DETAILED ACTION Notice of Pre-AIA or AIA Status The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. This action is responsive to the application filed on August 4 th , 2023 . Claims 1-20 are pending in the case. Claims 1, 8, and 15 are independent claims. Information Disclosure Statement The information disclosure statement filed August 21 st , 2023 fails to comply with 37 CFR 1.98(a)(2), which requires a legible copy of each cited foreign patent document; each non-patent literature publication or that portion which caused it to be listed; No copy of “ HOCHREITER, S.; and Schmidhuber , J. 1997. "Long short-term memory". Neural computation, 9(8): 1735–1780. ” ( l isted as citation No. 8 ) has been included, therefore, it has not been considered by the examiner. The remainder of the information disclosure statement has been considered. Claim Rejections - 35 USC § 101 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Regarding claim SEQ claimNum 1 : Step 1 : Claim SEQ claimNum \c 1 is directed to [a] method , therefore it falls under the statuary category of a process. Step 2A Prong 1 : The claim recites, in part: “ partitioning the time series into a plurality of partitions ” this encompasses the mental portioning of observed time series into a plurality of partitions. Further, this limitation is a mathematical concept. “ generating…a plurality of tokens, wherein the plurality of tokens are based on the time series ” this encompasses the mental creation of a plurality of tokens based on observed time series. “ generating…a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens ” this encompasses the mental creation of a transformer vector based on observed relationships among observed tokens. “ assigning…a classification to the transformer vector ” this encompasses the mental assignment of a classification to an observed transformer vector. Step 2A Prong 2 : The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows: “ receiving…a time series ” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). “ at a forecasting platform ” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h). “ processing the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ processing the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2). Step 2B : The additional elements “ at a forecasting platform ”, “ processing the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ processing the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” , taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Further, “ receiving…a time series ” limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc. , 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015). as well as receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec , 838 F.3d at 1321, 120 USPQ2d. See MPEP § 2106.05(d)/(II). Therefore, the claim is ineligible. Regarding claim SEQ claimNum 2 , the rejection of claim 1 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ each of the plurality of tokens corresponds to a partition of the plurality of partitions ” a continuation of the abstract idea identified in the parent claim. Step 2A Prong 2 : The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more. Regarding claim SEQ claimNum 3 , the rejection of claim 2 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ the plurality of tokens includes a multidimensional vector space ” this limitation is a mathematical concept. Step 2A Prong 2 : The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more. Regarding claim SEQ claimNum 4 , the rejection of claim 3 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ a number of dimensions in the multidimensional vector space corresponds to a number of patterns ” a continuation of the abstract idea identified in the parent claim. Step 2A Prong 2 : The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows: “ the convolutional neural network machine learning model is trained to predict ” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h). Step 2B : The additional elements, taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Therefore, the claim is ineligible. Regarding claim SEQ claimNum 5 , the rejection of claim 4 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ partition position embedding value is added to a value of the multidimensional vector space ” this limitation is a mathematical concept. Step 2A Prong 2 : The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more. Regarding claim SEQ claimNum 6 , the rejection of claim 1 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ the classification is a sign prediction ” a continuation of the abstract idea identified in the parent claim. Step 2A Prong 2 : The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more. Regarding claim SEQ claimNum 7 , the rejection of claim 6 is incorporated and further : Step 2A Prong 1 : The claim recites, in part: “ the sign prediction is an upward indication ” a continuation of the abstract idea identified in the parent claim. Step 2A Prong 2 : The claim does not recite any additional limitations, thus does not further recite any additional elements that integrates the judicial exception into a practical application or amount to significantly more. Regarding claim SEQ claimNum 8 : Step 1 : Claim SEQ claimNum \c 8 is directed to [a] system , therefore it falls under the statuary category of a machine. Step 2A Prong 1 : The claim recites, in part: “ partition the time series into a plurality of partitions ” this encompasses the mental portioning of observed time series into a plurality of partitions. Further, this limitation is a mathematical concept. “ generate…a plurality of tokens, wherein the plurality of tokens are based on the time series ” this encompasses the mental creation of a plurality of tokens based on observed time series. “ generate…a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens ” this encompasses the mental creation of a transformer vector based on observed relationships among observed tokens. “ assign…a classification to the transformer vector ” this encompasses the mental assignment of a classification to an observed transformer vector. Step 2A Prong 2 : The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows: “ receive…a time series ” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). “ at a forecasting platform ” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h). “ process the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ process the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2). Step 2B : The additional elements “ at a forecasting platform ”, “ processing the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ processing the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” , taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Further, “ receiving…a time series ” limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc. , 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015). as well as receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec , 838 F.3d at 1321, 120 USPQ2d. See MPEP § 2106.05(d)/(II). Therefore, the claim is ineligible. Regarding claims 9-14: The rejection of claim SEQ claimNum \c 8 is further incorporated, the rejection of claims 2-7 are applicable to claims 9-14, respectively. Regarding claim 15: Step 1 : Claim 15 is directed to [a] non-transitory computer readable storage medium , therefore it falls under the statuary category of a manufacture. Step 2A Prong 1 : The claim recites, in part: “ partitioning the time series into a plurality of partitions ” this encompasses the mental portioning of observed time series into a plurality of partitions. Further, this limitation is a mathematical concept. “ generating…a plurality of tokens, wherein the plurality of tokens are based on the time series ” this encompasses the mental creation of a plurality of tokens based on observed time series. “ generating…a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens ” this encompasses the mental creation of a transformer vector based on observed relationships among observed tokens. “ assigning…a classification to the transformer vector ” this encompasses the mental assignment of a classification to an observed transformer vector. Step 2A Prong 2 : The judicial exception is not integrated into a practical application; the remaining limitations of the claim are as follows: “ receiving…a time series ” the limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). “ at a forecasting platform ” the limitation is an additional element that generally links the use of the judicial exception to a particular technological environment or field of use. See MPEP § 2106.05(h). “ processing the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ processing the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” the limitations are an additional element that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or merely uses a computer in its ordinary capacity as a tool to perform an existing process. See MPEP § 2106.05(f)(2). Step 2B : The additional elements “ at a forecasting platform ”, “ processing the time series with a convolutional neural network machine learning model ”, “ by the convolutional neural network machine learning model ”, “ processing the plurality of tokens with a transformer machine learning model ”, “ by the transformer machine learning model ”, “ determined by the transformer machine learning model ”, “ by a multilayer perceptron classifier ” , taken individually and in combination, do not provide an inventive concept of significantly more than the abstract idea itself for the reasons set forth in step 2A prong 2 above. Further, “ receiving…a time series ” limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Furthermore the additional element is directed to storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc. , 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015). as well as receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec , 838 F.3d at 1321, 120 USPQ2d. See MPEP § 2106.05(d)/(II). Therefore, the claim is ineligible. Regarding claims 16-20: The rejection of claim 15 is further incorporated, the rejection of claims 2-6 are applicable to claims 16-20, respectively. The limitation is an additional element that amounts to adding insignificant extra-solution activity to the judicial exception. See MPEP § 2106.05(g). Claim Rejections - 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claims 1-5, 8-12 and 15-19 are rejected under 35 U.S.C. 103 as being unpatentable over Shen et al. (“TCCT: Tightly-coupled convolutional transformer on time series forecasting”, Shen et al., 21 January 2022) ( hereinafter “Shen”) in view of Dosovitskiy et al. (“An Image Is Worth 16x16 Words: Transformers For Image Recognition At Scale”, Dosovitskiy et al., 3 Jun 2021) ( as cited in the IDS, hereinafter “ Dosovitskiy ”) in further view of Wu et al. (“Visual Transformers: Token-based Image Representation and Processing for Computer Vision”, Wu et al., 20 Nov 2020) ( hereinafter “Wu”). Regarding claim SEQ claimMapNum 1 : Shen teaches [a] method comprising: receiving, at a forecasting platform, a time series (Shen, pages 2-3, col 2-1, section 3.1, ¶1 “Suppose we have a fixed input window z i,1: t 0 ⅈ=1 N ,the task is to predict corresponding fixed target window z i, t 0 +1: t 0 +T ⅈ=1 N .N refers to the number of related univariate timeseries, t0 is the input window size and T is the prediction window size.”) ; processing the time series with a convolutional neural network machine learning model (Shen, page 5, col 1, section 4.2, ¶1 “Informer employs a convolutional layer and a max-pooling layer between each two self-attention blocks to trim the input length.”) ; processing the plurality of tokens with a transformer machine learning model (Shen, page 6, col 2, section 4.4, ¶1 “All above architectures can seamlessly cooperate with Transformer or Transformer-like time series forecasting models, including canonical Transformer, LogTrans , Informer, etc.”) ; generating, by the transformer machine learning model, a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens determined by the transformer machine learning model (Shen, page 3, col 1, section 3.2, ¶2 “ ProbSparse self-attention allows each key only attend to dominant queries while performing the scaled dot-product to improve the efficiency. The way to judge the dominant queries is through the Kullback-Leibler divergence of query-key attention probability distribution and the uniform distribution. Queries owning larger KL divergence are regarded as more dominant ones.” furthermore, “No extra encoders are added and all three feature maps outputted by three self-attention blocks are fused and then transited to the final output of proper dimension.”) ; Shen does not teach “ partitioning the time series into a plurality of partitions; assigning, by a multilayer perceptron classifier, a classification to the transformer vector ” However, Dosovitskiy teaches partitioning the time series into a plurality of partitions ( Dosovitskiy , page 3, Figure 1, ¶1 “We split an image into fixed-size patches, linearly embed each of them, add position embeddings, and feed the resulting sequence of vectors to a standard Transformer encoder.”) ; assigning, by a multilayer perceptron classifier, a classification to the transformer vector ( Dosovitskiy , page 3, section 3.1, ¶2 “The classification head is implemented by a MLP with one hidden layer at pre-training time and by a single linear layer at fine-tuning time.”) Shen and Dosovitskiy are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen’s system to incorporate the partitions and MLP taught by Dosovitskiy . The motivation for doing so would have been to attain excellent results with fewer computational resources as stated in Dosovitskiy , page 1, Abstract “ Vision Transformer ( ViT ) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train. ” . Shen in view of Dosovitskiy does not teach “ generating, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series ” However, Wu teaches generating, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series (Wu, page 3, col 2, section 3.1, ¶1 “Formally, we denote the input feature map by X∈ R HW×C (height H, width W, channels C) and visual tokens by T∈ R L ×C s.t. L≪HW (L represents the number of tokens).” Furthermore Wu, page 3, col 2, section 3.1.1, ¶1 “A filter-based tokenizer, also adopted by [47, 6, 26], utilizes convolutions to extract visual tokens.” in light of the specification, ¶31 “In accordance with aspects, a CNN model may act as a filter that passes over the input time series.”) ; Shen in view of Dosovitskiy and Wu are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen/ Dosovitskiy ’s system to incorporate the convolutional neural network token processing taught by Wu . The motivation for doing so would have been Wu, page 1, col 2, ¶3 “ To overcome the above challenges, we address the root cause, the pixel-convolution paradigm, and introduce the Visual Transformer (VT) (Figure 1), a new paradigm to rep resent and process high-level concepts in images. ” . Regarding claim SEQ claimMapNum 2 : Shen in view of Dosovitskiy in further view of Wu teaches [t]he method of claim 1, wherein each of the plurality of tokens corresponds to a partition of the plurality of partitions ( Dosovitskiy , page 3, section 3.1, ¶2 “To handle 2D images, we reshape the image x∈ R H×W×C into a sequence of flattened 2D patches x P ∈ R N× P 2 ⋅C , where (H,W) is the resolution of the original image, C is the number of channels, (P,P) is the resolution of each image patch, and N = HW/P 2 is the resulting number of patches, which also serves as the effective input sequence length for the Transformer.” Here, each patch can be considered a partition which corresponds to the input sequence length ) Regarding claim SEQ claimMapNum 3 : Shen in view of Dosovitskiy in further view of Wu teaches [t]he method of claim 2, wherein each of the plurality of tokens includes a multidimensional vector space (Shen, page 3, col 2, section 4.1, ¶1 “The input ϵ R L×D , where L is the input length and d is the input dimension, is split into two parts through dimension X= X 1 L× ⅆ 1 , X 2 L× d 2 .”) . Regarding claim SEQ claimMapNum 4 : Shen in view of Dosovitskiy in further view of Wu teaches [t]he method of claim 3, wherein a number of dimensions in the multidimensional vector space corresponds to a number of patterns that the convolutional neural network machine learning model is trained to predict (Wu, page 3, col 2, section 3.1, ¶1 “Formally, we denote the input feature map by X∈ R HW×C (height H, width W, channels C) and visual tokens by T∈ R L ×C s.t. L≪HW (L represents the number of tokens).” here the number of dimensions corresponding to the number of channels can be considered the dimensions corresponding to the number of patterns in light of the specification, ¶33 “For instance, if a CNN model has been trained to detect 10 discrete patterns in various partitions of an input time series, then the patterns may be labeled or identified as 1st, 2nd, 3rd, . . . 10th in an ordinal sequence of recognizable patterns. Moreover, an output token for a corresponding partition in a time series input may include a 10-dimensional vector space.”) . Regarding claim SEQ claimMapNum 5 : Shen in view of Dosovitskiy in further view of Wu teaches [t]he method of claim 4, wherein a partition position embedding value is added to a value of the multidimensional vector space ( Dosovitskiy , page 3, section 3.1, ¶3 “Position embeddings are added to the patch embeddings to retain positional information. We use standard learnable 1D position embeddings, since we have not observed significant performance gains from using more advanced 2D-aware position embeddings (Appendix D.4). The resulting sequence of embedding vectors serves as input to the encoder.”) . Regarding claim SEQ claimMapNum 8 : Shen teaches [a] system comprising at least one computer including a processor and a memory (Shen, page 7, col 2, section 5.3, ¶1 “All experiments are conducted on a single Nvidia GTX 1080Ti 12GB GPU.”) , wherein the at least one computer is configured to: receive, at a forecasting platform, a time series (Shen, pages 2-3, col 2-1, section 3.1, ¶1 “Suppose we have a fixed input window z i,1: t 0 ⅈ=1 N ,the task is to predict corresponding fixed target window z i, t 0 +1: t 0 +T ⅈ=1 N .N refers to the number of related univariate timeseries, t0 is the input window size and T is the prediction window size.”) ; process the time series with a convolutional neural network machine learning model (Shen, page 5, col 1, section 4.2, ¶1 “Informer employs a convolutional layer and a max-pooling layer between each two self-attention blocks to trim the input length.”) ; process the plurality of tokens with a transformer machine learning model (Shen, page 6, col 2, section 4.4, ¶1 “All above architectures can seamlessly cooperate with Transformer or Transformer-like time series forecasting models, including canonical Transformer, LogTrans , Informer, etc.”) ; generate, by the transformer machine learning model, a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens determined by the transformer machine learning model (Shen, page 3, col 1, section 3.2, ¶2 “ ProbSparse self-attention allows each key only attend to dominant queries while performing the scaled dot-product to improve the efficiency. The way to judge the dominant queries is through the Kullback-Leibler divergence of query-key attention probability distribution and the uniform distribution. Queries owning larger KL divergence are regarded as more dominant ones.” furthermore, “No extra encoders are added and all three feature maps outputted by three self-attention blocks are fused and then transited to the final output of proper dimension.”) ; Shen does not teach “ partition the time series into a plurality of partitions; assigning, by a multilayer perceptron classifier, a classification to the transformer vector ” However, Dosovitskiy teaches partitioning the time series into a plurality of partitions ( Dosovitskiy , page 3, Figure 1, ¶1 “We split an image into fixed-size patches, linearly embed each of them, add position embeddings, and feed the resulting sequence of vectors to a standard Transformer encoder.”) ; assign, by a multilayer perceptron classifier, a classification to the transformer vector ( Dosovitskiy , page 3, section 3.1, ¶2 “The classification head is implemented by a MLP with one hidden layer at pre-training time and by a single linear layer at fine-tuning time.”) Shen and Dosovitskiy are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen’s system to incorporate the partitions and MLP taught by Dosovitskiy . The motivation for doing so would have been to attain excellent results with fewer computational resources as stated in Dosovitskiy , page 1, Abstract “ Vision Transformer ( ViT ) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train. ” . Shen in view of Dosovitskiy does not teach “ generate, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series ” However, Wu teaches generate, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series (Wu, page 3, col 2, section 3.1, ¶1 “Formally, we denote the input feature map by X∈ R HW×C (height H, width W, channels C) and visual tokens by T∈ R L ×C s.t. L≪HW (L represents the number of tokens).” Furthermore Wu, page 3, col 2, section 3.1.1, ¶1 “A filter-based tokenizer, also adopted by [47, 6, 26], utilizes convolutions to extract visual tokens.” in light of the specification, ¶31 “In accordance with aspects, a CNN model may act as a filter that passes over the input time series.”) ; Shen in view of Dosovitskiy and Wu are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen/ Dosovitskiy ’s system to incorporate the convolutional neural network token processing taught by Wu . The motivation for doing so would have been Wu, page 1, col 2, ¶3 “ To overcome the above challenges, we address the root cause, the pixel-convolution paradigm, and introduce the Visual Transformer (VT) (Figure 1), a new paradigm to rep resent and process high-level concepts in images. ” . Regarding claims 9-12: Claims 9-12 are rejected under the same rationale as claims 2-5. Regarding claim 15: Shen teaches [a] non-transitory computer readable storage medium, including instructions ( s (Shen, page 7, col 2, ¶1 “The source code is available at https://github.com/OrigamiSL/TCCT2021”) stored thereon, which instructions, when read and executed by one or more computer processors (Shen, page 7, col 2, section 5.3, ¶1 “All experiments are conducted on a single Nvidia GTX 1080Ti 12GB GPU.”) , cause the one or more computer processors to perform steps comprising:: receiving, at a forecasting platform, a time series (Shen, pages 2-3, col 2-1, section 3.1, ¶1 “Suppose we have a fixed input window z i,1: t 0 ⅈ=1 N ,the task is to predict corresponding fixed target window z i, t 0 +1: t 0 +T ⅈ=1 N .N refers to the number of related univariate timeseries, t0 is the input window size and T is the prediction window size.”) ; processing the time series with a convolutional neural network machine learning model (Shen, page 5, col 1, section 4.2, ¶1 “Informer employs a convolutional layer and a max-pooling layer between each two self-attention blocks to trim the input length.”) ; processing the plurality of tokens with a transformer machine learning model (Shen, page 6, col 2, section 4.4, ¶1 “All above architectures can seamlessly cooperate with Transformer or Transformer-like time series forecasting models, including canonical Transformer, LogTrans , Informer, etc.”) ; generating, by the transformer machine learning model, a transformer vector, wherein the transformer vector is based on relationships among the plurality of tokens determined by the transformer machine learning model (Shen, page 3, col 1, section 3.2, ¶2 “ ProbSparse self-attention allows each key only attend to dominant queries while performing the scaled dot-product to improve the efficiency. The way to judge the dominant queries is through the Kullback-Leibler divergence of query-key attention probability distribution and the uniform distribution. Queries owning larger KL divergence are regarded as more dominant ones.” furthermore, “No extra encoders are added and all three feature maps outputted by three self-attention blocks are fused and then transited to the final output of proper dimension.”) ; Shen does not teach “ partitioning the time series into a plurality of partitions; assigning, by a multilayer perceptron classifier, a classification to the transformer vector ” However, Dosovitskiy teaches partitioning the time series into a plurality of partitions ( Dosovitskiy , page 3, Figure 1, ¶1 “We split an image into fixed-size patches, linearly embed each of them, add position embeddings, and feed the resulting sequence of vectors to a standard Transformer encoder.”) ; assigning, by a multilayer perceptron classifier, a classification to the transformer vector ( Dosovitskiy , page 3, section 3.1, ¶2 “The classification head is implemented by a MLP with one hidden layer at pre-training time and by a single linear layer at fine-tuning time.”) Shen and Dosovitskiy are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen’s system to incorporate the partitions and MLP taught by Dosovitskiy . The motivation for doing so would have been to attain excellent results with fewer computational resources as stated in Dosovitskiy , page 1, Abstract “ Vision Transformer ( ViT ) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train. ” . Shen in view of Dosovitskiy does not teach “ generating, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series ” However, Wu teaches generating, by the convolutional neural network machine learning model, a plurality of tokens, wherein the plurality of tokens are based on the time series (Wu, page 3, col 2, section 3.1, ¶1 “Formally, we denote the input feature map by X∈ R HW×C (height H, width W, channels C) and visual tokens by T∈ R L ×C s.t. L≪HW (L represents the number of tokens).” Furthermore Wu, page 3, col 2, section 3.1.1, ¶1 “A filter-based tokenizer, also adopted by [47, 6, 26], utilizes convolutions to extract visual tokens.” in light of the specification, ¶31 “In accordance with aspects, a CNN model may act as a filter that passes over the input time series.”) ; Shen in view of Dosovitskiy and Wu are analogous art because both references concern methods for convolutional neural network-transformer hybrid architectures . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen/ Dosovitskiy ’s system to incorporate the convolutional neural network token processing taught by Wu . The motivation for doing so would have been Wu, page 1, col 2, ¶3 “ To overcome the above challenges, we address the root cause, the pixel-convolution paradigm, and introduce the Visual Transformer (VT) (Figure 1), a new paradigm to rep resent and process high-level concepts in images. ” . Regarding claims 16-19 Claims 16-19 are rejected under the same rationale as claims 2-5. Claims 6, 7, 13, 14 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Shen in view of Dosovitskiy in view of Wu in further view of Sirignano et al. (“Universal features of price formation in financial markets: perspectives from Deep Learning”, Sirignano et al., 19 Mar 2018) ( hereinafter “ Sirignano ”). Regarding claim SEQ claimMapNum 6 : Shen in view of Dosovitskiy in further view of Wu teaches [t]he method of claim 1 Shen in view of Dosovitskiy in view of Wu does not teach “ wherein the classification is a sign prediction ” However, Sirignano teaches wherein the classification is a sign prediction ( Sirignano , page 10, figure 4, ¶1 “Models are trained to predict the direction {−1,+1} of next mid-price move.”) . Shen in view of Dosovitskiy in further view of Wu and Sirignano are analogous art because both references concern methods for forecasting . Accordingly, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Shen/ Dosovitskiy /Wu ’s system to incorporate the sign prediction taught by Sirignano . The motivation for doing so would have been to have a stable out-of-sample prediction as stated in Sirignano , page 1, abstract “ The universal price formation model exhibits a remarkably stable out-of-sample prediction accuracy across time, for a wide range of stocks from different sectors. ” . Regarding claim SEQ claimMapNum 7 : 838200 1545590 0 0 Shen in view of Dosovitskiy in view of Wu in further view of Sirignano teaches [t]he method of claim 6, wherein the sign prediction is an upward indication ( Sirignano , page 9, section 3,1 ¶3, “The linear (VAR) model may be formulated as follows: at each observation we update a vector of linear features h t and then use a probit model for the conditional probability of an upward price move given the state variables: where G depends on the distributional assumptions on the innovations in the linear model.”) . Regarding claims 13 and 14: Claims 13 and 14 are rejected under the same rationale as claims 6 and 7. Regarding claim 20: Claim 20 is rejected under the same rationale as claim 6. Conclusion The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Rostamian et al. (" Event prediction within directional change framework using a CNN-LSTM model ", Rostamian et al., 16 August 2022 ) teaches a CNN-LSTM model to investigate its predictive competence within the Directional Change (DC) framework to predict DC event prices. Any inquiry concerning this communication or earlier communications from the examiner should be directed to FILLIN "Enter examiner's name" \* MERGEFORMAT JACOB Z SUSSMAN MOSS whose telephone number is (571) 272-1579 . The examiner can normally be reached FILLIN "Work schedule?" \* MERGEFORMAT Monday - Friday, 9 a.m. - 5 p.m. ET . Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, FILLIN "SPE Name?" \* MERGEFORMAT Kakali Chaki can be reached on FILLIN "SPE Phone?" \* MERGEFORMAT (571) 272-3719 . The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /J.S.M./ Examiner, Art Unit 2122 /KAKALI CHAKI/ Supervisory Patent Examiner, Art Unit 2122