मेटा के लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़न वेब सेवाएँ

मेटा के लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़न वेब सेवाएँ

आज, हमें यह घोषणा करते हुए खुशी हो रही है कि मेटा द्वारा विकसित लामा 2 फाउंडेशन मॉडल ग्राहकों के लिए उपलब्ध हैं अमेज़न SageMaker जम्पस्टार्ट. बड़े भाषा मॉडल (एलएलएम) का लामा 2 परिवार 7 अरब से 70 अरब मापदंडों के पैमाने पर पूर्व-प्रशिक्षित और ठीक-ठीक जेनरेटिव टेक्स्ट मॉडल का एक संग्रह है। फाइन-ट्यून किए गए एलएलएम, जिन्हें लामा-2-चैट कहा जाता है, संवाद उपयोग के मामलों के लिए अनुकूलित हैं। आप इन मॉडलों को आसानी से आज़मा सकते हैं और सेजमेकर जंपस्टार्ट के साथ उनका उपयोग कर सकते हैं, जो एक मशीन लर्निंग (एमएल) हब है जो एल्गोरिदम, मॉडल और एमएल समाधान तक पहुंच प्रदान करता है ताकि आप जल्दी से एमएल के साथ शुरुआत कर सकें।

इस पोस्ट में, हम सेजमेकर जम्पस्टार्ट के माध्यम से लामा 2 मॉडल का उपयोग करने के तरीके के बारे में जानेंगे।

लामा 2 क्या है?

लामा 2 एक ऑटो-रिग्रेसिव भाषा मॉडल है जो एक अनुकूलित ट्रांसफार्मर आर्किटेक्चर का उपयोग करता है। लामा 2 अंग्रेजी में व्यावसायिक और अनुसंधान उपयोग के लिए है। यह पैरामीटर आकारों की श्रेणी में आता है - 7 अरब, 13 अरब, और 70 अरब - साथ ही पूर्व-प्रशिक्षित और सुव्यवस्थित विविधताएं। मेटा के अनुसार, ट्यून किए गए संस्करण सहायता और सुरक्षा के लिए मानवीय प्राथमिकताओं को संरेखित करने के लिए पर्यवेक्षित फाइन-ट्यूनिंग (एसएफटी) और मानव प्रतिक्रिया (आरएलएचएफ) के साथ सुदृढीकरण सीखने का उपयोग करते हैं। लामा 2 को सार्वजनिक रूप से उपलब्ध स्रोतों से 2 ट्रिलियन टोकन डेटा पर पूर्व-प्रशिक्षित किया गया था। ट्यून किए गए मॉडल सहायक-जैसी चैट के लिए अभिप्रेत हैं, जबकि पूर्व-प्रशिक्षित मॉडल को विभिन्न प्राकृतिक भाषा निर्माण कार्यों के लिए अनुकूलित किया जा सकता है। इस बात पर ध्यान दिए बिना कि डेवलपर मॉडल के किस संस्करण का उपयोग करता है मेटा से जिम्मेदार उपयोग मार्गदर्शिका अतिरिक्त फ़ाइन-ट्यूनिंग का मार्गदर्शन करने में सहायता कर सकता है जो उचित सुरक्षा शमन के साथ मॉडलों को अनुकूलित और अनुकूलित करने के लिए आवश्यक हो सकता है।

सेजमेकर जम्पस्टार्ट क्या है?

सेजमेकर जम्पस्टार्ट के साथ, एमएल व्यवसायी ओपन सोर्स फाउंडेशन मॉडल के व्यापक चयन में से चुन सकते हैं। एमएल प्रैक्टिशनर फाउंडेशन मॉडल को समर्पित करने के लिए तैनात कर सकते हैं अमेज़न SageMaker नेटवर्क पृथक वातावरण से उदाहरण और मॉडल प्रशिक्षण और तैनाती के लिए सेजमेकर का उपयोग करके मॉडल को अनुकूलित करें।

अब आप कुछ ही क्लिक में लामा 2 को खोज और तैनात कर सकते हैं अमेज़ॅन सैजमेकर स्टूडियो या सेजमेकर पायथन एसडीके के माध्यम से प्रोग्रामेटिक रूप से, आपको सेजमेकर सुविधाओं के साथ मॉडल प्रदर्शन और एमएलओपीएस नियंत्रण प्राप्त करने में सक्षम बनाता है जैसे अमेज़न SageMaker पाइपलाइन, अमेज़न SageMaker डिबगर, या कंटेनर लॉग। मॉडल को AWS सुरक्षित वातावरण में और आपके VPC नियंत्रण में तैनात किया गया है, जिससे डेटा सुरक्षा सुनिश्चित करने में मदद मिलती है। लामा 2 मॉडल आज शुरुआत में अमेज़न सेजमेकर स्टूडियो में उपलब्ध हैं us-east 1 और us-west 2 क्षेत्रों के।

मॉडल खोजें

आप सेजमेकर स्टूडियो यूआई और सेजमेकर पायथन एसडीके में सेजमेकर जम्पस्टार्ट के माध्यम से फाउंडेशन मॉडल तक पहुंच सकते हैं। इस अनुभाग में, हम सेजमेकर स्टूडियो में मॉडलों की खोज कैसे करें, इसके बारे में जानेंगे।

सेजमेकर स्टूडियो एक एकीकृत विकास वातावरण (आईडीई) है जो एकल वेब-आधारित विज़ुअल इंटरफ़ेस प्रदान करता है जहां आप डेटा तैयार करने से लेकर निर्माण, प्रशिक्षण और अपने एमएल मॉडल को तैनात करने तक सभी एमएल विकास चरणों को करने के लिए उद्देश्य-निर्मित टूल तक पहुंच सकते हैं। सेजमेकर स्टूडियो कैसे शुरू करें और स्थापित करें, इसके बारे में अधिक जानकारी के लिए देखें अमेज़ॅन सैजमेकर स्टूडियो.

एक बार जब आप सेजमेकर स्टूडियो पर हों, तो आप सेजमेकर जम्पस्टार्ट तक पहुंच सकते हैं, जिसमें पूर्व-प्रशिक्षित मॉडल, नोटबुक और पूर्वनिर्मित समाधान शामिल हैं। पूर्वनिर्मित और स्वचालित समाधान.

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.

सेजमेकर जम्पस्टार्ट लैंडिंग पृष्ठ से, आप समाधान, मॉडल, नोटबुक और अन्य संसाधन ब्राउज़ कर सकते हैं। आप इसमें दो प्रमुख लामा 2 मॉडल पा सकते हैं फाउंडेशन मॉडल: टेक्स्ट जनरेशन हिंडोला. यदि आपको लामा 2 मॉडल दिखाई नहीं देते हैं, तो बंद करके और पुनः आरंभ करके अपने सेजमेकर स्टूडियो संस्करण को अपडेट करें। संस्करण अपडेट के बारे में अधिक जानकारी के लिए देखें स्टूडियो ऐप्स बंद करें और अपडेट करें.

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.

आप चुनकर अन्य चार मॉडल वेरिएंट भी पा सकते हैं सभी टेक्स्ट जेनरेशन मॉडल का अन्वेषण करें या खोज रहे हैं llama खोज बॉक्स में

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.

आप मॉडल के बारे में विवरण देखने के लिए मॉडल कार्ड चुन सकते हैं जैसे लाइसेंस, प्रशिक्षण के लिए उपयोग किया गया डेटा और उपयोग कैसे करें। आप दो बटन भी पा सकते हैं, तैनाती और नोटबुक खोलें, जो आपको मॉडल का उपयोग करने में मदद करता है।

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.

जब आप कोई भी बटन चुनते हैं, तो एक पॉप-अप आपको स्वीकार करने के लिए अंतिम-उपयोगकर्ता लाइसेंस अनुबंध और स्वीकार्य उपयोग नीति दिखाएगा।

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.

स्वीकार करने पर, आप मॉडल का उपयोग करने के लिए अगले चरण पर आगे बढ़ेंगे।

एक मॉडल तैनात करें

जब आप चुनते हैं तैनाती और शर्तें स्वीकार करें, मॉडल परिनियोजन शुरू हो जाएगा। वैकल्पिक रूप से, आप उदाहरण नोटबुक के माध्यम से तैनात कर सकते हैं जो चुनने पर दिखाई देता है नोटबुक खोलें. उदाहरण नोटबुक अनुमान लगाने और संसाधनों को साफ करने के लिए मॉडल को कैसे तैनात किया जाए, इस पर शुरू से अंत तक मार्गदर्शन प्रदान करता है।

नोटबुक का उपयोग करके तैनात करने के लिए, हम द्वारा निर्दिष्ट एक उपयुक्त मॉडल का चयन करके शुरुआत करते हैं model_id. आप निम्नलिखित कोड के साथ सेजमेकर पर किसी भी चयनित मॉडल को तैनात कर सकते हैं:

from sagemaker.jumpstart.model import JumpStartModel
my_model = JumpStartModel(model_id = "meta-textgeneration-llama-2-70b-f")
predictor = my_model.deploy()

यह मॉडल को सेजमेकर पर डिफ़ॉल्ट कॉन्फ़िगरेशन के साथ तैनात करता है, जिसमें डिफ़ॉल्ट इंस्टेंस प्रकार और डिफ़ॉल्ट वीपीसी कॉन्फ़िगरेशन शामिल हैं। आप गैर-डिफ़ॉल्ट मान निर्दिष्ट करके इन कॉन्फ़िगरेशन को बदल सकते हैं जम्पस्टार्टमॉडल. इसे तैनात करने के बाद, आप सेजमेकर भविष्यवक्ता के माध्यम से तैनात समापन बिंदु के खिलाफ अनुमान चला सकते हैं:

payload = { “inputs”: [ [ {"role": "system", "content": "Always answer with Haiku"}, {"role": "user", "content": "I am going to Paris, what should I see?"}, ] ], "parameters":{"max_new_tokens":256, "top_p":0.9, "temperature":0.6}
}

फाइन-ट्यून किए गए चैट मॉडल (Llama-2-7b-chat, Llama-2-13b-chat, Llama-2-70b-chat) उपयोगकर्ता और चैट सहायक के बीच चैट के इतिहास को स्वीकार करते हैं, और बाद की चैट उत्पन्न करते हैं। पूर्व-प्रशिक्षित मॉडल (Llama-2-7b, Llama-2-13b, Llama-2-70b) को एक स्ट्रिंग प्रॉम्प्ट की आवश्यकता होती है और दिए गए प्रॉम्प्ट पर टेक्स्ट पूरा करना होता है। निम्नलिखित कोड देखें:

predictor.predict(payload, custom_attributes="accept_eula=true")

ध्यान दें कि डिफ़ॉल्ट रूप से, accept_eula गलत पर सेट है. आपको सेट करना होगा accept_eula=true समापन बिंदु को सफलतापूर्वक लागू करने के लिए। ऐसा करके, आप उपयोगकर्ता लाइसेंस अनुबंध और स्वीकार्य उपयोग नीति को स्वीकार करते हैं जैसा कि पहले बताया गया है। आप भी कर सकते हैं डाउनलोड लाइसेंस समझौता.

Custom_attributes EULA को पास करने के लिए कुंजी/मूल्य जोड़े का उपयोग किया जाता है। कुंजी और मान को अलग किया जाता है = और जोड़ियों को अलग किया जाता है ;. यदि उपयोगकर्ता एक ही कुंजी को एक से अधिक बार पास करता है, तो अंतिम मान रखा जाता है और स्क्रिप्ट हैंडलर को भेज दिया जाता है (यानी, इस मामले में, सशर्त तर्क के लिए उपयोग किया जाता है)। उदाहरण के लिए, यदि accept_eula=false; accept_eula=true फिर सर्वर को भेज दिया जाता है  accept_eula=true रखा जाता है और स्क्रिप्ट हैंडलर को भेज दिया जाता है।

अनुमान पैरामीटर अंतिम बिंदु पर पाठ निर्माण प्रक्रिया को नियंत्रित करते हैं। अधिकतम नए टोकन नियंत्रण मॉडल द्वारा उत्पन्न आउटपुट के आकार को संदर्भित करता है। ध्यान दें कि यह शब्दों की संख्या के समान नहीं है क्योंकि मॉडल की शब्दावली अंग्रेजी भाषा की शब्दावली के समान नहीं है, और प्रत्येक टोकन अंग्रेजी भाषा का शब्द नहीं हो सकता है। तापमान आउटपुट में यादृच्छिकता को नियंत्रित करता है। उच्च तापमान के परिणामस्वरूप अधिक रचनात्मक और मतिभ्रमपूर्ण आउटपुट मिलते हैं। सभी अनुमान पैरामीटर वैकल्पिक हैं.

निम्नलिखित तालिका सेजमेकर जम्पस्टार्ट में उपलब्ध सभी लामा मॉडलों को सूचीबद्ध करती है model_ids, डिफ़ॉल्ट उदाहरण प्रकार, और इनमें से प्रत्येक मॉडल के लिए समर्थित कुल टोकन की अधिकतम संख्या (इनपुट टोकन की संख्या और उत्पन्न टोकन की संख्या का योग)।

मॉडल का नाम मॉडल आईडी अधिकतम कुल टोकन डिफ़ॉल्ट इंस्टेंस प्रकार
लामा-2-7बी मेटा-टेक्स्टजेनरेशन-लामा-2-7बी 4096 एमएल.g5.2xबड़ा
लामा-2-7बी-चैट मेटा-टेक्स्टजेनरेशन-लामा-2-7बी-एफ 4096 एमएल.g5.2xबड़ा
लामा-2-13बी मेटा-टेक्स्टजेनरेशन-लामा-2-13बी 4096 एमएल.g5.12xबड़ा
लामा-2-13बी-चैट मेटा-टेक्स्टजेनरेशन-लामा-2-13बी-एफ 4096 एमएल.g5.12xबड़ा
लामा-2-70बी मेटा-टेक्स्टजेनरेशन-लामा-2-70बी 4096 एमएल.g5.48xबड़ा
लामा-2-70बी-चैट मेटा-टेक्स्टजेनरेशन-लामा-2-70बी-एफ 4096 एमएल.g5.48xबड़ा

ध्यान दें कि सेजमेकर एंडपॉइंट्स की टाइमआउट सीमा 60s है। इस प्रकार, भले ही मॉडल 4096 टोकन उत्पन्न करने में सक्षम हो, यदि पाठ निर्माण में 60 से अधिक समय लगता है, तो अनुरोध विफल हो जाएगा। 7बी, 13बी और 70बी मॉडल के लिए, हम सेट करने की अनुशंसा करते हैं max_new_tokens टोकन की कुल संख्या 1500K से कम रखते हुए क्रमशः 1000, 500 और 4 से अधिक नहीं।

लामा-2-70बी के लिए अनुमान और उदाहरण संकेत

आप पाठ के किसी भी भाग को पूरा करने के लिए लामा मॉडल का उपयोग कर सकते हैं। पाठ निर्माण के माध्यम से, आप विभिन्न प्रकार के कार्य कर सकते हैं, जैसे प्रश्नों का उत्तर देना, भाषा अनुवाद, भावना विश्लेषण और भी बहुत कुछ। समापन बिंदु पर इनपुट पेलोड निम्नलिखित कोड जैसा दिखता है:

payload = { “inputs”: str, (optional)"parameters":{"max_new_tokens":int, "top_p":float, "temperature":float}
}

निम्नलिखित कुछ नमूना उदाहरण संकेत और मॉडल द्वारा उत्पन्न पाठ हैं। सभी आउटपुट अनुमान मापदंडों के साथ उत्पन्न होते हैं {"max_new_tokens":256, "top_p":0.9, "temperature":0.6}.

Input: Simply put, the theory of relativity states that Output: 1) the laws of physics are the same for all observers in uniform motion relative to one another, and 2) the speed of light is the same for all observers, regardless of their relative motion or of the motion of the source of the light.
In the late 1800s, scientists believed that light travelled through a medium they called the luminiferous ether. But in 1887, Albert Michelson and Edward Morley designed an experiment to measure the speed of light in the ether, and found that the speed of light was the same no matter how they oriented their equipment. This result seemed to contradict the idea of the ether, and Michelson and Morley’s experiment became one of the most famous failed experiments in history.
In 1905, Albert Einstein published a paper that used the results of the Michelson-Morley experiment to develop the theory of special relativity. In 1915, Einstein published a paper on general relativity, which is a theory of gravitation.
In physics, the speed of light is considered to be a constant, and its value is 299,792,458 metres Input: I believe the meaning of life is Output: to be happy, to find joy in life, to love and be loved. To have the courage to be yourself and to be kind to others. To be able to express yourself, to create and to be inspired. To be able to give and receive. To be able to learn and grow. To be able to find beauty in the world and to be able to appreciate it. To be able to find peace and contentment. To be able to find meaning and purpose. To be able to make a difference in the world. To be able to find joy in the simple things. To be able to find joy in the difficult things. To be able to find joy in the journey. To be able to find joy in the destination. To be able to find joy in the present moment. To be able to find joy in the future. To be able to find joy in the past. To be able to find joy in the present. To be able to find joy in the future. To be able to find joy in the past. To be able to find joy in the present. To be able to find joy in the future. To be able to find joy in the past. To be able to find joy in the present. To be able to find joy Input: A brief message congratulating the team on the launch: Hi everyone, I just Output: wanted to send a quick note to congratulate everyone on the launch of our new website. I know it’s been a long time coming and I’m sure we’re all relieved that it’s finally out there. Thanks to everyone for their hard work and dedication over the last few months. I know it wasn’t always easy, but I think we can all be proud of the final product. I’m looking forward to seeing how our users respond to the new site and I’m confident that it will be a success. Thanks again, and enjoy the rest of your day! Best, [Your Name] [Your Title] [Your Company] [Your Email] [Your Phone Number] [Your Website] [Your Social Media Handles] [Your Address] [Your City, State, Zip Code] [Your Country] [Your Postal Code] [Your Website] [

अगले उदाहरण में, हम दिखाते हैं कि कुछ-शॉट-इन-संदर्भ सीखने के साथ लामा मॉडल का उपयोग कैसे करें, जहां हम मॉडल के लिए उपलब्ध प्रशिक्षण नमूने प्रदान करते हैं। ध्यान दें कि हम केवल तैनात मॉडल पर अनुमान लगाते हैं और इस प्रक्रिया के दौरान, मॉडल का वजन नहीं बदलता है।

Input: Translate English to French: sea otter => loutre de mer peppermint => menthe poivrée plush girafe => girafe peluche cheese => Output: fromage chocolate => chocolat car => voiture hamster => hamster beer => bière bicycle => bicyclette banana => banane strawberry => fraise pizza => pizza cheeseburger => cheeseburger pineapple => ananas carrot => carotte watermelon => pastèque whale => baleine eggplant => aubergine cat => chat penguin => manchot lion => lion elephant => éléphant panda => panda tiger => tigre salmon => saumon squirrel => écureuil kangaroo => kangourou turtle => tortue octopus => poulpe frog => grenouille pig => cochon goldfish => poisson rouge tiger shark => requin tigre gorilla => gorille dol

लामा-2-70बी-चैट के लिए अनुमान और उदाहरण संकेत

लामा-2-चैट मॉडल के साथ, जो संवाद उपयोग के मामलों के लिए अनुकूलित हैं, चैट मॉडल एंडपॉइंट का इनपुट चैट सहायक और उपयोगकर्ता के बीच का पिछला इतिहास है। आप अब तक हुई बातचीत से संबंधित प्रश्न पूछ सकते हैं। आप सिस्टम कॉन्फ़िगरेशन भी प्रदान कर सकते हैं, जैसे कि व्यक्तित्व जो चैट सहायक के व्यवहार को परिभाषित करते हैं। समापन बिंदु पर इनपुट पेलोड निम्नलिखित कोड जैसा दिखता है:

payload = { “inputs”: [ [ (optional){"role": "system", "content": str}, (repeated k times) {"role": "user", "content": str}, {"role": "assistant", "content": str}, {"role": "user", "content": str}, ] ], (optional)"parameters":{"max_new_tokens":int, "top_p":float, "temperature":float}
}

निम्नलिखित कुछ नमूना उदाहरण संकेत और मॉडल द्वारा उत्पन्न पाठ हैं। सभी आउटपुट अनुमान मापदंडों के साथ उत्पन्न होते हैं {"max_new_tokens": 512, "top_p": 0.9, "temperature": 0.6}.

Input:
User: what is the recipe of mayonnaise? Output:
Assistant: Sure, I can help you with that! Mayonnaise is a thick, creamy condiment made from a mixture of egg yolks, oil, vinegar or lemon juice, and seasonings. Here's a basic recipe for homemade mayonnaise: Ingredients: * 2 egg yolks
* 1/2 cup (120 ml) neutral-tasting oil, such as canola or grapeseed
* 1 tablespoon (15 ml) vinegar or lemon juice
* 1/2 teaspoon salt
* 1/4 teaspoon black pepper Instructions: 1. In a medium-sized bowl, whisk together the egg yolks, salt, and black pepper until well combined.
2. Slowly pour the oil into the bowl while continuously whisking the mixture. It's important to add the oil very slowly, otherwise the mixture can break and you'll end up with a separated, curdled mess.
3. Continue whisking until the mixture thickens and emulsifies, which should take about 5-7 minutes. You'll know it's ready when the mixture has doubled in volume and has a thick, creamy consistency.
4. Add the vinegar or lemon juice and continue whisking for another minute or two, until the flavors are fully incorporated.
5. Taste the mayonnaise and adjust the seasoning as needed. You can add more salt, pepper, or acid if desired.
6. Cover the bowl with plastic wrap and refrigerate the mayonnaise for at least 30 minutes before serving. This will allow the flavors to meld together and the mayonnaise to thicken further. That's it! This recipe makes about 1 1/2 cups (360 ml) of mayonnaise, which should be enough for a small to medium-sized batch of sandwiches or salads. You can also customize the recipe by adding different flavorings, such as garlic, mustard, or herbs, to create a unique taste. Enjoy!

निम्नलिखित उदाहरण में, उपयोगकर्ता ने सहायक के साथ पेरिस में पर्यटक स्थलों के बारे में बातचीत की है। इसके बाद, उपयोगकर्ता चैट सहायक द्वारा अनुशंसित पहले विकल्प के बारे में पूछताछ कर रहा है।

Input:
User: I am going to Paris, what should I see?
Assistant: Paris, the capital of France, is known for its stunning architecture, art museums, historical landmarks, and romantic atmosphere. Here are some of the top attractions to see in Paris: 1. The Eiffel Tower: The iconic Eiffel Tower is one of the most recognizable landmarks in the world and offers breathtaking views of the city.
2. The Louvre Museum: The Louvre is one of the world's largest and most famous museums, housing an impressive collection of art and artifacts, including the Mona Lisa.
3. Notre-Dame Cathedral: This beautiful cathedral is one of the most famous landmarks in Paris and is known for its Gothic architecture and stunning stained glass windows. These are just a few of the many attractions that Paris has to offer. With so much to see and do, it's no wonder that Paris is one of the most popular tourist destinations in the world.
User: What is so great about #1? Output:
Assistant: The Eiffel Tower is a world-renowned iconic landmark that symbolizes Paris and France. It was built for the World's Fair in 1889 and was initially intended to be a temporary structure, but it has become a permanent part of the Parisian skyline. Here are some reasons why the Eiffel Tower is so great: 1. Engineering marvel: The Eiffel Tower was a groundbreaking feat of engineering at the time of its construction. It stands at 324 meters (1,063 feet) tall and was the tallest structure in the world for over 40 years.
2. Iconic design: The Eiffel Tower's distinctive shape and design make it instantly recognizable. It has been the subject of countless works of art, photographs, and films, and has become a symbol of Paris and France.
3. Panoramic views: The Eiffel Tower offers breathtaking views of the city from its observation decks. Visitors can enjoy a 360-degree view of Paris, including famous landmarks such as the Seine River, the Arc de Triomphe, and the Montmartre neighborhood.
4. Historical significance: The Eiffel Tower has played a significant role in Paris's history, having been the site of numerous important events, including the famous "Battle of the Eiffel Tower" during World War II.
5. Cultural impact: The Eiffel Tower has had a significant impact on popular culture, inspiring countless works of art, literature, and music. It has also been the subject of numerous documentaries and films, including the classic movie "An American in Paris."
6. Romantic setting: The Eiffel Tower is known for its romantic atmosphere, making it a popular destination for couples and newlyweds. It has been the site of countless marriage proposals and has been featured in numerous romantic films and literature.
7. Accessibility: The Eiffel Tower is easily accessible by public transportation, making it a convenient destination for visitors. It is also wheelchair-friendly, with elevators and ramps providing access to all levels. Overall, the Eiffel Tower is a must-see destination for anyone visiting Paris, offering a unique and unforgettable experience that combines history, culture, engineering, and rom

निम्नलिखित उदाहरणों में, हम सिस्टम का कॉन्फ़िगरेशन सेट करते हैं:

Input:
System: Always answer with Haiku
User: I am going to Paris, what should I see? Output:
Assistant: Eiffel Tower high
Art in the Louvre Museum
River Seine flows Input:
System: Always answer with emojis
User: How to go from Beijing to NY? Output:
Assistant: Here are the emojis that represent the steps to go from Beijing to New York: 🇨🇳🛫🛬🇺🇸🗽🏙️🚕💨🛩️🛬🛫😴💤🛩️🛬🛫😍

क्लीन अप

नोटबुक चलाने के बाद, सभी संसाधनों को हटाना सुनिश्चित करें ताकि प्रक्रिया में आपके द्वारा बनाए गए सभी संसाधन हटा दिए जाएं और आपकी बिलिंग बंद हो जाए:

predictor.delete_model()
predictor.delete_endpoint()

निष्कर्ष

इस पोस्ट में, हमने आपको दिखाया कि सेजमेकर स्टूडियो में लामा 2 मॉडल के साथ शुरुआत कैसे करें। इसके साथ, आपके पास छह लामा 2 फाउंडेशन मॉडल तक पहुंच है जिसमें अरबों पैरामीटर शामिल हैं। क्योंकि फाउंडेशन मॉडल पूर्व-प्रशिक्षित होते हैं, वे प्रशिक्षण और बुनियादी ढांचे की लागत को कम करने और आपके उपयोग के मामले में अनुकूलन को सक्षम करने में भी मदद कर सकते हैं। सेजमेकर जम्पस्टार्ट के साथ शुरुआत करने के लिए, निम्नलिखित संसाधनों पर जाएँ:


लेखक के बारे में

मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ.जून जीता सेजमेकर जम्पस्टार्ट में एक उत्पाद प्रबंधक है। वह ग्राहकों को जेनरेटिव एआई एप्लिकेशन बनाने में मदद करने के लिए फाउंडेशन मॉडल को आसानी से खोजने योग्य और प्रयोग करने योग्य बनाने पर ध्यान केंद्रित करता है। अमेज़ॅन में उनके अनुभव में मोबाइल शॉपिंग एप्लिकेशन और लास्ट माइल डिलीवरी भी शामिल है।

विवेक मदान डॉ विवेक मदनी अमेज़ॅन सेजमेकर जम्पस्टार्ट टीम के साथ एक अनुप्रयुक्त वैज्ञानिक है। उन्होंने अर्बाना-शैंपेन में इलिनोइस विश्वविद्यालय से पीएचडी प्राप्त की और जॉर्जिया टेक में पोस्ट डॉक्टरेट शोधकर्ता थे। वह मशीन लर्निंग और एल्गोरिथम डिज़ाइन में एक सक्रिय शोधकर्ता हैं और उन्होंने EMNLP, ICLR, COLT, FOCS और SODA सम्मेलनों में पेपर प्रकाशित किए हैं। मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ. डॉ काइल उलरिच अमेज़ॅन सेजमेकर जम्पस्टार्ट टीम के साथ एक एप्लाइड साइंटिस्ट हैं। उनकी शोध रुचियों में स्केलेबल मशीन लर्निंग एल्गोरिदम, कंप्यूटर विज़न, टाइम सीरीज़, बायेसियन नॉन-पैरामीट्रिक्स और गॉसियन प्रक्रियाएं शामिल हैं। उनकी पीएचडी ड्यूक यूनिवर्सिटी से है और उन्होंने न्यूरिप्स, सेल और न्यूरॉन में पेपर प्रकाशित किए हैं। मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ. डॉ आशीष खेतानी अमेज़ॅन सेजमेकर जम्पस्टार्ट के साथ एक वरिष्ठ एप्लाइड वैज्ञानिक हैं और मशीन लर्निंग एल्गोरिदम विकसित करने में मदद करते हैं। उन्होंने इलिनोइस विश्वविद्यालय अर्बाना-शैंपेन से पीएचडी की उपाधि प्राप्त की। वह मशीन लर्निंग और सांख्यिकीय अनुमान में एक सक्रिय शोधकर्ता हैं, और उन्होंने न्यूरिप्स, आईसीएमएल, आईसीएलआर, जेएमएलआर, एसीएल और ईएमएनएलपी सम्मेलनों में कई पत्र प्रकाशित किए हैं। मेटा से लामा 2 फाउंडेशन मॉडल अब अमेज़न सेजमेकर जम्पस्टार्ट | पर उपलब्ध हैं अमेज़ॅन वेब सेवाएँ प्लेटोब्लॉकचेन डेटा इंटेलिजेंस। लंबवत खोज. ऐ. सुंदर रंगनाथन AWS में GenAI/फ्रेमवर्क GTM विशेषज्ञों के वैश्विक प्रमुख हैं। वह अमेज़ॅन ईसी2, ईकेएस, ईएफए, एडब्ल्यूएस बैच और अमेज़ॅन सेजमेकर जैसी एडब्ल्यूएस सेवाओं में बड़े भाषा मॉडल, जेनएआई और बड़े पैमाने पर एमएल वर्कलोड के लिए जीटीएम रणनीति विकसित करने पर ध्यान केंद्रित करते हैं। उनके अनुभव में नेटएप, माइक्रोन टेक्नोलॉजी, क्वालकॉम और मेंटर ग्राफिक्स में उत्पाद प्रबंधन और उत्पाद विकास में नेतृत्व भूमिकाएं शामिल हैं।

समय टिकट:

से अधिक AWS मशीन लर्निंग

अमेज़ॅन सेजमेकर कैनवस के साथ डेटा प्रोसेसिंग, प्रशिक्षण और अनुमान में 70% प्रदर्शन सुधार के साथ व्यावसायिक परिणामों में तेजी लाएं | अमेज़न वेब सेवाएँ

स्रोत नोड: 1870829
समय टिकट: अगस्त 3, 2023