaller file with less overall control. Small values like 4 or 8 are great for stylistic guidance, high values like 128 or 256 are good for teaching content upgrades. Higher ranks also require higher VRAM.') + lora_alpha = gr.Slider(label='LoRA Alpha', value=16, minimum=1, maximum=2048, step=4, info='LoRA Alpha. This divided by the rank becomes the scaling of the LoRA. Higher means stronger. A good standard value is twice your Rank.') # TODO: Better explain what this does, in terms of real world effect especially. - loraDropout = gr.Slider(label='LoRA Dropout', minimum=0.0, maximum=1.0, step=0.025, value=0.05, info='Percentage probability for dropout of LoRA layers.') - cutoffLen = gr.Slider(label='Cutoff Length', minimum=1,maximum=2048, value=256, step=32, info='Cutoff length for text input. Essentially, how long of a line of text to feed in at a time. Higher values require drastically more VRAM.') + lora_dropout = gr.Slider(label='LoRA Dropout', minimum=0.0, maximum=1.0, step=0.025, value=0.05, info='Percentage probability for dropout of LoRA layers.') + cutoff_len = gr.Slider(label='Cutoff Length', minimum=1,maximum=2048, value=256, step=32, info='Cutoff length for text input. Essentially, how long of a line of text to feed in at a time. Higher values require drastically more VRAM.') + with gr.Row(): - datasetFunction = get_json_dataset('training/datasets') - dataset = gr.Dropdown(choices=datasetFunction(), value='None', label='Dataset', info='The dataset file to use for training.') - ui.create_refresh_button(dataset, lambda : None, lambda : {'choices': datasetFunction()}, 'refresh-button') - evalDataset = gr.Dropdown(choices=datasetFunction(), value='None', label='Evaluation Dataset', info='The dataset file used to evaluate the model after training.') - ui.create_refresh_button(evalDataset, lambda : None, lambda : {'choices': datasetFunction()}, 'refresh-button') - formatsFunction = get_json_dataset('training/formats') - format = gr.Dropdown(choices=formatsFunction(), value='None', label='Data Format', info='The format file used to decide how to format the dataset input.') - ui.create_refresh_button(format, lambda : None, lambda : {'choices': formatsFunction()}, 'refresh-button') + dataset_function = get_json_dataset('training/datasets') + dataset = gr.Dropdown(choices=dataset_function(), value='None', label='Dataset', info='The dataset file to use for training.') + ui.create_refresh_button(dataset, lambda : None, lambda : {'choices': dataset_function()}, 'refresh-button') + eval_dataset = gr.Dropdown(choices=dataset_function(), value='None', label='Evaluation Dataset', info='The dataset file used to evaluate the model after training.') + ui.create_refresh_button(eval_dataset, lambda : None, lambda : {'choices': dataset_function()}, 'refresh-button') + formats_function = get_json_dataset('training/formats') + format = gr.Dropdown(choices=formats_function(), value='None', label='Data Format', info='The format file used to decide how to format the dataset input.') + ui.create_refresh_button(format, lambda : None, lambda : {'choices': formats_function()}, 'refresh-button') + with gr.Row(): - startButton = gr.Button("Start LoRA Training") - stopButton = gr.Button("Interrupt") + start_button = gr.Button("Start LoRA Training") + stop_button = gr.Button("Interrupt") + output = gr.Markdown(value="Ready") - startEvent = startButton.click(do_train, [loraName, microBatchSize, batchSize, epochs, learningRate, loraRank, loraAlpha, loraDropout, cutoffLen, dataset, evalDataset, format], [output]) - stopButton.click(doInterrupt, [], [], cancels=[], queue=False) + startEvent = start_button.click(do_train, [lora_name, micro_batch_size, batch_size, epochs, learning_rate, lora_rank, lora_alpha, lora_dropout, cutoff_len, dataset, eval_dataset, format], [output]) + stop_button.click(doInterrupt, [], [], cancels=[], queue=False) def doInterrupt(): global WANT_INTERRUPT @@ -74,108 +79,119 @@ class Callbacks(transformers.TrainerCallback): control.should_epoch_stop = True control.should_training_stop = True -def cleanPath(basePath: str, path: str): +def cleanPath(base_path: str, path: str): """"Strips unusual symbols and forcibly builds a path as relative to the intended directory.""" # TODO: Probably could do with a security audit to guarantee there's no ways this can be bypassed to target an unwanted path. # Or swap it to a strict whitelist of [a-zA-Z_0-9] path = path.replace('\\', '/').replace('..', '_') - if basePath is None: + if base_path is None: return path - return f'{Path(basePath).absolute()}/{path}' + return f'{Path(base_path).absolute()}/{path}' -def do_train(loraName: str, microBatchSize: int, batchSize: int, epochs: int, learningRate: float, loraRank: int, loraAlpha: int, loraDropout: float, cutoffLen: int, dataset: str, evalDataset: str, format: str): +def do_train(lora_name: str, micro_batch_size: int, batch_size: int, epochs: int, learning_rate: float, lora_rank: int, lora_alpha: int, lora_dropout: float, cutoff_len: int, dataset: str, eval_dataset: str, format: str): global WANT_INTERRUPT, CURRENT_STEPS, MAX_STEPS, CURRENT_GRADIENT_ACCUM WANT_INTERRUPT = False CURRENT_STEPS = 0 MAX_STEPS = 0 - yield "Prepping..." + # == Input validation / processing == + yield "Prepping..." # TODO: --lora-dir PR once pulled will need to be applied here - loraName = f"loras/{cleanPath(None, loraName)}" + lora_name = f"loras/{cleanPath(None, lora_name)}" if dataset is None: return "**Missing dataset choice input, cannot continue.**" if format is None: return "**Missing format choice input, cannot continue.**" - gradientAccumulationSteps = batchSize // microBatchSize - CURRENT_GRADIENT_ACCUM = gradientAccumulationSteps - actualLR = float(learningRate) + gradient_accumulation_steps = batch_size // micro_batch_size + CURRENT_GRADIENT_ACCUM = gradient_accumulation_steps + actual_lr = float(learning_rate) shared.tokenizer.pad_token = 0 shared.tokenizer.padding_side = "left" + # == Prep the dataset, format, etc == with open(cleanPath('training/formats', f'{format}.json'), 'r') as formatFile: - formatData: dict[str, str] = json.load(formatFile) + format_data: dict[str, str] = json.load(formatFile) + def tokenize(prompt): - result = shared.tokenizer(prompt, truncation=True, max_length=cutoffLen + 1, padding="max_length") + result = shared.tokenizer(prompt, truncation=True, max_length=cutoff_len + 1, padding="max_length") return { "input_ids": result["input_ids"][:-1], "attention_mask": result["attention_mask"][:-1], } + def generate_prompt(data_point: dict[str, str]): - for options, data in formatData.items(): + for options, data in format_data.items(): if set(options.split(',')) == set(x[0] for x in data_point.items() if len(x[1].strip()) > 0): for key, val in data_point.items(): data = data.replace(f'%{key}%', val) return data - raise RuntimeError(f'Data-point "{data_point}" has no keyset match within format "{list(formatData.keys())}"') + raise RuntimeError(f'Data-point "{data_point}" has no keyset match within format "{list(format_data.keys())}"') + def generate_and_tokenize_prompt(data_point): prompt = generate_prompt(data_point) return tokenize(prompt) + print("Loading datasets...") data = load_dataset("json", data_files=cleanPath('training/datasets', f'{dataset}.json')) train_data = data['train'].shuffle().map(generate_and_tokenize_prompt) - if evalDataset == 'None': - evalData = None + + if eval_dataset == 'None': + eval_data = None else: - evalData = load_dataset("json", data_files=cleanPath('training/datasets', f'{evalDataset}.json')) - evalData = evalData['train'].shuffle().map(generate_and_tokenize_prompt) + eval_data = load_dataset("json", data_files=cleanPath('training/datasets', f'{eval_dataset}.json')) + eval_data = eval_data['train'].shuffle().map(generate_and_tokenize_prompt) + # == Start prepping the model itself == if not hasattr(shared.model, 'lm_head') or hasattr(shared.model.lm_head, 'weight'): print("Getting model ready...") prepare_model_for_int8_training(shared.model) + print("Prepping for training...") config = LoraConfig( - r=loraRank, - lora_alpha=loraAlpha, + r=lora_rank, + lora_alpha=lora_alpha, # TODO: Should target_modules be configurable? target_modules=[ "q_proj", "v_proj" ], - lora_dropout=loraDropout, + lora_dropout=lora_dropout, bias="none", task_type="CAUSAL_LM" ) - loraModel = get_peft_model(shared.model, config) + lora_model = get_peft_model(shared.model, config) trainer = transformers.Trainer( - model=loraModel, + model=lora_model, train_dataset=train_data, - eval_dataset=evalData, + eval_dataset=eval_data, args=transformers.TrainingArguments( - per_device_train_batch_size=microBatchSize, - gradient_accumulation_steps=gradientAccumulationSteps, + per_device_train_batch_size=micro_batch_size, + gradient_accumulation_steps=gradient_accumulation_steps, # TODO: Should more of these be configurable? Probably. warmup_steps=100, num_train_epochs=epochs, - learning_rate=actualLR, + learning_rate=actual_lr, fp16=True, logging_steps=20, - evaluation_strategy="steps" if evalData is not None else "no", + evaluation_strategy="steps" if eval_data is not None else "no", save_strategy="steps", - eval_steps=200 if evalData is not None else None, + eval_steps=200 if eval_data is not None else None, save_steps=200, - output_dir=loraName, + output_dir=lora_name, save_total_limit=3, - load_best_model_at_end=True if evalData is not None else False, + load_best_model_at_end=True if eval_data is not None else False, # TODO: Enable multi-device support ddp_find_unused_parameters=None ), data_collator=transformers.DataCollatorForLanguageModeling(shared.tokenizer, mlm=False), callbacks=list([Callbacks()]) ) - loraModel.config.use_cache = False - old_state_dict = loraModel.state_dict - loraModel.state_dict = ( + + lora_model.config.use_cache = False + old_state_dict = lora_model.state_dict + lora_model.state_dict = ( lambda self, *_, **__: get_peft_model_state_dict(self, old_state_dict()) - ).__get__(loraModel, type(loraModel)) + ).__get__(lora_model, type(lora_model)) + if torch.__version__ >= "2" and sys.platform != "win32": - loraModel = torch.compile(loraModel) + lora_model = torch.compile(lora_model) # == Main run and monitor loop == # TODO: save/load checkpoints to resume from? @@ -210,11 +226,11 @@ def do_train(loraName: str, microBatchSize: int, batchSize: int, epochs: int, le yield f"Running... **{CURRENT_STEPS}** / **{MAX_STEPS}** ... {timerInfo}, `{timeElapsed:.0f}`/`{totalTimeEstimate:.0f}` seconds" print("Training complete, saving...") - loraModel.save_pretrained(loraName) + lora_model.save_pretrained(lora_name) if WANT_INTERRUPT: print("Training interrupted.") - yield f"Interrupted. Incomplete LoRA saved to `{loraName}`" + yield f"Interrupted. Incomplete LoRA saved to `{lora_name}`" else: print("Training complete!") - yield f"Done! LoRA saved to `{loraName}`" + yield f"Done! LoRA saved to `{lora_name}`"