Update README.md
Browse files
README.md
CHANGED
@@ -1,201 +1,106 @@
|
|
1 |
---
|
2 |
library_name: transformers
|
3 |
-
|
|
|
|
|
|
|
4 |
---
|
5 |
|
6 |
-
# Model Card for Model ID
|
7 |
|
8 |
-
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
|
10 |
|
|
|
11 |
|
12 |
-
|
|
|
|
|
13 |
|
14 |
-
### Model Description
|
15 |
|
16 |
-
<!-- Provide a longer summary of what this model is. -->
|
17 |
|
18 |
-
|
19 |
|
20 |
-
-
|
21 |
-
- **Funded by [optional]:** [More Information Needed]
|
22 |
-
- **Shared by [optional]:** [More Information Needed]
|
23 |
-
- **Model type:** [More Information Needed]
|
24 |
-
- **Language(s) (NLP):** [More Information Needed]
|
25 |
-
- **License:** [More Information Needed]
|
26 |
-
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
|
28 |
-
|
|
|
|
|
|
|
|
|
29 |
|
30 |
-
|
31 |
|
32 |
-
- **
|
33 |
-
- **Paper [optional]:** [More Information Needed]
|
34 |
-
- **Demo [optional]:** [More Information Needed]
|
35 |
|
36 |
-
##
|
37 |
|
38 |
-
|
39 |
|
40 |
-
|
41 |
|
42 |
-
|
43 |
|
44 |
-
|
45 |
|
46 |
-
|
47 |
|
48 |
-
|
49 |
|
50 |
-
[
|
51 |
|
52 |
-
### Out-of-Scope Use
|
53 |
|
54 |
-
|
55 |
|
56 |
-
|
57 |
|
58 |
-
|
|
|
|
|
|
|
|
|
59 |
|
60 |
-
|
61 |
|
62 |
-
|
|
|
|
|
63 |
|
64 |
-
|
65 |
|
66 |
-
|
67 |
|
68 |
-
|
69 |
|
70 |
-
|
|
|
|
|
71 |
|
72 |
-
|
73 |
|
74 |
-
|
75 |
|
76 |
-
|
|
|
|
|
|
|
77 |
|
78 |
-
|
79 |
|
80 |
-
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
|
82 |
-
[More Information Needed]
|
83 |
|
84 |
-
|
85 |
|
86 |
-
|
87 |
-
|
88 |
-
#### Preprocessing [optional]
|
89 |
-
|
90 |
-
[More Information Needed]
|
91 |
-
|
92 |
-
|
93 |
-
#### Training Hyperparameters
|
94 |
-
|
95 |
-
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
-
|
97 |
-
#### Speeds, Sizes, Times [optional]
|
98 |
-
|
99 |
-
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
-
|
101 |
-
[More Information Needed]
|
102 |
-
|
103 |
-
## Evaluation
|
104 |
-
|
105 |
-
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
-
|
107 |
-
### Testing Data, Factors & Metrics
|
108 |
-
|
109 |
-
#### Testing Data
|
110 |
-
|
111 |
-
<!-- This should link to a Dataset Card if possible. -->
|
112 |
-
|
113 |
-
[More Information Needed]
|
114 |
-
|
115 |
-
#### Factors
|
116 |
-
|
117 |
-
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
-
|
119 |
-
[More Information Needed]
|
120 |
-
|
121 |
-
#### Metrics
|
122 |
-
|
123 |
-
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
-
|
125 |
-
[More Information Needed]
|
126 |
-
|
127 |
-
### Results
|
128 |
-
|
129 |
-
[More Information Needed]
|
130 |
-
|
131 |
-
#### Summary
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
## Model Examination [optional]
|
136 |
-
|
137 |
-
<!-- Relevant interpretability work for the model goes here -->
|
138 |
-
|
139 |
-
[More Information Needed]
|
140 |
-
|
141 |
-
## Environmental Impact
|
142 |
-
|
143 |
-
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
-
|
145 |
-
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
-
|
147 |
-
- **Hardware Type:** [More Information Needed]
|
148 |
-
- **Hours used:** [More Information Needed]
|
149 |
-
- **Cloud Provider:** [More Information Needed]
|
150 |
-
- **Compute Region:** [More Information Needed]
|
151 |
-
- **Carbon Emitted:** [More Information Needed]
|
152 |
-
|
153 |
-
## Technical Specifications [optional]
|
154 |
-
|
155 |
-
### Model Architecture and Objective
|
156 |
-
|
157 |
-
[More Information Needed]
|
158 |
-
|
159 |
-
### Compute Infrastructure
|
160 |
-
|
161 |
-
[More Information Needed]
|
162 |
|
163 |
#### Hardware
|
164 |
|
165 |
-
|
166 |
|
167 |
#### Software
|
168 |
|
169 |
-
|
170 |
-
|
171 |
-
## Citation [optional]
|
172 |
-
|
173 |
-
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
-
|
175 |
-
**BibTeX:**
|
176 |
-
|
177 |
-
[More Information Needed]
|
178 |
-
|
179 |
-
**APA:**
|
180 |
-
|
181 |
-
[More Information Needed]
|
182 |
-
|
183 |
-
## Glossary [optional]
|
184 |
-
|
185 |
-
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
-
|
187 |
-
[More Information Needed]
|
188 |
-
|
189 |
-
## More Information [optional]
|
190 |
-
|
191 |
-
[More Information Needed]
|
192 |
-
|
193 |
-
## Model Card Authors [optional]
|
194 |
-
|
195 |
-
[More Information Needed]
|
196 |
-
|
197 |
-
## Model Card Contact
|
198 |
-
|
199 |
-
[More Information Needed]
|
200 |
|
|
|
201 |
|
|
|
|
|
|
1 |
---
|
2 |
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
language:
|
5 |
+
- es
|
6 |
+
pipeline_tag: text-generation
|
7 |
---
|
8 |
|
|
|
9 |
|
|
|
10 |
|
11 |
|
12 |
+
# LLM-AviationV2: Innovaci贸n AI en los Cielos
|
13 |
|
14 |
+
<p align="center">
|
15 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/6419c2f6b4adb0e101b17b6c/fhU5yTQKH9nHN_zues186.png" style="width: 50%; max-width: 500px; height: auto;" alt="LLM-AviationV2: Innovaci贸n AI en los Cielos"/>
|
16 |
+
</p>
|
17 |
|
|
|
18 |
|
|
|
19 |
|
20 |
+
### Descripci贸n del Modelo
|
21 |
|
22 |
+
Desarrollado por Edison Bejarano y Nicolas Potes, este modelo representa un avance revolucionario en la utilizaci贸n de la tecnolog铆a de Modelos de Lenguaje (LM) dentro del sector aeron谩utico, espec铆ficamente dise帽ado para mejorar la comprensi贸n y accesibilidad del Reglamento Aeron谩utico Colombiano (RAC). Entrenado en una Tesla V100-SXM2-16GB, el modelo `LLM-AviationV2` se embarca en un viaje para navegar el complejo panorama regulatorio con una eficiencia y perspicacia sin precedentes.
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
|
24 |
+
- **Desarrollado por:** [Edison Bejarano](https://huggingface.co/ejbejaranos) - [Sergio Nicolas](https://huggingface.co/SergioMadridF)
|
25 |
+
- **Tipo de modelo:** Versi贸n afinada de `google/gemma-2b-it`
|
26 |
+
- **Idiomas (NLP):** Espa帽ol (es)
|
27 |
+
- **Licencia:** Apache-2.0
|
28 |
+
- **Afinado a partir del modelo:** `google/gemma-2b-it`
|
29 |
|
30 |
+
### Fuentes del Modelo
|
31 |
|
32 |
+
- **URL en Hugging Face:** [ejbejaranos/LLM-AviationV2](https://huggingface.co/ejbejaranos/LLM-AviationV2)
|
|
|
|
|
33 |
|
34 |
+
## Usos
|
35 |
|
36 |
+
### Uso Directo
|
37 |
|
38 |
+
El modelo `LLM-AviationV2` est谩 dise帽ado para aplicaciones directas en tareas de generaci贸n de texto, con el objetivo de simplificar la interpretaci贸n y aplicaci贸n de las regulaciones aeron谩uticas. Su funci贸n principal es servir a profesionales y entusiastas del campo de la aeron谩utica, proporcionando acceso inmediato a informaci贸n comprensible extra铆da del RAC.
|
39 |
|
40 |
+
## Detalles de Entrenamiento
|
41 |
|
42 |
+
## Datos de Entrenamiento
|
43 |
|
44 |
+
El modelo `LLM-AviationV2` fue afinado utilizando el dataset `RAC_Colombia_QualityImproved025`, el cual representa una versi贸n mejorada en t茅rminos de calidad del Reglamento Aeron谩utico Colombiano. Este dataset fue curado y mejorado por el equipo de [SomosNLP](https://huggingface.co/somosnlp), con el objetivo de proporcionar una base de datos m谩s precisa y relevante para tareas de procesamiento de lenguaje natural relacionadas con la aviaci贸n.
|
45 |
|
46 |
+
Para m谩s detalles sobre este dataset, puedes consultar la documentaci贸n y los metadatos a trav茅s del siguiente enlace:
|
47 |
|
48 |
+
[Dataset `RAC_Colombia_QualityImproved025` en Hugging Face](https://huggingface.co/datasets/somosnlp/RAC_Colombia_QualityImproved025)
|
49 |
|
|
|
50 |
|
51 |
+
### Procedimiento de Entrenamiento
|
52 |
|
53 |
+
#### Hiperpar谩metros de Entrenamiento
|
54 |
|
55 |
+
- **Tipo de GPU:** Tesla V100-SXM2-16GB
|
56 |
+
- **Tiempo Total de Entrenamiento:** Aprox. 31 minutos (1860 segundos)
|
57 |
+
- **Tasa de Aprendizaje:** 0.00005
|
58 |
+
- **Optimizador:** Paged AdamW 8bit
|
59 |
+
- **Pasos M谩ximos:** 516
|
60 |
|
61 |
+
#### Velocidades, Tama帽os, Tiempos
|
62 |
|
63 |
+
- **Tiempo de Entrenamiento:** 882.68 segundos
|
64 |
+
- **Muestras por Segundo en Entrenamiento:** 2.338
|
65 |
+
- **Pasos por Segundo en Entrenamiento:** 0.585
|
66 |
|
67 |
+
## Evaluaci贸n
|
68 |
|
69 |
+
### Resultados
|
70 |
|
71 |
+
El modelo ha demostrado una capacidad significativa para comprender y generar contenido regulatorio aeron谩utico en espa帽ol, convirti茅ndose en un valioso recurso para la industria.
|
72 |
|
73 |
+
<p align="center">
|
74 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/6419c2f6b4adb0e101b17b6c/dCaCtryupyHfxy7LCNxJ7.png" style="width: 50%; max-width: 500px; height: auto;" alt="M茅trica de perdida: Innovaci贸n AI en los Cielos"/>
|
75 |
+
</p>
|
76 |
|
77 |
+
## Impacto Ambiental
|
78 |
|
79 |
+
El entrenamiento de `LLM-AviationV2` se llev贸 a cabo con una consideraci贸n cuidadosa de su huella ambiental, optimizando para la eficiencia y minimizando el gasto computacional innecesario.
|
80 |
|
81 |
+
- **Tipo de Hardware:** Tesla V100-SXM2-16GB
|
82 |
+
- **Horas Utilizadas:** Aproximadamente 0.52 horas
|
83 |
+
- **Consumo de Energ铆a:** Aproximadamente 0.156 kWh
|
84 |
+
- **Emisiones de CO2 Estimadas:** Aproximadamente 0.0741 kg
|
85 |
|
86 |
+
Estas cifras subrayan nuestro compromiso con la sostenibilidad y la reducci贸n del impacto ambiental en el desarrollo de tecnolog铆as de inteligencia artificial.
|
87 |
|
|
|
88 |
|
|
|
89 |
|
90 |
+
## Especificaciones T茅cnicas
|
91 |
|
92 |
+
### Infraestructura de C贸mputo
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
|
94 |
#### Hardware
|
95 |
|
96 |
+
El entrenamiento se realiz贸 en una Tesla V100-SXM2-16GB, elegida por su equilibrio entre rendimiento y eficiencia energ茅tica.
|
97 |
|
98 |
#### Software
|
99 |
|
100 |
+
- **Versi贸n de Transformers:** 4.38.0
|
101 |
+
- **Entorno de Entrenamiento:** Proporcionado por la biblioteca Hugging Face Transformers.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
102 |
|
103 |
+
## M谩s Informaci贸n
|
104 |
|
105 |
+
Para obtener informaci贸n m谩s detallada sobre `LLM-AviationV2`, incluido el acceso al modelo y sus capacidades completas, por favor visita nuestro [repositorio en Hugging Face](https://huggingface.co/ejbejaranos/LLM-AviationV2).
|
106 |
+
LLM-AviationV2).
|