MinimalGPT
v2.0.0
https://github.com/abhaskumarsinha/Corpus2GPT
Logotipo MinimalGPT" ancho="20%" estilo="ancho máximo: 100%;">
[ GPT-1 Paper ] [ 1002 short stories from project guttenberg ] [ logo.com ] [ Transformer - Paper ] [ Huggingface Transformers ] [ TensorFlow ] [ BPE Tokenizer: subword-nmt ]
MinimalGPT es un marco de código conciso, adaptable y optimizado que abarca los componentes esenciales necesarios para la construcción, entrenamiento, inferencia y ajuste del modelo GPT. Este marco se implementa exclusivamente utilizando Keras y TensorFlow, lo que garantiza la compatibilidad y coherencia dentro del ecosistema de aprendizaje profundo más amplio.
NUEVO: ¡Soporte de CPU/GPU/TPU y soporte para cargar conjuntos de datos de archivos grandes!
En el repositorio, introducimos dos archivos integrales que componen nuestro marco propuesto. El primer archivo, GPT.py , sirve como marco fundamental y abarca componentes cruciales como bloques y capas. Estos componentes abarcan atención de múltiples cabezales, mecanismos de retroalimentación, atención de producto escalado, codificación posicional, salida softmaxed y una función de inferencia para la predicción del modelo. El segundo archivo, MinimalGPT .py , agiliza la utilización de nuestro marco al ofrecer una interfaz de línea de comandos concisa. Esta interfaz permite a los usuarios realizar sin esfuerzo operaciones esenciales, incluida la creación, el entrenamiento, el guardado, la carga, el ajuste y la inferencia del modelo, todo condensado en una única ejecución de línea de comando. Además, los archivos se pueden importar cómodamente al código Python, lo que permite a los usuarios incorporarlos sin problemas en sus proyectos mediante una simple llamada de función.
pip install -r requirements.txt
La arquitectura del modelo se rige por varios parámetros críticos, incluidos GPT_INPUT, D_MODEL, MULTI_HEAD y DECODER_STACKS . Es imperativo garantizar la coherencia en estos parámetros para evitar problemas relacionados con la carga del modelo para procesos posteriores de reentrenamiento o inferencia. En situaciones en las que surge incertidumbre, consultar el archivo de configuración generado durante la ejecución anterior puede proporcionar información valiosa. Además, los parámetros VOCABULARY_START y VOCABULARY_END desempeñan un papel crucial en la definición de los marcadores de ventana para el corpus. Estos marcadores ayudan a generar la capa Vectorizador, que extrae el vocabulario del corpus dentro de los recuentos de tokens INICIO y FINALIZACIÓN especificados. Es esencial tener en cuenta que los tokens dentro del corpus están separados por espacios en blanco y la inclusión de VOCABULARY_START y VOCABULARY_END se vuelve especialmente relevante cuando no se especifica explícitamente un archivo de token.
Además, tenga en cuenta que AMBOS, el archivo tokenizador y los pesos del modelo, se guardan/cargan a la vez. Actualmente, el código no admite guardar/cargar estos dos archivos por separado.
El modo de inferencia (-i) no solo requiere parámetros del modelo y un tokenizador y un archivo de pesos guardados para generar datos de inferencia. Debe usarse con el interruptor (-ol).
usage: MinimalGPT .py [-h] [-d DATA_PATH] [-l LEARNING_RATE]
[-ol OUTPUT_LENGTH] [-e EPOCHS] [-b BATCH_SIZE]
[-s GPT_INPUT] [-dm D_MODEL] [-p MULTI_HEAD]
[-ds DECODER_STACKS] [-ts TOKEN_START] [-te TOKEN_END]
[-vs VOCABULARY_START] [-ve VOCABULARY_END] [-sd]
[-lt LOAD_TOKENIZER] [-lw LOAD_WEIGHTS]
[-st SAVE_TOKENIZER] [-sw SAVE_WEIGHTS] [-ot OPTIMIZER]
[-i] [-mv] [-mvo]
optional arguments:
-h, --help show this help message and exit
-d DATA_PATH, --data-path DATA_PATH
File: Corresponding to corpus or training text
[String]
-l LEARNING_RATE, --learning-rate LEARNING_RATE
Float: Learning Rate. The model will train ONLY IF the
rate is > 0, skip otherwise [Float]
-ol OUTPUT_LENGTH, --output-length OUTPUT_LENGTH
Length of the output sequence to be generated
-e EPOCHS, --epochs EPOCHS
Number of training Epochs [Int]
-b BATCH_SIZE, --batch-size BATCH_SIZE
Size of each batch [Int]
-s GPT_INPUT, --gpt-input GPT_INPUT
Number of Tokens of text the model inputs at a time
[Int]
-dm D_MODEL, --d-model D_MODEL
Embedding layer output dimensions [Int]
-p MULTI_HEAD, --multi-head MULTI_HEAD
Number of Multi-head Attention layer in parallel [Int]
-ds DECODER_STACKS, --decoder-stacks DECODER_STACKS
Number of stacked Decoder layer [Int]
-ts TOKEN_START, --token-start TOKEN_START
The token number in the corpus to mark it as the
starting point of the training [Int]
-te TOKEN_END, --token-end TOKEN_END
The token number in the corpus to mark it as the end
point of the training [Int]
-vs VOCABULARY_START, --vocabulary-start VOCABULARY_START
Token number from the corpus to mark the starting
point of vocabulary data [Int]
-ve VOCABULARY_END, --vocabulary-end VOCABULARY_END
Token number from the corpus to mark the end point of
vocabulary data [Int]
-sd, --save Save the Model and Vectorizer data to disk
[True/False]
-lt LOAD_TOKENIZER, --load-tokenizer LOAD_TOKENIZER
File: Vectorization layer [File]
-lw LOAD_WEIGHTS, --load-weights LOAD_WEIGHTS
File: Model Weights [File]
-st SAVE_TOKENIZER, --save-tokenizer SAVE_TOKENIZER
File: Saving Vectorizer File [File]
-sw SAVE_WEIGHTS, --save-weights SAVE_WEIGHTS
File: Saving Model Weights[File]
-ot OPTIMIZER, --optimizer OPTIMIZER
Optimizer consistent to TensorFlow optimizer class
[tf.keras.optimizers]
-i, --inference-only Only Print the output of the model in Inference Mode
[True/False]
-mv, --model-vectorizer
Return Model, Vectorizer Tuple [True/False]
-mvo, --model-vectorizer-output
Return Model, Vectorizer, Output Tuple [True/False]
Suponiendo que las especificaciones del modelo deseadas implican GPT_INPUT = 10, D_MODEL = 128, MULTI_HEAD = 8 y DECODER_STACKS = 1, y el rango de token de corpus para el entrenamiento abarca desde TOKEN_START = 0 a TOKEN_END = 40000, y genere la capa de vectorizador a partir del intervalo de corpus de VOCABULARY_START = 0 a VOCABULARY_END = 200000, se ejecuta el siguiente comando para iniciar el proceso de entrenamiento del modelo. Los pesos resultantes y los datos del tokenizador se guardan en la carpeta designada. Los resultados siguientes ilustran el resultado de la ejecución de este comando.
PS C:gpt> python MinimalGPT .py -d './dataset/output_dataset.txt' -l 0.001 -ol 200 -e 4 -b 512 -s 10 -dm 128 -p 8 -ds 1 -ts 0 -te 40000 -vs 0 -ve 200000 -sd -st './models/tokenizer.mgt' -sw './models/weights.mgw'
Total tokens: 40000
100%|██████████████████████████████████████████████████████████████████████████████| 200000/200000 [02:02<00:00, 1636.38it/s]
New Vectorizer created successfully...
Vocabulary Size: 14270
100%|██████████████████████████████████████████████████████████████████████████████| 39989/39989 [00:00<00:00, 302926.25it/s]
100%|█████████████████████████████████████████████████████████████████████████████| 39989/39989 [00:00<00:00, 1289942.19it/s]
(None, 10, 128)
Epoch 1/4
79/79 [==============================] - 88s 1s/step - loss: 7.8692
Epoch 2/4
79/79 [==============================] - 92s 1s/step - loss: 3.8066
Epoch 3/4
79/79 [==============================] - 93s 1s/step - loss: 1.1487
Epoch 4/4
79/79 [==============================] - 92s 1s/step - loss: 0.2900
100%|██████████████████████████████████████████████████████████████████████████████████████| 190/190 [00:05<00:00, 34.70it/s]
Vocabulary size saved: 14270
and her eyes in the library. She was the rather large woman, although not fat, and when she wore high heels--which sh
e was not prone to do, because although Cutter would not have cared, she kept trying to project into other people's minds and
trying, as she said, "Not to do anything to them, that I wouldn't want them to do you me."--she rose a good inch above Cutter.
She was pleasant humored, and cooperative, and the one great irritant about her that annoyed Cutter, was the fact that she wa
s not capable of meeting life wholeheartedly and with strength. She steadily worried about other people's feelings and thought
s, so that Cutter wondered if she were capable of the slightest personal conviction. Yet that weakness was an advantage at the
same time, to him, because she worked constantly toward making him happy. The house was run to his minutest liking, and the s
ervants liked her, so that while she did not use a strong enough
Supongamos que queremos ajustar el modelo anterior (o volver a entrenarlo), luego el comando para volver a cargar el tokenizador y los pesos y volver a entrenarlo en un nuevo texto de un rango de ventana específico del corpus se proporciona a continuación:
PS C:gpt> python MinimalGPT .py -d './dataset/output_dataset.txt' -l 0.00005 -ol 200 -e 1 -b 512 -s 10 -dm 128 -p 8 -ds 1 -ts 80000 -te 120000 -sd -st './models/tokenizer2.mgt' -sw './models/weights2.mgw' -lt './models/tokenizer.mgt' -lw './models/weights.mgw'
Total tokens: 40000
100%|██████████████████████████████████████████████████████████████████████████████| 39989/39989 [00:00<00:00, 302923.51it/s]
100%|█████████████████████████████████████████████████████████████████████████████| 39989/39989 [00:00<00:00, 1428099.68it/s]
(None, 10, 128)
79/79 [==============================] - 81s 993ms/step - loss: 7.9725
100%|██████████████████████████████████████████████████████████████████████████████████████| 190/190 [00:06<00:00, 30.29it/s]
Vocabulary size saved: 14270
of her own the black of my own and my wife had could seen the house at the same moment her mind caught the first sugg
estion of the folded paper. “But he must have a name! Where is the paper?” She moved to the desk, and began to turn over the s
cattered documents that littered it. The first that caught her eye was an unfinished letter in her husband’s hand, with his pe
n lying across it, as though dropped there at a sudden summons. “My dear Parvis,”--who was Parvis?--“I have just received your
letter announcing Elwell’s death, and while I suppose there is now no farther risk of trouble, it might be safer--” That was
all. The “risk of trouble” was easily explained by the newspaper clipping which had apprised Mary of the suit brought against
her husband by one of his associates in the Blue Star enterprise. The only new information conveyed in the letter was the fact
of its showing Boyne,
El modo de inferencia implica la carga de pesos y vectorizadores previamente entrenados. Luego, estos componentes se utilizan para ejecutar el modelo, generando resultados de una longitud específica según lo especificado.
PS C:gpt> python MinimalGPT .py -i -ol 500 -e 6 -b 512 -s 10 -dm 128 -p 8 -ds 1 -lt './models/tokenizer2.mgt' -lw './models/weights2.mgw'
(None, 10, 128)
100%|██████████████████████████████████████████████████████████████████████████████████████| 490/490 [00:13<00:00, 35.93it/s]
of her own “on the other from the inel’--a little sensational, of course. But I guess you’d better look it over.” He
held out a newspaper to Mary, who unfolded it slowly, remembering, as she did so, the evening when, in that same room, the per
usal of a clipping from the “Sentinel” had first shaken the depths of her security. As she opened the paper, her eyes, shrinki
ng from the glaring head-lines, “Widow of Boyne’s Victim Forced to Appeal for Aid,” ran down the column of text to two portrai
ts inserted in it. The first was her husband’s, taken from a photograph made the year they had come to England. It was the pic
ture of him that she liked best, the one that stood on the writing-table up-stairs in her bedroom. As the eyes in the photogra
ph met hers, she felt it would be impossible to read what was said of him, and closed her lids with the sharpness of the pain.
“I thought if you felt disposed to put your name down--” she heard Parvis continue. She opened her eyes with an effort, and t
hey fell on the other portrait. It was that of a youngish man, slightly built, in rough clothes, with features somewhat blurre
d by the shadow of a projecting hat-brim. Where had she seen that outline before? She stared at it confusedly, her heart hamme
ring in her throat and ears. Then she gave a cry. “This is the man--the man who came for my husband!” She heard Parvis start t
o his feet, and was dimly aware that she had slipped backward into the corner of the sofa, and that he was bending above her i
n alarm. With an intense effort she straightened herself, and reached out for the paper, which she had dropped. “It’s the man!
I should know him anywhere!” she cried in a voice that sounded in her own ears like a scream. Parvis’s voice seemed to come t
o her from far off, down endless, fog-muffled windings. “Mrs. Boyne, you’re not very well. Shall I call somebody? Shall I get
a glass of water?” “No, no, no!” She threw herself toward him, her hand frantically clenching the newspaper. “I tell you, it’s
the man! I KNOW him! He spoke to me in the garden!” Parvis took the journal from her, directing his glasses to the portrait.
“It can’t be, Mrs. Boyne. It’s Robert Elwell.” “Robert Elwell?” Her white
Incorporar los modelos entrenados generados mediante la utilización de MinimalGPT .py en su proyecto es un proceso sencillo que se facilita importando la función MinimalGPT y configurándola de acuerdo con las especificaciones deseadas. Esto se puede lograr configurando los parámetros return_model_and_vectorizer = True o return_model_and_vectorizer_and_output = True dentro del marco inference_only = True (modo de inferencia). Además, el entrenamiento, la creación y la exportación del modelo se pueden lograr utilizando un enfoque similar, en paralelo al modo de línea de comandos. Para obtener una ilustración completa de estos procedimientos, el Jupyter Notebook adjunto proporciona una demostración ejemplar.
from MinimalGPT import MinimalGPT model = MinimalGPT (output_length = 200, gpt_input = 10, d_model = 128, h = 8, decoder_stacks = 1, load_tokenizer = './models/tokenizer3.mgt', load_weights = './models/weights3.mgw', inference_only = True, return_model_and_vectorizer_and_output = True) model[0].summary()
Model: "model"
Layer (type) Output Shape Param
================================================================= input_1 (InputLayer) [(None, 10)] 0
embedding (Embedding) (None, 10, 128) 1826816
positional_embedding (Posit (None, 10, 128) 0
ionalEmbedding)
decoder (Decoder) (None, 10, 128) 37160
flatten (Flatten) (None, 1280) 0
dense (Dense) (None, 14273) 18283713
tf.nn.softmax (TFOpLambda) (None, 14273) 0
================================================================= Total params: 20,147,689 Trainable params: 20,147,689 Non-trainable params: 0
El modelo implementado aquí difiere un poco en comparación con la implementación original en papel. La matriz formada después de concatenar las cabezas de la salida del producto escalado se multiplica por el parámetro de matriz de tamaño clave dimensión x d_modelo. A efectos prácticos, este pequeño ajuste para reducir la cantidad de parámetros conduciría a un pequeño aumento en el rendimiento debido a la optimización de parámetros entrenables.
